Dounin Vladimir v 201011 MSc Thesis

Innovative Approaches for the Electrochemical

Detection of Acetylcholinesterase Inhibitors

Vladimir Dounin

A thesis submitted in conformity with the requirements for

the degree of Master of Science

Graduate Department of Chemistry

University of Toronto

© Copyright by Vladimir Dounin (2010)

ii

Innovative Approaches for the Electrochemical Detection of

Acetylcholinesterase Inhibitors

Vladimir Dounin

Master of Science

Department of Chemistry University of Toronto

2010

Abstract

This document describes research conducted during 2009-2010 in the Kerman

Group laboratory at the University of Toronto Scarborough to investigate the

application of electrochemical techniques for the detection of acetylcholinesterase

inhibitors in aqueous samples. Two main projects were completed and are

discussed herein. The first project demonstrated that the new unmodified, nano-

structured gold disposable electrochemical printed (DEP) chips produced by

BioDevice Technology can compete with surface-modified electrode configurations

to detect trace concentrations of insecticides. This was achieved through the

measurement of acetylcholinesterase-catalyzed production of thiocholine after

incubation of the enzyme with low concentrations of paraoxon (10 ppb) and

carbofuran (8 ppb). The second project featured the novel application of a glassy

carbon (GC) electrode to monitor the changes in availability of Thioflavin T (ThT) for

oxidation at the electrode surface, which is non-linearly modulated by the presence

of acetylcholinesterase and the enzyme’s pre-treatment with trace concentrations of

paraoxon and carbachol.

iii

Acknowledgments

I would like to thank my Supervisor, Professor Kagan Kerman, for all of his support

and guidance throughout this past year of exciting research and for many good

memories. My time as a Masters student with Professor Kerman has proved to be

an invaluable experience where I have learned a great deal not only about research

in electrochemistry but also about the interface of research with industry, academic

careers, education, and everyday life in this world around us.

None of my graduate study experience with the Department of Chemistry would

have taken place without the recommendation of Dr. Svetlana Mikhaylichenko, who

has been one of the most influential people in my life and a great source of advice,

encouragement, and a great role model since my undergraduate years.

I would also like to extend my thanks to all of the students in the Kerman group –

particularly those who I have had the privilege of working with the most: Anthony,

Andrea, Christopher, Tiffiny, and Vinci – for being key players in keeping a very

productive working environment in the group that is full of lively discussions,

inspiring ideas, and much-needed and well-timed humour. I have been fortunate to

be associated with such a talented group of people and I am certain that this talent

will lead them to many great accomplishments.

Furthermore, I would like to thank the administrative staff in the University of

Toronto’s Department of Chemistry at the St. George Campus, the Department of

Physical and Environmental Sciences at the Scarborough Campus, and of course

the School of Graduate Studies for their patience and support throughout my time as

a graduate student. Finally, a grateful thank-you to Professor Ulrich Krull - the

second reader of this document. Professor Krull’s teachings in his Chemical Sensors

course have really put into perspective the many possibilities that exist (and have

yet to be discovered) for the construction of a sensor and the logical steps that can

be taken in the process.

I wish everybody the best in moving forward from here to greater achievements and

successes in all future enterprises.

iv

Table of Contents

Abstract ii

Acknowledgments iii

List of Figures v

Commonly Used Abbreviations vii

1. Introduction 1

1.1 Towards the Detection of Acetylcholinesterase Inhibitors 1

1.2 Biological Sensing 2

1.3 Acetylcholinesterase Inhibitors: The Target Analytes 3

1.4 The Biological Recognition Element 6

1.4.1 Acetylcholinesterase 6

1.4.2 Operation of the Biological Recognition Element 8

1.5 Literature Review of Acetylcholinesterase Inhibitor Detection 11

1.6 Electrochemical Transduction: Voltammetry 19

1.6.1 Electrode Materials: Manufacturing Process,

Characteristics, and Uses

22

1.7 Literature Review: Electrochemical Transduction for Detection of

AChE Inhibitors

25

1.8 Contributions of this Research Work 29

2. Results and Discussion 31

2.1 Project #1: Gold disposable electrochemical printed chips for the

analysis of acetylcholinesterase inhibition using differential pulse

voltammetry

31

2.2 Project #2: Electrochemical Detection of Thioflavin T’s Interaction

with the Acetylcholinesterase Peripheral Binding Site: Application to

the Detection of Acetylcholinesterase Inhibitors

44

2.3 Concluding Remarks and Future Directions 55

3. Experimental Description and Supporting Material 58

3.1 Project #1: DEP gold chips for the analysis of AChE inhibition

using DP voltammetry

58

3.2 Project #2: Electrochemical Detection of ThT’s Interaction with the

AChE Peripheral Binding Site: Application to the Detection of AChE

Inhibitors

61

4. Literature Cited 73

v

List of Figures

CHAPTER 1

Figure 1.1 Illustration of a general biosensor flowchart 2

Figure 1.2 Chemical structures for organophosphate and carbamate AChE

inhibitors

4

Figure 1.3 Mechanisms of AChE inhibition by paraoxon and carbofuran 5

Figure 1.4 Chemical formula for the AChE-catalyzed breakdown of ACh 6

Figure 1.5 An illustration of the AChE enzyme structure for visualization 7

Figure 1.6 General reaction scheme of OPH function of an organophosphate 8

Figure 1.7 Diffusion versus kinetic controlled conditions for enzyme function 10

Figure 1.8 An illustration of the screen-printing process to make disposable

electrochemically printed chips

23

Figure 1.9 An illustration of masks that can be used in the screen-printing

process

24

CHAPTER 2

Figure 2.1 Photograph and diagram of BioDevice Technology gold DEP chip 31

Figure 2.2 DP voltammograms of 2.1 mM ATCh on gold DEP chip 34

Figure 2.3 DP voltammograms of a solution of 0.5 nM AChE and 2.1 mM

ATCh on a gold DEP chip after a 10 min incubation

34

Figure 2.4 Comparison of DP voltammograms after ATCh incubation with

AChE that has or has not been pre-treated with 50 ppb carbofuran

36

Figure 2.5 Calibration plots for paraoxon and carbofuran obtained for the

ATCh-AChE biosensor using the gold DEP chips

37

Figure 2.6 Background DP voltammogram measurements for milk and river

water real samples

38

Figure 2.7 Comparison DP voltammograms of AChE and ATCh after a 10 min

incubation in real samples

39

Figure 2.8 Illustration of a proposed biosensor design featuring AChE

attached to magnetic microbeads

43

Figure 2.9 Chemical structure of ThT 44

Figure 2.10 Experimental setup of the three-electrode system used to conduct

experiments with ThT

45

Figure 2.11 DP voltammogram of 280 nM ThT on GC electrode 45

vi

Figure 2.12 Calibration plot of ThT oxidation peak current concentration

dependence

46

Figure 2.13 Calibration plot of 280 nM ThT oxidation peak current pH

dependence

48

Figure 2.14 Calibration plot of 280 nM ThT oxidation peak current dependence

on AChE concentrations

49

Figure 2.15 Comparison of ThT and BTA-1 chemical structures 50

Figure 2.16 Sample DP voltammograms illustrating changes in 860 nM BTA-1

oxidation peak currents in the presence of AChE and paraoxon

51

Figure 2.17 Calibration plots for the ThT-based AChE inhibitor biosensor and

DP voltammograms illustrating changes occurring to ThT oxidation

peak currents during measurements

52

Figure 2.18 Illustration of ThT oxidation at the GC electrode surface while in the

presence of AChE and inhibitor molecules

53

CHAPTER 3

Figure 3.1 Flowchart representation of procedural steps completed during

measurements for the gold DEP chip research

60

Figure 3.2 Comparison of DP voltammograms for 280 nM ThT taken directly in

solution or after a wash step in PBS buffer

64

Figure 3.3 Comparison of DP voltammograms for a solution of 280 nM ThT

and 12.5 nM AChE taken directly in solution or after a wash step in

PBS buffer

64

Figure 3.4 Comparison of CVs for 20 mM ferri-ferrocyanide on a GC electrode

in PBS buffer and in solution with 12.5 nM AChE

66


in solutions with varying AChE concentrations

67


in PBS buffer versus a solution of 100 ppm carbachol

69


in a solution of 12.5 nM AChE versus a solution of 12.5 nM AChE

with 100 ppm carbachol

70


in a solution of 200 nM AChE versus a solution of 200 nM AChE

with 280 nM ThT

71

vii

Commonly Used Abbreviations

ACh: Acetylcholine

AChE: Acetylcholinesterase

ATCh: Acetylthiocholine

BTA-1: Benzothiazole-1

Ch: Choline

CNT: Carbon nanotubes

CV: Cyclic voltammetry

DEP chips: Disposable electrochemical printed chips

DPV: Differential pulse voltammetry

FET: Field effect transistor

GC electrode: Glassy carbon electrode

NMR: Nuclear magnetic resonance

OPH: Organophosphorous hydrolase

PBS: Phosphate buffer solution (saline)

TCh: Thiocholine

ThT: Thioflavin T

1

1. INTRODUCTION

1.1 Towards the Detection of Acetylcholinesterase Inhibitors

The acetylcholine (ACh) neurotransmitter performs very important functions in the

peripheral and central nervous systems. The presence of the ACh neurotransmitter

is primarily regulated by the acetylcholinesterase (AChE) enzyme, whose native

function is to cleave ACh into choline and acetate. AChE inhibitors disrupt ACh

regulation and consequently promote elevated levels of ACh in nervous and muscle

tissues with complex physiological effects1-3. Some of these effects are understood

well enough to warrant medicinal uses of AChE inhibitors such as in Alzheimer’s

disease treatments. Critical elevation of ACh produces a fatal disruption of the

nervous system. This fact has led to the successful introduction of AChE inhibitors to

maximize crop yields in agriculture by killing insects and other pests that destroy

crops. The use of AChE inhibitors in warfare has also occurred such as with the

infamous Sarin gas, which was dispersed by a radical religious cult to target

innocent people in the 1995 “Sarin Subway Incident” in Tokyo4. Furthermore, poor

control of AChE inhibitors used in agriculture has resulted in many accidental

poisonings and the introduction of these chemicals into water and food resources.

The effects of consuming AChE inhibitors at sub-critical concentrations over

prolonged periods of time are currently poorly understood. Existing data, particularly

on organophosphate AChE inhibitors, suggests that there is a range of sensitive

non-AChE targets in living systems5 in addition to AChE and that the

phosphorylation of these additional targets leads to detrimental effects in the fields

of development and behaviour6. Regulatory bodies in Europe and North America

have limited the maximum allowable concentration of many AChE inhibitors in food

and water to be at most in the mid-ppb range and usually at 20 ppb or lower7-9.

Early efforts towards the detection of AChE inhibitors were traditionally

accomplished through the chromatographic separation of samples followed by

analysis through nuclear magnetic resonance and mass spectrometry10. These

approaches provide accurate determination of AChE inhibitor identities and can

quantify their sample concentrations precisely. However, these traditional

2

approaches take hours to complete, need sample preparation steps, and require a

skilled operator to run the appropriate test procedures11. In modern times, alternative

detection methods continue to be developed in the form of biosensors that are

relatively cost-effective, sensitive, and offer rapid analysis of samples for AChE

inhibitors with a minimal amount of required training of personnel. This chapter will

provide an overview of biological sensing for these inhibitors before proceeding on

to the details of the featured research completed in the Kerman Group laboratory in

2009-2010.

1.2 Biological Sensing

The term ‘biosensor’ describes a device designed for the detection of a particular

analyte through a measurable interaction of the analyte with a biological recognition

element. The biological recognition element can consist of structures that are

relevant to living organisms, ranging from entire cells to complicated proteins or

even to something as small as a short oligonucleotide. The interaction of analyte

with the biological recognition element can be detected through a suitable

transducer, which is a device integrated into the biosensor that can respond to the

biological interaction events in the form of a change in some measurable quantity,

usually presenting itself as an electrical signal 12. The signal from the transducer can

then be amplified and processed for interpretation by the end-user in analog or

digital formats13.

Figure 1.1. A general flowchart representation of biosensor operation.

The choice of the biological recognition element and transducer ultimately

determines the capacity of the assembled device for biological sensing in terms of

3

the simplicity of use, assembly and operation costs, portability, and the overall

performance of measurements in selectivity, sensitivity, speed, and stability.

The characteristics of the biological recognition element are most important in

determining the biosensor’s performance in selectivity – the capacity to detect target

analytes without significant signal interference from non-target analytes – and

specificity – the capacity to only detect the target analyte in the presence of non-

target analytes13. The detection occurs either through a selective binding event,

such as an interaction of an antigen with an antibody, or through a selective reaction

such as that observed for enzymes and their substrates12. Sensitivity and sensor

measurement speed are usually inversely related due to the limiting rates of the

interactions taking place at the biological recognition layer. Finally, the biological

recognition element may undergo degradation and structural changes over time,

which affects stability and reusability. Immobilization of the biological component is

possible using a variety of approaches, including gel entrapment, adsorption,

membrane confinement, and chemical functionalization13. Meanwhile, the

technology that is used for the transducer limits the sensitivity and portability of the

biosensor. Technological improvements in transducer development can improve

sensitivity and allow for smaller sensor size for better portability. The transducer also

affects the speed of measurements made with the biosensor since some transducer

technologies take longer to work than others.

1.3 Acetylcholinesterase Inhibitors: The Target Analytes

AChE inhibitors are chemical or biological compounds that can interact with the

AChE enzyme to inhibit its function of breaking down ACh. A variety of chemical

structures show AChE-inhibiting activity. Several different categories of AChE

inhibitors exist, of which the most popular are the carbamates and

organophosphates for their applications in agriculture. The carbamate category

includes a variety of compounds that contain the carbamate ester functional group.

4

In contrast, organophosphate AChE inhibitors are ester derivatives of phosphoric

acid. The chemical structures of these two categories of inhibitors are shown below.

Figure 1.2. General chemical structures for the most popular categories of AChE inhibitors,

namely organophosphates (left) and carbamates (right).

The routes of entry for AChE inhibitors are through ingestion, absorption, and

respiration. The absorption of these compounds into the body results in the targeting

of AChE enzymes that exist in the muscle, blood, and nervous system14. As the

AChE inhibitor encounters an AChE enzyme, it usually interacts with the enzyme’s

catalytic site, which is located at the bottom of a deep and narrow gorge15. This

interaction can be irreversible with the covalent modification of a serine residue in

the catalytic site or can be reversible if the interaction is based on temporary affinity

binding. For example, the carbamate class of AChE inhibitors reacts with the serine

residue through a carbamylation reaction, which transfers the methylcarbamoyl ester

group to the serine. In contrast, the organophosphate class of AChE inhibitors

undergo a phosphorylation reaction with the same serine residue, leaving a

phosphate group attached.

Phosphorylation of the serine residue is not readily reversible without an antidote

such as 2-pralidoxime (2-PAM) and becomes permanent within 10 h of exposure as

“aging” – the process of dealkylation at the attached organophosphate’s R groups –

takes place16. In this process, the alkoxy-O-P bonds of the attached

organophosphate are broken and replaced with weaker hydroxy leaving groups

either through a general acid catalyzed nucleophilic substitution reaction or assisted

by stabilization of leaving groups by amino acids within the catalytic site of the

enzyme17. In contrast to what happens with organophosphates, a carbomylated

5

serine is very unstable and undergoes hydrolysis in a matter of hours to regenerate

the serine residue18.

Figure 1.3. Mechanisms involved in the action of paraoxon (an organophosphate AChE

inhibitor) and carbofuran (a carbamate AChE inhibitor) on AChE’s active site serine residue.

Apart from carbamates and organophosphates, other types of AChE inhibitors

include phenanthrene, piperidine, and indanone. Their mechanisms of action are

affinity-based with ionic and hydrogen bond interactions between functional groups

and amino acid residues in the catalytic site of AChE. In addition to the catalytic site,

the peripheral binding site exists near the entrance to the gorge that leads to the

catalytic site. It is another possible target for AChE inhibitors that function by

blocking the entrance of ACh to the enzyme’s catalytic site. Inhibitors that interact

with the peripheral site include small molecules such as propidium and also peptide

toxins like fasciculin15. Certain compounds have also been specifically designed for

medicinal purposes to interact weakly with both the catalytic site and peripheral

binding site, such as Donepezil for Alzheimer’s disease treatment.

6

The toxicity of AChE inhibitors to an organism varies depending on several factors

including the chemical structure of the inhibitor and the species variant of the AChE

enzyme exposed to the inhibitor. In general, AChE inhibitors that target the catalytic

site have a toxicity that is primarily determined by the conformational freedom of the

leaving group that is removed during the alkylation step and also by the inhibitor’s

hydrophobicity. These properties determine the accessibility of the catalytic site to

the inhibitor, since the catalytic gorge consists largely of hydrophobic amino acid

residues19. Variations in the DNA sequence encoding for the AChE enzyme between

different species can also make particular AChE enzymes more susceptible to

certain AChE inhibitors than others. This has been exploited for the production of

AChE mutants that can be applied for the purpose of AChE inhibitor detection with

some degree of selectivity for particular inhibitor structures.

1.4 The Biological Recognition Element

1.4.1 Acetylcholinesterase

The AChE enzyme has an asymmetric, usually globular ellipsoidal protein structure

that consists of a large central α/β-sheet core with 8 β-sheets that are connected to

one another by α-helices, which designates it structurally as a α/β-fold enzyme20.

This central core is surrounded by another 15 α-helices21. The primary function of

AChE is to break down the neurotransmitter ACh into choline (Ch) and acetate at

cholinergic synapses as indicated in the following equation

Figure 1.4. The AChE-catalyzed cleavage reaction of acetylcholine (ACh) into choline (Ch)

and acetate.

7

The AChE enzyme appears most abundantly in a tetrameric form with average

dimensions of 25 x 18 x 1.6 nm (~720’000 Å3)21. Each enzyme has an active site

volume of ~300 Å3 at the bottom of a ~20 Å deep hydrophobic gorge. The gorge is

lined with mostly aromatic amino acid residues along with a few acidic residues,

which are known to affect the affinity of AChE enzymes from different species to

their substrates and inhibitors20. Near the edge of this hydrophobic gorge lies an

anionic peripheral binding site, whose amino acid residues form an electric field that

attracts the cationic acetylcholine substrate into the gorge and down towards the

active site with the help of dipole-dipole interactions with the aromatic amino acid

residues20.

Figure 1.5. An illustration of the AChE enzyme’s structure for visualization purposes, with

the green area representing the active site and a yellow molecule shown occupying the

peripheral binding site. X-ray diffraction results showing the interaction of Thioflavin T with

electric eel AChE are available on the Protein Data Bank at the DOI: 10.2210/pdb2j3q/pdb

The AChE enzyme is classified as a serine hydrolase, with a catalytic triad present

at its active site that consists of serine, histidine, and glutamate amino acid residues.

The latter acidic residue is usually found to be aspartate in most serine hydrolase

enzymes. The aspartate residue stabilizes the histidine residue’s intermediate

imidazolium cation, which isolates the choline group of the natural substrate20. This

is followed by a hydrolysis reaction that cleaves off acetate. In addition to AChE’s

native function in modulating ACh concentrations in the nervous system, its other

functions include neuritogenesis, synaptogenesis, and amyloid-β complex

formation22.

8

1.4.2 Operation of the Biological Recognition Element

The biological recognition element of a biological sensor can contain sugars,

proteins (including enzymes such as AChE), nucleic acids, receptors, or entire cells.

The mode of action is classified as being either catalytic or affinity-based23. Catalytic

recognition is made possible by the selective affinity of enzymes for their substrates.

When the enzymes are activated or inhibited, the quantity of product made over time

changes and the difference can be measured24. Furthermore, enzymes can be

involved in transforming the target analyte as a substrate into a different measurable

product. An example of catalytic recognition applied for AChE inhibitor detection is

the use of organophosphorous hydrolase (OPH), an enzyme that breaks P-O bonds

of organophosphate compounds to make alcohol and acid products25. These

products can then be detected with a variety of transducers.

P

O

O

O

O

+ H2O +R3

OHR3

R1

R2

P

HO

O

O

O

R1

R2

OPH

Figure 1.6. General reaction scheme of an organophosphate compound with the OPH

enzyme.

Affinity-based recognition involves the irreversible and non-catalytic binding of a

target species to the biological recognition element23. Affinity-based recognition of

AChE inhibitors has traditionally been achieved with immunoassays by using

antibodies that have affinity for certain inhibitor compounds. In the first research

project described in this document, catalytic-based recognition was applied for the

detection of AChE inhibitors using the AChE enzyme.

In order to understand the processes taking place at the catalytic-based biological

recognition element, enzymes are modelled with fundamental enzyme kinetics and

Michaelis-Menten kinetics in mind. The observed initial rate of reaction depends on

the concentration of the substrate (S), enzyme (E), and the Michaelis (KM) and

9

dissociation (kd) constants respectively. The constants reflect the relative rates of

substrate-enzyme association/dissociation and product formation.

As product P is created through the enzyme catalyzed reaction, there is an initial

linear relationship between the change in product concentration d[P] and time.

However, as [S] decreases and [P] increases, the rate of enzyme activity decreases

due to a lack of saturation of the enzyme by substrate and/or competitive inhibition

due to the affinity of the product to the enzyme’s active site. Therefore, depending

on how an enzyme-based biosensor is calibrated in terms of incubation times with

the substrate, different types of responses can be obtained for the same measured

phenomenon. This is complicated when AChE is the chosen enzyme since AChE

inhibitors may also be substrates of AChE. Furthermore, some AChE inhibitors

permanently inhibit the enzyme (organophosphates), some do so temporarily

(carbamates), and yet others simply compete with ACh for the catalytic site.

There are also some subtle details about the structure and function of enzymes that

affect biosensor measurements. Very low concentrations of inhibitors can

sometimes tend to activate instead of inhibit AChE. This is likely due to interactions

of the inhibitors with the peripheral binding site of AChE, causing conformational

changes that improve accessibility of the active site. These interactions also appear

to be time-dependent, with prolonged peripheral site binding leading to decreased

accessibility of substrates to the active site26. The effectiveness of AChE inhibitors

decreases with increasing concentrations of the inhibitors due to the development of

steric blockades around the entrance to the enzyme active site27. These facts

explain the wide variety of possible output responses, linear and non-linear,

10

obtained from different biosensor designs that use various concentrations of enzyme

and substrate along with different incubation times.

Realizing the need for a standardized approach to achieving linear sensor

responses in the design and calibration of AChE-based AChE inhibitor biosensors,

Zhang et al.28 recommended that AChE and its substrate should exist at

concentrations that ensure the rate of product formation is not contingent on

substrate concentration. In this case, kinetically controlled conditions would be

maintained. The substrate should saturate the enzyme (i.e. [S] >> KM) to ensure that

it operates with zero-order kinetics29. In contrast, the incubation of AChE with

inhibitors in the experimental sample should take place under diffusion controlled

conditions, so that the enzymes are not saturated by the inhibitors at any time. If

saturation of AChE occurs during this step, the rate of inhibition is no longer linearly

dependent on inhibitor concentration since there is competition between inhibitor

molecules for access to the enzyme active site.

Figure 1.7. Schematic of diffusion controlled conditions (a), where diffusion of substrate to the enzyme determines the rate of reaction, unlike kinetic controlled conditions (b), where the enzymes are saturated and working as fast as they can. Sometimes, in kinetically controlled conditions, steric hindrance of the substrates near the active site can slow the exit of the enzyme product, thus decreasing the aggregate rate of reaction.

The suggested standardized approach by Zhang et al.28 may not necessarily work

for all different types of enzymes that may appear at the biological recognition

11

element. In general, enzyme interactions with the target analyte that is being

quantified should take place under diffusion controlled conditions to ensure linearity

in sensor response. Saturation of the enzyme by the analyte creates a situation

where not all of the analyte molecules have an opportunity to interact with the

enzyme during the incubation step. However, when the enzyme-catalyzed reaction

is used strictly to amplify the biosensor signal, kinetic controls are necessary to

ensure that the enzymes are consistently creating products for maximum biosensor

signal response. Without kinetic control, inhibited enzymes may not even have any

significant impact on overall product formation during the incubation step with their

substrate. Under kinetic controls, all of the enzymes work throughout the substrate

incubation step and a decrease in product due to enzyme inhibition will be more

visible. In practice, saturation of the enzyme at the biological recognition layer leads

to kinetic control but, if too much substrate is present, reaction rates decrease due to

the presence of steric blockades at the enzyme active sites.

1.5 Literature Review of Acetylcholinesterase Inhibitor Detection

The use of NMR and mass spectrometry is a well-established analytical approach

for the detection and identification of AChE inhibitors. However, detection

techniques which are faster, cheaper, and portable have been developed over the

past thirty years. In the early-to-mid 1980s, immunoassays had come into popularity

and some antibodies had been developed for certain AChE inhibitors, such as

soman and paraoxon, either directly or as a part of haptens when the inhibitor

molecules are too small to be recognized by antibodies on their own10, 30. Using a

competitive inhibition enzyme immunoassay format, Hunter et al.10 showed that it

was possible to detect paraoxon in low-ppb ranges both in solution and in serum.

Detection usually involves the enzyme-linked immunosorbent assay format or an

assay featuring fluorescent- or chemiluminescent-tagged antibodies31. Currently,

research is ongoing to produce recombinant antibodies for a variety of small

molecules such as AChE inhibitors, although the overall number of useful antibodies

remains low to this day. Unfortunately, the impressive detection abilities of

12

immunoassays also usually require multiple preparation steps and a long incubation

times that require multiple hours to complete11. However, it has been recently shown

that immunoassays can actually be used to detect AChE inhibitors in as little as 10

min and in a single step but with somewhat higher detection limits nearing 250 ppb.

This was accomplished by Zhou et al.32 in the form of a gold

immunochromatographic assay using carbofuran monoclonal antibodies labelled

with colloidal gold particles. Although this detection limit is not as low as other more

sophisticated sensing platforms, this immunoassay approach is sufficient to test for

toxic levels of many AChE inhibitors. It would also be useful to test for regulatory

compliance in the United States, where AChE inhibitor concentrations are regulated

to be in the mid-ppb to low-ppm ranges9. Further developments of this type of

immunochromatographic technique may also improve detection limits in the near

future, such as with silver enhancement of the gold nanocolloid, which allows for

double labelling of the same antibody31. Thus, the application of rapid

immunoassays for AChE detection is not an idea that should be readily dismissed.

There remains an overall scarcity of useful antibodies that respond to AChE

inhibitors even to this day. Once this scarcity is addressed, the usefulness of

immunoassay techniques will be better recognized.

The slow detection times of traditional immunoassay techniques existing in the

1980s drew attention to the exploration of other biological recognition and

transduction approaches for the detection of pesticides including AChE inhibitors. At

about the same time as immunoassays were being developed for this purpose,

research groups were beginning to apply the AChE and organophosphate hydrolase

(OPH) enzymes as the biological recognition elements coupled with a variety of

common transducers in optical, electrochemical, and mass-sensitive sensor

designs. OPH is an enzyme that only breaks down organophosphate compounds

into an alcohol and an acid, both of which are more useful for transduction purposes

than the original triester compounds themselves25. In the literature, there are many

peer-reviewed articles that document research on AChE inhibitor detection featuring

AChE and OPH. Furthermore, excellent efforts have been made to develop methods

13

of preserving enzyme activity over time so that the potential biosensors would have

a substantial shelf life on the order of months to years.

Sensors featuring optical transduction mostly rely on the interaction of an indicator

or sensor surface either directly with the analyte or indirectly with other species in

the sensor environment that can report on analyte concentrations. Sensing occurs

when resulting changes in absorbance or fluorescence are detected. Optical

transduction systems are very diverse and include many flavours of absorbance,

bioluminescence, chemiluminescence, evanescence and fluorescence33. In 2005,

White and Harmon demonstrated that portable and rapid optical solid-state detection

of organophosphates was feasible using OPH immobilized on glass microscope

slides with detection limits in the ppt-range and detection times of 10 seconds34. This

technique featured monitoring the absorbance of copper metalloporphyrins that

interact with the OPH enzyme and get displaced by trace concentrations of the

organophosphate substrates. A similar concept using AChE was applied by

Nagatani et al.35 where 5,5 dithiobis-2-nitrobenzoic acid was converted through a

reaction with thiocholine into 5 -mercapto-2-nitrobenzoic acid, where the latter

chemical compound was detected optically as a yellow dye that absorbed light at

410 nm. In the presence of AChE inhibitors, AChE would produce fewer thiocholine

molecules from its cleavage of acetylthiocholine (ATCh) substrate, producing lower

concentrations of the yellow dye. This approach allowed for visual discrimination

between 0.1 and 0.2 ppm DZN-oxon and down to the low-ppb range with a hand-

held photometer. More recently in 2007, Vamvakaki and Chaniotakis36 applied

liposomes to trap AChE and a pH-sensitive fluorescent dye called pyranine while

allowing the transport of ACh and AChE inhibitors through porins in the liposome. In

the presence of AChE inhibitors, AChE activity decreased inside the liposomes and

pyranine fluorescence decreased. Using this approach, detection limits in the mid-

ppq range were established with measurement times in as little as 5 min. Then, in

2009, Dale and Rebek37 achieved millisecond detection times without a biological

recognition element, opting instead to use oxime ring chemistry to detect

organophosphate nerve gas agents at ppm levels and lower. The presence of

organophosphates produces a ring-closing reaction that red shifts oxime

14

fluorescence maxima by 30 nm. This sensor has excellent potential to be used for

real-time personnel monitoring in hazardous conditions where AChE inhibitors may

be present.

The use of semiconductor nanoparticles (quantum dots) for AChE inhibitor detection

has also just recently been realized as biosensor research continued to develop

through 2003. Quantum dots are desirable for use in optical detection systems due

to their resistance to photo-bleaching, their wide excitation wavelength ranges, and

narrow size-dependent emission wavelengths that allow for multiplexing applications

when compared to conventional fluorescent dyes31. An excellent review of quantum

dots and their applications in optical detection systems can be found in the cited

literature12, 38. Using quantum dots functionalized with OPH, Constantine et al.39

found that paraoxon’s binding with OPH resulted in changes in quantum dot

photoluminescence. This type of biosensor yielded detection limits in the low ppb-

range for paraoxon almost instantly upon introduction of the sample to the quantum

dots as the OPH underwent conformational changes. However, more development

is required to improve selectivity of this type of sensor by experimenting with

different quantum dot coatings, possibly changing the structural properties of OPH

or replacing OPH with other biomolecules that interact with AChE inhibitors31.

However, quantum dots are definitely not limited to use with optical transduction,

since they are also applicable in electronic transduction systems due to their

capacity for electron exchange. Quantum dots have also been shown to be useful in

photoelectrochemical transduction designs in AChE inhibitor biosensors. Pardo-

Yissar et al.40 used AChE-derivatized quantum dots that were then covalently linked

to a gold electrode surface to create a photoelectrochemically active biosensor that

responds to thiocholine. Thiocholine interacts with the quantum dots, enabling them

to produce a stable photocurrent upon excitation by wavelengths of light in their

absorption band. In the presence of AChE inhibitors, tested in the ppm-range for this

sensor, less thiocholine was produced and the photocurrent decreased significantly.

Although the detection limit was not reported in this study, it is nevertheless a very

unique application of quantum dots and nanomaterials towards the detection of

AChE inhibitors.

15

Continuing on the topic of nanomaterials, the growth processes of nanoparticles

have also been utilized in biosensor designs for the detection of AChE inhibitors.

When AChE breaks ATCh into thiocholine, the latter product’s presence apparently

serves to slow the growth of gold-silver nanoparticles originating from the deposition

of silver on seed gold nanoparticles in the presence of a reducing agent such as

ascorbic acid. Thiocholine binds to the seed gold nanoparticle surface and blocks

the access of silver atoms to the gold nanoparticle surfaces. In the presence of

AChE inhibitors, less thiocholine is produced by AChE and an increase in the rate of

nanoparticle growth is seen. Virel et al.41 applied this observation to make a

biosensor using a colorimetric assay measuring absorbance at 400 nm for the gold-

silver nanoparticle plasmon band. AChE inhibitors such as paraoxon were tested

and detected in low ppb-levels in a matter of 5-10 min. A similar system in which

osmium complexes are made to play a role in gold nanoparticle growth was realized

by Xiao et al.42. When AChE cleaved ACh into Ch, this allowed for the reduction of

an oxidized osmium complex through the concurrent oxidation of Ch into betaine.

The reduced osmium complex promotes nanoparticle growth. The presence of

AChE inhibitors decreased the production of choline, the necessary reducing agent

to regenerate the required osmium complex, subsequently slowing nanoparticle

growth42.

Interesting work has also been conducted using carbon nanotubes (CNTs) for

application in biosensor design. CNTs represent a nanomaterial with many possible

applications in sensor design due to their large surface area, electrical conductivity,

stability, self-assembly, and capacity for surface modifications. CNTs have mostly

found applications in sensors using electric and electrochemical transduction

schemes, the latter of which will be introduced later in this document after a primer

introduction to electrochemical transduction. As of yet, there are very few notable

cases that are not related to electrochemical transduction where CNTs were applied

for the detection of AChE inhibitors, such as with the sensor developed by Ishii et

al.43 that features afield effect transistor (FET)- based AChE inhibitor biosensor

using AChE immobilized on carbon nanotubes. This biosensor exploits the fact that

CNTs are useful as semiconductor materials. When AChE inhibitors bind to the

16

immobilized enzymes on the CNT surface, the effective potential at the surface of

the CNTs is changed and can be detected through the source-drain current in the

microampere range. With an incubation time of 10 min, the sensor could achieve

detection limits of 200 ppq, which makes it one of the most sensitive devices for

measuring the presence of the AChE inhibitors acephate and fenitrothion.

Mass-sensitive transduction schemes have also been applied for the purpose of

AChE inhibitor detection. These include piezoelectric transducers and surface

plasmon resonance transducers (both the standard and localized varieties).

Piezoelectric transducers feature materials, such as quartz, which respond to

mechanical stresses by producing an electric potential that can be detected,

amplified and analyzed as a quantitative signal. A good example of an AChE

inhibitor biosensor using a piezoelectric transducer is the precipitation biosensor

reported by Kim et al.44 in 2007. A quartz crystal microbalance was used with a gold

modified quartz surface, which allowed for the immobilization of AChE in close

proximity to the surface via a sulphur-tagged linker. The substrate used was 3-

indolyl acetate, which is cleaved into 3-hydroxyindole by AChE. This product

oxidizes into a blue precipitate that settles onto the quartz surface and thus creates

a shift in the resonance frequency detected on the microbalance. Using this sensor

configuration, Kim et al.44 achieved mid-ppt to low-ppb detection limits for carbofuran

and EPN in 10 min. In contrast to piezoelectric transduction, surface plasmon

resonance (SPR) relies on the sensitivity of surface plasmons to changes in the

refractive index of bulk solution (or air) within nanometers of the metal surface45.

Using SPR, Rajan et al.46 developed a flow-based sensor that detected binding

events of an AChE inhibitor with AChE that was immobilized on the silver core of a

plastic-cladded silica optical fiber. In the presence of acetylcholine substrate, the

introduction of chlorphyrifos changed the refraction index in the vicinity of the silver

core and thus led to changes in the SPR wavelength. This effect was believed to be

occurring due to expulsion of acetylcholine from the AChE active site. This approach

led to detection limits in the low-ppb range in less than 10 min. Using a similar flow-

based SPR approach and similar measurement timelines, Mauriz et al.47 obtained

17

ppt-range detection limits for chlorphyrifos by replacing AChE with anti-chlorphyrifos

antibodies.

In closing of this literature review on AChE inhibitor biosensor designs, it is worth

mentioning the application of photothermal transduction for the purpose of detecting

AChE inhibitors due to the distinct characteristics of the transduction method, which

measures temperature changes in samples. Using an argon ion laser, Pogacnik et

al.48 irradiated thiocholine and measured the temperature change of the sample to

give an indication of existing AChE activity versus control experiments. The

presence of AChE inhibitors in a sample would decrease AChE activity. This would

in turn lead to lower amounts of thiocholine in solutions and lower detected changes

in temperature versus controls. With this approach, it was possible to achieve

detection of low-ppb ranges of AChE inhibitors in 6 min with excellent matching to

real sample concentrations detected by GC-MS measurements. Although the

thermal lens spectrometer required for these measurements is quite bulky, there is

definitely room for miniaturization as current portable models come in the size of a

suitcase (15 cm x 10 cm x 3 cm) containing all of the components including laser

diodes and bioanalytical column49. This type of biosensor would be most applicable

to work with samples that do not require substantial preparation steps, such as river

water, since foodstuffs typically require lengthy procedures including pureeing and

centrifugation prior to injection into the bioanalytical column of the biosensor.

It is evident from this brief literature review that many different biosensor designs

exist for the purpose of detecting AChE inhibitors. These rely on a variety of different

transduction schemes but many share the common biological recognition element

that is the AChE enzyme or OPH enzyme. These enzymes allow for limited

specificity in a biosensor given that OPH responds to organophosphates whereas

AChE is inhibited by AChE inhibitors. However, the enzymes are also affected by

their solvents, pH, ionic strength, and proteases existing in solution that degrade the

enzyme protein structure. This is critical when such biosensors are tested using real

samples as it can lead to questions about the reliability of the devices for on-field

use. It is not surprising then that very few (~20%) peer-reviewed academic papers

18

describing biosensors for AChE inhibitors include any real sample analyses in their

research reports50. Furthermore, where the enzymes can be demonstrated to work

as intended in real samples, they cannot distinguish between different

organophosphates or AChE inhibitors in any given unknown sample. The activity of

OPH only indicates the presence of any solvated organophosphate substrates

whereas loss of AChE activity only indicates the presence of some kind of AChE

inhibitor. Thus, in real samples where the identity of AChE inhibitors present in

solution is unknown, the sensor output has very limited value beyond indicating that

some species are present that interact with the enzyme at the biological recognition

element. Composite samples containing more than one type of AChE inhibitor would

require additional modifications to a biosensor in order for the device to provide

meaningful information about the individual components. Some attempts have been

made to achieve this through favourable modification of enzyme conformation.

Using directed evolution and amino acid substitutions, it is possible to isolate AChE

mutants that are more or less sensitive to particular AChE inhibitor species. For

example, Bachmann et al.51 created genetically engineered variants of the

Drosophila melanogaster AChE enzyme, which they subsequently applied in the

development of an artificial neural network (ANN) that could identify the component

concentrations of composite solutions of paraoxon and carbofuran. The ANN

processed collected readings from four different D. melanogaster AChE enzymes to

predict the individual concentrations of paraoxon and carbofuran in each sample

solution. The group succeeded in establishing a detection range between 0-5 ppb

for each compound with ~10% prediction error.

In the research described within this document, electrochemical transduction was

utilized along with AChE as the biological recognition element. This research

direction is not meant to reflect a particular bias for electrochemical techniques or

the biological recognition element, since the variety of options reported in this

literature review have all shown very impressive capabilities for AChE inhibitor

detection. It serves as a contribution to our current understanding of electrode

surface modifications and explores a new way of monitoring AChE inhibitors through

19

the oxidation of molecules that weakly intercalate with AChE. This is described in

further detail in Section 1.8.

1.6 Electrochemical Transduction: Voltammetry

This primer on electrochemical transduction was assembled based on the reference

material presented in an excellent introductory book to electrochemistry techniques

called Analytical Electrochemistry by Joseph Wang 52. Readers who wish to become

more familiarized with electrochemistry as it applies to chemical and biological

sensors are encouraged to consult this resource in the cited literature. In addition,

the electrochemistry scholar seeking a thorough understanding of electrochemical

techniques may refer to Alan J. Bard’s and Larry R. Faulkner’s book Electrochemical

Methods: Fundamentals and Applications53. The use of electrochemical transduction

offers the benefits of low costs, short measurement times, and an excellent potential

for the miniaturization and portability of the final assembled biosensor24.

Electrochemical transduction is not affected by the turbidity of sample solutions,

which is a major problem in the application of optical transduction platforms to such

samples54. Furthermore, the transduction process is relatively simple compared to

other technologies. In addition to the sample being analyzed, the components

required for electrochemical transduction include an electrode system, a voltage

source, and a potentiostat to collect electrical measurements. The electrode system

usually consists of three electrodes: working, reference, and counter. There are

many forms of electrode systems that are available in a variety of shapes and sizes

of which two were used in the research described herein: rod-shaped individual

electrodes and screen-printed electrodes.

Electrochemical transduction involves the monitoring of redox (reduction-oxidation)

reactions at the working electrode surface under various conditions of applied

potential from the voltage source. In general, a redox reaction at the electrode

surface involves the transfer of electrons to and from the members of the redox

species at a particular value of applied potential as described by the Nernst

equation,

where

electrons involved in the redox reaction per molecule, F is

is the universal gas constant, T is the Kelvin scale temperature, C

oxidized analyte concentration and C

The current observed at the applied potential depends on the flu

(O) or reduced (R)

where D represents the diffusion coefficient for the analyte (in cm

of the redox active species, V(x,t) is the hydrodynamic velocity in the x

C(x,t) is the concentration

added to negate flux due to a potential gradient and the sample solution is not

stirred during

After addressing the

diffusional flux, current can eventually be expressed through the Cottrell equation

(given here for a planar

where

applied past and above the value of E

become depleted and a diffusion layer

called

on the x

exists within 59 mV of E

solution over the voltage scanning period

where Eo is the standard redox potential for the redox reaction,





(O) or reduced (R)



represents the concentration gradient,



stirred during the

After addressing the


en here for a planar

here A is the area of a planar electrode (in cm



called a voltammogram

on the x-axis. The result of the sweep is a

exists within 59 mV of E

olution over the voltage scanning period

is the standard redox potential for the redox reaction,





(O) or reduced (R) analyte to the electrode surface,






the measurement

After addressing the mathematical expression for the


en here for a planar electrode):

A is the area of a planar electrode (in cm



voltammogram can be plotted with current

The result of the sweep is a

exists within 59 mV of Eo, demonstrating the oxidation of reduced species existing in







analyte to the electrode surface,




C(x,t) is the concentration of O or R at a particular position and time. If excess salt is


measurement, only diffusion plays a role in determining the flux.

mathematical expression for the


electrode):




can be plotted with current

The result of the sweep is a

, demonstrating the oxidation of reduced species existing in





oxidized analyte concentration and CR is the initial reduced analyte concentration.






at a particular position and time. If excess salt is


, only diffusion plays a role in determining the flux.




applied past and above the value of Eo, the reduced species at the electrode surface

become depleted and a diffusion layer of redox

can be plotted with current

The result of the sweep is a voltammogram with a current


olution over the voltage scanning period. At a constant applied potential or with a




is the initial reduced analyte concentration.





represents the concentration gradient, is the potential gradient, and






A is the area of a planar electrode (in cm2). When a pote

, the reduced species at the electrode surface

redox species is

can be plotted with current (I) on the y

voltammogram with a current


. At a constant applied potential or with a

is the standard redox potential for the redox reaction, n is the number of

electrons involved in the redox reaction per molecule, F is the Faraday constant,



The current observed at the applied potential depends on the flux (J)

where D represents the diffusion coefficient for the analyte (in cm2/s), z is the charge


is the potential gradient, and




mathematical expression for the time-dependence of the


When a pote


species is established.

on the y-axis and potential

voltammogram with a current



n is the number of

the Faraday constant,

is the universal gas constant, T is the Kelvin scale temperature, CO is the initial


(J) of the oxidized

/s), z is the charge

of the redox active species, V(x,t) is the hydrodynamic velocity in the x-direction,





dependence of the


When a potential sweep is


established. A graph

axis and potential

voltammogram with a current peak that



20

n is the number of

the Faraday constant, R

is the initial


oxidized

/s), z is the charge

direction,





dependence of the


ntial sweep is


A graph

axis and potential (V)

peak that



21

linear potential sweep, the area of this peak can be used to quantify the

concentration of the analyte in solution. However, in many applied electrochemical

experiments, including the ones performed in the described research, non-linear

potential sweeps were applied to produce peaks that cannot be theoretically linked

to analyte concentration. However, this is a trade-off for sharper and more distinct

peaks than seen in linear potential sweeps. This is the case for the technique of

differential pulse voltammetry (DP Voltammetry), which involves the application of

pulsed potentials superimposed on a linear potential sweep. Current is recorded just

before and right after the pulse is applied so as to allow background charging

processes at the electrode surface that are unrelated to the presence of analyte in

solution to be completed first, thus decreasing measurement noise. The resulting

voltammogram yields peaks whose height rather than area is directly proportional to

analyte concentration,

where σ is and ∆E is the pulse amplitude. The peak potential, Ep, is related

to the polarographic half-wave potential E0.5 by

Inorganic compounds, such as the model ferri-ferrocyanide redox couple, are found

to yield reproducible oxidation and reduction peaks theoretically separated by 59

mV. With cyclic, linear potential sweeps, the peak currents appear proportional to

the square-root of the scan rate (in V/s). The application of potential sweeps,

whether linear or not, to organic molecules usually yields peaks that are not

reversible. An oxidation sweep on organic molecules usually yields peaks that do

not have a symmetrical matching peak in a subsequent reduction sweep and vice

versa. This property usually extends for most biological molecules including

proteins, sugars, fats, and nucleic acids.

22

1.6.1 Electrode Materials: Manufacturing Process, Characteristics,

and Uses

The electrochemistry research that is described in this document was conducted

using disposable electrochemically printed chips from BioDevice Technology and a

renewable glassy carbon electrode from CH Instruments. Each type of electrode has

a distinct manufacturing process and physical characteristics.

The glassy carbon electrode is made through a process called carbonization.

Starting from a pre-moulded phenolic polymer resin (phenol-formaldehyde), the

material is exposed to high temperatures of above 300 °C and up to 1200 °C in an

inert atmosphere over a long period of time55. The process must occur gradually in

order to allow for gaseous oxygen, nitrogen, hydrogen, and other species to slowly

diffuse to the surface in order to avoid the formation of defects (cavities) in the

glassy carbon material56. The result is the very smooth (glassy) carbon surface that

defines this type of electrode. The microscopic structure of the surface consists of

many layers of cross-linked graphite-like sheets arranged as tangled ribbons55.

Although not as conductive as metal electrodes, carbon-based electrodes are

relatively inexpensive and are frequently used in electrochemical studies to take

advantage of their large potential window and the many possibilities of implementing

surface chemistry55. Glassy carbon is very pure, conductive, and impermeable to

gas and chemical reactions57. Upon oxidation, the surface does end up containing

some redox active groups with the chemical adsorption of oxygen and these include

carbonyls and quinone/hydroquinone56. The physical characteristics of glassy

carbon electrodes require a compromise for its use. The glassy carbon surface is

easily renewable after the completion of a measurement through the application of a

polishing step with alumina powder to remove adsorbed species. However, the

glassy carbon surface is not perfect due to the formation of defects in the

carbonization process. There may be small cavities left by escaping gas molecules

that will thereafter harbour adsorbed species such that they cannot be removed

through polishing and may create interfering signals in measurements. Furthermore,

the glassy carbon surface is also very easy to scratch. The scratch would then serve

23

as a defect. The removal of scratches is possible but requires extensive polishing

steps with several different grades of emery paper and alumina powder56.

The advent of screen-printing techniques ushered in the era of disposable

electrochemically printed chips. These printed chips can be produced to include all

three electrodes on one chip or only the working electrode if it is so desired. The

manufacturing procedure is completed in a sequence of steps which include the

application of screens/masks on a plastic substrate. Each mask is patterned to allow

for the deposition of an ink on a specific region of the electrode. This allows for

conductive elements and insulating elements to be placed precisely where needed

in order to make the final product.

Figure 1.8. An illustration of the screen-printing process to make disposable

electrochemically printed chips. Grooves in the mask allow for the squeegee to move the ink

onto precise locations on the plastic substrate.

Usually, entire sheets of electrodes are made simultaneously with screen-printing.

First, a conductive ink, such as silver, gold, or platinum, is used to define the

electrode connectors from the surfaces of the working, counter, and reference

electrodes. Next, a graphite ink is applied with a mask that permits for the deposition

of the working electrode surface58. The graphite ink may then be modified with

biomolecules, metals, nanoparticles, and a variety of other functionalization options

that can improve the detection capability of the finished screen printed electrode. For

24

example, metal nanoparticle electro-deposition is possible on a carbon paste screen

printed electrode surface at -0.4 V by covering the working electrode in a highly

acidic metal chloride solution of gold and platinum59. Finally, an insulating ink may

be applied with a mask that helps to prevent short-circuits between the conductive

leads traveling from the different electrodes when an aqueous sample is applied on

their surface.

Figure 1.9. An illustration of a possible sequence of mask steps to create a screen-printed

electrode. In order, (1) deposits the counter electrode and its lead, (2) deposits the working

electrode and its lead, (3) deposits the reference electrode and its lead, and (4) deposits an

insulating hydrophobic layer across all three of the electrode leads.

The benefits of disposable electrochemically printed chips include their low cost and

disposability, which avoids the possibility of electrode surface contamination

between measurements. The accuracy of the modern screen-printing process,

which is automated and subjected to quality control procedures, means that the

electrodes that are prepared on a single sheet are very similar and should allow for

very reproducible measurements.

25

1.7 Literature Review: Electrochemical Transduction for Detection

of AChE Inhibitors

The detection of AChE inhibitors using electrochemical techniques mostly appears

in academic literature in the form of amperometric and voltammetric measurements

of products from reactions catalyzed by OPH or AChE. Potentiometric,

conductimetric, and impedimetric electrochemical detection of AChE inhibitors has

also occasionally been reported. Potentiometric detection of AChE inhibitors has

traditionally been focused on the measurement of pH changes that result from the

activity of the AChE enzyme. When AChE is inhibited, fewer acetate molecules are

produced from the breakdown of ACh and thus the pH of the solution remains more

acidic (or less basic) when compared to a sample containing uninhibited AChE60.

Not much appears to have changed with potentiometric detection methods of AChE

inhibitors since 1977. In 1998, Mulchandani et al.61 immobilized cell membrane OPH

enzyme-expressing Escherichia coli cells on the glass surface of a pH electrode, an

exquisite step up from previous designs that immobilized the enzyme onto the

electrode surface via covalent bonding and cross-linking. The bacterial cell

immobilization effectively bypassed the necessity to express, purify, collect, and

immobilize the OPH enzyme – all of which are tedious, costly, and time consuming

steps. OPH produces two protons for each organophosphate molecule that is

cleaved by the enzyme and the acidification of the test solution is detectable by the

pH electrode down to the mid-ppb range for several organophosphate compounds

including paraoxon and methyl parathion62. The advent of nanoscale potentiometric

sensors designed to work with metal amplification labels may yet revive the use of

potentiometric detection for AChE inhibitors by greatly improving detection limits63,

but thus far such a system has yet to be designed and reported.

Conductimetric detection of AChE inhibitors is very rarely reported and involves the

measurement of solution conductivity and changes in this property as it reflects the

target analyte. Using AChE as the biological recognition element, Suwansa-ard et

al.64 showed that the performance of conductimetric detectors for AChE inhibitors

26

was comparable to potentiometric detectors. As AChE cleaves ACh into Ch and

acetate, the acetate anions alter the conductivity of the solution. Using

conductimetry and potentiometry, the group obtained the same detection limits and

linear range for carbaryl and carbofuran in the mid-to-low ppb range respectively.

However, conductimetric detection proved to be less sensitive than potentiometric

detection in the defined linear range for each carbamate AChE inhibitor. Another

electrochemical technique seldom seen applied for the detection of AChE inhibitors

is electrochemical impedance spectroscopy (EIS). This technique measures

changes in impedance at the surface of the electrode by means of interpreting the

response of the electrode system to applied alternating voltage at different

frequencies. The current response to the applied voltage will be a sinusoidal wave

that has a phase difference from the voltage sinusoid. The ratio of voltage to current

represents the impedance65. The impedance depends on events that occur at the

electrode surface, such as molecular interactions, adsorption, and redox reactions.

Diffusion and electrode kinetics also play a role in determining the impedance

response. Most commonly, a Nyquist plot is made up of the impedance values for

each scanned frequency with the imaginary component of impedance on the y-axis

and real component on the x-axis. Briefly described, the shape of the Nyquist plot,

usually a semi-circle with a tail, provides information about the state of the electrode

surface. EIS is a very versatile tool for studying events at electrode surfaces, but

perhaps the technique is too complicated for easy integration into a biosensor

design for AChE inhibitor detection when compared to other competing

electrochemical techniques. Instead, voltammetric and amperometric designs are

much more prevalently found in peer-reviewed publications.

Both amperometric and voltammetric measurement systems are easily applied to

AChE inhibitor detection. Amperometric detection involves applying a constant

potential over time that will be sufficient to oxidize or reduce a reporter species in

solution. The current recorded from the redox reaction of the reporter species should

give some indication about the concentration of AChE inhibitors in the sample. Thus,

amperometry allows for real-time measurements of samples as they come into

contact with the electrode surface. Voltammetric detection, in contrast to

27

amperometry, allows the user to scan through an applied potential range in order to

create a voltammogram that shows the peak potential and peak current of the

reporter species undergoing a redox reaction at the electrode surface. Although

voltammetric detection does not offer real-time monitoring of samples, it is not a far

stretch to collect multiple measurements over a short period of time through an

automated process which would provide a pseudo-real-time measurement system.

The advantage of voltammetry over amperometry is the additional information

provided about both the peak potential and peak shape, either of which can change

depending on measurement conditions. With amperometry, only current is

measured without the additional information obtained in voltammetry. If the peak

potential changes, the amperometric current measurements continue to be recorded

at an off-peak potential unless there is an in-built correction system in place to

prevent this from happening. However, amperometry is generally known to be the

more sensitive technique over voltammetry.

The performance of AChE inhibitor sensors featuring amperometric transduction is

impressive. In 1991, Kulys66 reported a simple electrochemically-printed sensor with

a TCNQ-modified graphite surface. TCNQ serves as a mediator that is reduced by

choline or thiocholine and whose oxidation signal intensity can thereafter be

interpreted as a measurement of acetylcholinesterase activity. Butyrylcholinesterase

enzyme was immobilized on the electrode surface and was exposed to S-

butyrylthiocholine as the chosen substrate. It was possible to record measurements

at 100 mV (vs. Ag/AgCl reference) and to detect a decrease in thiocholine

production in the presence of 60 ppb paraoxon concentrations and higher. Schulze

et al.67 demonstrated a 500 ppt detection limit for paraoxon by comparing

amperometric data of thiocholine produced by immobilized acetylcholinesterase

mutants on graphite screen printed electrodes in the presence and absence of

incubation with the insecticide. Laschi et al.68 showed in 2007 that screen-printed

cobalt(II) phthalocyanine- modified carbon electrodes could also be used to measure

thiocholine production by AChE at an applied potential of 100 mV. Using this

technique, a detection limit of 110 ppt was reached for carbofuran with a

measurement time of 15 min. Real-time inhibition measurements have also been

28

demonstrated using an amperometry-based system with gelatin-immobilized

acetylcholinesterase electrodes by Pohanka et al.69 with reported achievable real-

time detection of 200 ppb injected paraoxon with no prior electrode incubation over 1

min of measurement. Marinov et al.70 recently published their research in 2010

where they immobilized AChE on a Poly(acrylonitrile-methylmethacrylate-

sodiumvinylsulfonate) membrane (PAN membrane) loaded with gold nanoparticles.

The PAN membrane protected the enzyme from degradation and prevented

biological fouling of the platinum working electrode surface during measurements of

ATCh at an applied potential of 0.8 V. Between measurements, the PAN membrane

was replaced. This approach was reported to have yielded a detection limit for

paraoxon in the low-ppq range after a 20 min incubation, which is very surprising

considering the high applied potential which should oxidize the ATCh substrate,

thiocholine product, and the enzyme itself.

Although not as commonly seen as amperometry-based detection, voltammetry is a

competitive approach to AChE inhibitor detection. The core design of the biosensor

utilizing voltammetry is usually identical to that of one using the amperometric

detection technique in that an enzyme product’s redox activity is still monitored to

yield information about the activity of the enzyme at the biological recognition

element. Qiu et al.71 applied square wave voltammetry to measure choline

production via 2,6-dichloroindophenol as the redox indicator and reached low ppb

detection levels of parathion-methyl. By immobilizing AChE on TCNQ-modified

screen printed electrodes, Arvinte et al.72 were able to apply DP voltammetry to

detect methyl paraoxon at concentrations approaching 1 ppb. Hernandez et al. used

a similar approach with DP voltammetry using the TCNQ oxidation signal to report a

20 ppt detection limit for carbofuran with a 10 min incubation, which exceeded the

performance of a similar amperometric sensor described previously by Laschi et

al29.

29

1.8 Contributions of this Research Work

The first project was conducted to determine whether two components could be

applied successfully to electrochemically detect common AChE inhibitors. The first

component was the BioDevice Technology unmodified disposable gold

electrochemically printed chip, which would be used to oxidize thiocholine produced

from the AChE-catalyzed breakdown of acetylthiocholine substrate. This chip is

unique in that the gold nanostructures were electrodeposited onto the carbon paste

surface by the developers and no further modifications were performed by our

group. In the cited literature, the gold working electrodes used by various groups

were prepared through the surface immobilization of colloidal gold nanoparticles 70

and featured additional surface modifications of the working electrode with

mediators and the AChE enzyme 66-72. The second component was the AChE

enzyme itself, which was the F345Y mutant selected from a group of N. brasiliensis

enzymes donated to Professor Kerman by his colleague, Professor Bachmann. The

hypothesis was that we could obtain competitive detection limits since we would

utilize the full surface area of the gold nanostructured disposable electrodes in the

absence of any surface modifications, which might block thiocholine’s access to the

electrode surface. Using differential pulse voltammetry, the results showed low-ppb

detection limits of 8 ppb for carbofuran and 10 ppb for paraoxon in 20 mM Tris 100

mM NaCl buffer solution. We also spiked real samples of Highland Creek river water

and 6% milk with 50 ppb and 35 ppb paraoxon respectively to show that real

aqueous samples could be tested successfully with the sensor design. The real

samples conformed to the calibration curve obtained in the Tris buffer. However, in

the process, we discovered that our AChE enzyme worked better in the Highland

Creek river water sample than it did in Tris, which was likely due to the higher ionic

strength in the river sample than in the Tris buffer. This reinforced the importance of

conducting baseline AChE activity tests in each unique real sample composition

prior to operating the sensor in the same type of sample that also contains AChE

inhibitors.

30

The second project was aimed to translate fluorescence results that used Thioflavin

T (ThT), a fluorescent dye, to indirectly detect AChE inhibitors. ThT intercalates

weakly with AChE, which increases its fluorescence. In the presence of AChE

inhibitors, this intercalation is disrupted and the fluorescence of ThT decreases as a

result. We proceeded to examine this intercalation electrochemically through the

oxidation of ThT on a glassy carbon electrode surface. Electrochemical analysis

permits the use of very small concentrations of reagents. In this case, we used 280

nM ThT and 12.5 nM AChE, compared to micromolar amounts that are used in

fluorescence spectroscopy studies. We found that ThT oxidation decreases when

AChE is present on the electrode surface beyond the drop that is expected from

surface fouling. The results suggest that ThT is attaching to the surface of the AChE

enzymes that exist on the electrode surface beyond a 1:1 ratio, suggesting that

interactions occur between ThT and AChE that may remove ThT from solution but

not increase the dye’s fluorescence. We also found that ThT oxidation increases if

the AChE is pre-treated with carbachol or paraoxon, which are believed to disrupt

ThT’s intercalation with AChE. Interestingly, the calibration plots for each AChE

inhibitor were complex, with multiple maxima. The difference in ThT oxidation was

used to detect the presence of carbachol and paraoxon at concentrations down to

10 ppb.

31

2. RESULTS AND DISCUSSION

2.1 Project #1: Gold disposable electrochemical printed chips for

the analysis of acetylcholinesterase inhibition using differential

pulse voltammetry

The first research project discussed in this paper featured gold disposable

electrochemically printed (DEP) chips. These chips were used in a traditional

electrochemical biosensor configuration where AChE was the biological recognition

element and ATCh was the substrate. The enzyme-catalyzed breakdown product,

thiocholine, was oxidized directly on the electrode surface after an incubation step of

the substrate in solution with AChE. The oxidation reaction of thiocholine at a gold

working electrode surface takes advantage of gold-sulphur affinity that facilitates the

measurement of oxidation peak currents superior to that obtainable from carbon-

based electrode surfaces. The activity of AChE in solution was monitored by

comparing thiocholine oxidation currents between inhibitor-containing experimental

samples and control measurements.

Figure 2.1. Illustrations of the gold DEP chips used for the first research project. (a) A gold

DEP chip after sample-loading with 12 µL of solution. (b) A diagram of the gold DEP chip’s

surface features. The actual surface area of the working electrode on the chips was

approximately 3 mm2.

a. b.

32

The major differences between the biosensor reported in this project and other

previously published reports using AChE and ATCh were: 1. The application of new

nanostructured, unmodified gold DEP chips to take advantage of gold-sulphur

chemistry for improved biosensor signal measurements, 2. A lack of enzyme

immobilization on the electrode surface in order to maximize the available surface

area for thiocholine oxidation, and 3. the use of a mutant AChE enzyme at the

biological recognition element.

At the time when the decision was made to undertake this project, the gold DEP

chips had been made available from Bio Device Technology (Japan) compliments of

Professor E. Tamiya (Japan Advanced Institute of Science and Technology).

Furthermore, several Nippostrongylus brasiliensis AChE mutants were also

available at the time, having been kindly donated by Professor T. T. Bachmann

(University of Edinburgh). These mutants were expressed by Schulze et al.67 in an

effort to find AChE enzymes that would show greater sensitivity to particular AChE

inhibitors. Schulze et al.67 had evaluated each of the mutants by their ki constants (in

M-1min-1) relative to that of the wild-type N. brasiliensis AChE enzyme in response to

a variety of AChE inhibitors as shown in the following equation

where E is the concentration of active enzyme after incubation with the chosen

inhibitor, E0 is the initial concentration of the enzyme before incubation, t is the

elapsed time (chosen to be 1 min), and [Inhibitor] is the concentration of the chosen

inhibitor in solution. Higher ki values obtained in that experiment indicated that a

lower concentration of inhibitor was needed to inhibit the enzyme in question by

some arbitrary amount indicated by the natural logarithm ratio on the left side of the

equation. The ratio of constant’s values of any particular N. brasiliensis AChE

mutant (referred to here as ki*) to that of the wild-type AChE (ki) for a chosen AChE

inhibitor indicated whether the mutation made the resulting enzyme more vulnerable

or more protected against the inhibitor.

33

The F345Y mutant (phenylalanine switched to tyrosine at position 345), which was

used for our research project featuring gold DEPs, showed a lower value for its ki*

constant (less sensitivity) in response to both carbofuran and paraoxon compared to

the wild-type N. brasiliensis AChE. The F345Y mutation occurs at the choline

binding site of the enzyme. Although the consequences of this mutation on overall

enzyme structure were not clear, modifications to the choline binding site will affect

the docking ability of substrates and inhibitors at the catalytic triad. The ki*of F345Y

versus the wild-type enzyme was in the same order of magnitude for both of these

AChE inhibitors (1 <ki/ki*< 10), although paraoxon’s measured ki

* was approximately

1-2 times smaller than the ki* for carbofuran.

It was anticipated that lower paraoxon concentrations than carbofuran could be

needed to inhibit the F345Y AChE. This would already have been the case even

without using the mutant AChE since the ki values for the wild-type N. brasiliensis

AChE were reported to be 1.0x106 M-1min-1 for paraoxon and 4.4x105M-1min-1 for

carbofuran. Without the mutation, N. brasiliensis AChE is roughly twice as sensitive

to paraoxon compared to carbofuran. With F345Y, the difference in the enzyme’s

sensitivity between the two inhibitors is diminished almost to unity according to the

measurements made by Schulze et al.67 in 2005.

Electrochemistry of Acetylthiocholine

The electrochemical properties of ATCh were examined as the second step in this

project after determining the Tris buffer baseline control voltammogram, which

showed no significant contaminants that oxidized to produce current beyond the low

nanoampere range. The ATCh molecule has a distinct oxidation voltammogram

featuring a set of peaks at 0.44 V and 0.57 V. These appear to be the primary

diagnostic peaks that are indicative of the ATCh concentration present in solution.

ATCh stocks that have spent a considerable amount of time in storage will show

lower peak currents at these potential values compared to new ATCh stocks since

the molecule is hygroscopic and undergoes hydrolysis over time. Experimentally,

34

hydrolysis only became noticeable after many hours when ATCh was prepared in a

pH 7 buffered solution. In addition to the peaks at 0.44 V and 0.57 V, there is one

more peak that appeared at 0.75 V but usually fluctuated unpredictably with each

different measurement.

Figure 2.2. Three DP voltammograms of 2.1 mM ATCh measured on new gold DEP chips.

Figure 2.3. Three DP voltammograms of the 0.5 nM AChE – 2.1 mM ATCh solution, each

taken after a 10 min incubation period on a new gold DEP chip.

35

When AChE is added into solution with 2.1 mM ATCh and incubated (in this case a

concentration of ~0.5 nM was used for AChE, which does not contribute any

significant oxidation current on the voltammogram), the thiocholine product can be

seen having been oxidized at ~0.25 V on the voltammogram. This peak grows

predictably with increased incubation times, but the other peaks were not found to

be as diagnostically useful.

The general trends that were noticed between the substrate and product

voltammograms was the increase of the oxidation peak current at 0.44 V, the

disappearance of the peak at 0.57 V, and a slight but unstable decrease in the

current intensity of the peak at 0.75 V. These results suggest that the peak at 0.57 V

is likely representative of the ATCh substrate or at least one of its oxidation

processes. The peak at 0.44 V may represent the oxidation current from both ATCh

and the thiocholine product, likely related to the gold-sulphur interaction since this

peak shows a major difference in current intensity between carbon and gold DEP

chips. The gold DEP chips show 3-4 times greater oxidation peak currents at 0.44 V

compared to carbon DEP chips. It is possible that the peak at 0.57 V gets

superimposed at the 0.44 V position. This could be confirmed by oxidizing pure

thiocholine at a concentration that produces a peak of matching intensity at 0.25 V

and then observing the difference in the voltammogram at the 0.44 V position.

Finally, the peak at 0.75 V varies dramatically in peak currents recorded between

DEP chips, suggesting an oxidation process whose likelihood in this experiment

depends too heavily on variability in the chip surfaces. Additional investigation of this

peak using renewable GC or gold working electrodes may yield a better idea of what

this peak represents and any possibility of using it for detection purposes. In the

interest of saving time on measurements, the scanning range was limited to run from

0.1 to 0.6 V.

The thiocholine oxidation peak current at 0.25 V increases in magnitude with

increasing incubation times of the substrate but reaches a maximum either due to

competitive inhibition of the enzyme by the product or due to the depletion of the

substrate. If the AChE is first incubated in solution with an AChE inhibitor prior to

36

incubation with the ATCh substrate, the oxidation peak current of thiocholine at 0.25

V is measured to have a lower magnitude compared to controls with uninhibited

AChE. Furthermore, the peak at 0.44 V appeared to change in shape, again hinting

at a possibility that an oxidation potential overlap of substrate and product exists.

Figure 2.4. A comparison of DP voltammograms showing thiocholine production in a control

measurement (top, black line) with AChE versus AChE treated with 50 ppb carbofuran

(lower, gray line).

The different oxidation peak currents were then used to quantify the AChE enzyme

activity after incubation with various concentrations of the AChE inhibitors paraoxon

and carbofuran.

Biosensor Performance

Experimental results showed that it was possible to get good detection limits for both

paraoxon and carbofuran even when using an AChE mutant (F345Y) that is less

sensitive to both of these inhibitors than enzymes commonly found in such sensors.

Carbofuran was detected at 8 ppb and paraoxon at 10 ppb. The calibration plots that

were constructed from experimental data showed that F345Y was still slightly more

37

sensitive to paraoxon than to carbofuran, with 80 ppb of paraoxon leading to

(80±10)% inhibition versus (50±10)% for the same concentration of carbofuran.

Figure 2.5. The calibration plots of the featured biosensor on the gold DEP chips in

response to carbofuran and paraoxon concentrations in Tris buffer. On the right, in the

paraoxon calibration plot, real samples can be seen as the diamond data series appearing

at ~35 ppb (milk sample) and ~50 ppb (river sample) respectively.

Both calibration plots showed an overall logarithmic response tendency of the

average AChE activity with increasing concentrations, although linear regions were

seen between 10 and 40 ppb for both inhibitors. The logarithmic shape indicates that

the enzymes in solution are being saturated by inhibitors at the higher tested

concentrations during the incubation step. The shape of the calibration curve is

tuneable through the modification of the incubation step length and the

concentration of AChE. Using lower AChE concentrations potentially leads to better

detection limits and saturation of the enzymes at lower inhibitor concentrations.

However, under these conditions less substrate is broken down during a fixed

incubation time and thus less product can be oxidized, which decreases sensor

signal intensity. This has to be compensated by increasing the length of the

incubation step with the substrate. The concentration of the substrate should be

sufficiently in excess to maintain kinetic control and to ensure that product inhibition

of the enzyme is not significant during the incubation step with the substrate. In the

scenario where kinetic control is not achieved, the sensor will have poor

discrimination between concentration values at the lower end of the calibration plot

38

for an inhibitor. If product inhibition of AChE is significant, substrate cleavage will be

halted and the maximum current intensity that is generated during the oxidation of

the electrode will not be realized, which is detrimental to the sensor’s overall signal

to noise ratio.

During the collection of data for this research project, it was found that the solvent

used to dissolve the AChE inhibitor can play a major role in the overall activity of the

AChE enzymes. Carbofuran dissolves very poorly in aqueous solution, which is why

2% DMSO was used to aid in solubility. Higher concentrations of DMSO decrease

AChE activity substantially, such that 20% DMSO results in almost complete

enzyme inhibition in the featured biosensor design. However, other solvent effects

vary and lower concentrations of certain solvents have also been reported to

enhance AChE performance 73. Regardless of the solvent used, it is important to run

inhibitor-free controls using exactly the same solution composition as the sample.

Therefore, all carbofuran measurements of thiocholine production were compared

against carbofuran-free controls with AChE and ATCh in Tris buffer with 2% DMSO.

Figure 2.6. Background DP voltammogram measurements for 6% milk in Tris (dashed line)

and river water in Tris (solid line). Note that the current is in nano-amperes compared to the

micro-ampere values seen in the measurements of 2.1 mM ATCh.

The sensor was challenged with two real samples – a 6% milk sample spiked with

35 ppb paraoxon and a river sample with 50 ppb paraoxon. The milk sample was

39

made by diluting with Tris buffer from 20% cream stock, while the river sample was

diluted the same way. Background voltammogram measurements were taken for

each real sample before the addition of paraoxon and the results showed negligible

amounts of oxidation products in the low nanoampere current ranges.

When the spiked real samples were tested, these samples produced output data

that fit reasonably on the paraoxon calibration curve for the sensor. Interestingly, the

milk sample led to a greater degree of inhibition on average than did the river

sample even with a lower paraoxon concentration. This observation illustrates the

challenge of applying such sensors to real sample analyses, since the operation of

the sensor can be quite unpredictable without careful consideration of the

experimental conditions imposed by the real sample components. Furthermore, both

real samples produced measurements of baseline AChE activity (i.e. in clean, Tris

buffered milk and river water solutions) that was elevated beyond the activity of

AChE in Tris buffer controls.

Figure 2.7. A sample collection of DP voltammogram measurements after the 10 min

AChE-substrate incubation step in Tris buffer (dashed line), river water (black line), and 6%

milk (gray line).

The river sample yielded the greatest AChE production of thiocholine that was

measured to be 30% greater than Tris buffer controls. The milk sample allowed

40

AChE to produce only marginally less thiocholine with the same incubation length,

but the results were more variable possibly due to interference from other proteins

present in the milk. Furthermore, the thiocholine peak recorded in milk was much

sharper around 0.3 V. Additional testing is needed to determine why AChE

performance improves when Tris buffer is mixed with milk or river samples. Intuition

suggests that the increased ionic strength in such real samples plays a major role in

the increase of activity. It has been well-known for quite some time that lower ionic

strength promotes aggregation of AChE enzymes, which may reduce their

performance due to decreased accessibility of substrate to the enzyme active

sites74, 75. However, a more plausible explanation may simply be the effect of ionic

strength on the active site’s catalytic triad, where lower ionic strength destabilizes

the interaction between glutamate and histidine residues76. This leads to poorer

affinity of the substrate at the active site and thus decreased enzyme performance

compared to real samples with higher ionic strengths.

Finally, it is important to discuss the impact of experimental error sources on the

performance of this biosensor. Although the detection limits were stated as being 8

ppb for carbofuran and 10 ppb for paraoxon, it was actually possible to detect

paraoxon as low as 500 ppt and carbofuran at 2 ppb. However, the variability in

measured thiocholine oxidation at these concentrations meant that the error bars for

the data sets overlapped with the relevant control values. Standard deviations of

~10-15% were seen in most measurement sets for any particular inhibitor

concentration. This appears to be the norm for published studies of these types of

biosensors that use AChE as the biological recognition element. For example,

Laschi et al.68 showed standard deviations of ~8% for most of their measurements

with a similar biosensor configuration testing for carbofuran concentrations. With on-

field application of these types of sensors, these error bars will determine the

confidence of an operator in the inhibitor concentration value that is obtained from

any recorded measurement. With large error bars (e.g. my 40 ppb carbofuran

measurement) it becomes difficult to rely on the sensor to distinguish between

significant and insignificant concentrations of AChE inhibitors in a given sample. In

the calibration curves collected for this research project, the error bars of several

41

different measurements overlapped, especially for carbofuran. In retrospect, it would

have been more appropriate to collect measurements of inhibitor concentrations in

magnitude increments (i.e. 10xM) to obtain a data set that includes more signals that

do not have overlapping error bars. Further, more rigorous measurement conditions

could be implemented with more sophisticated experimental settings. These would

include auto-pipetting and controlled mixing of the samples throughout all incubation

steps to ensure that human error in the manipulation of instrumentation did not

contribute towards the standard deviation of measurement sets. The use of a

Faradaic cage enclosure around the DEP chips during electrochemical

measurements would also serve to decrease the influence of background

electromagnetic noise that may have contributed to fluctuations in sensor readings.

With the implementation of these changes, it should be possible to lower the

detection limits and show the true capabilities of this type of sensor.

Summary

This project demonstrated the viability of applying nanostructured gold DEP chips to

AChE inhibitor detection without any additional surface modifications. By keeping

AChE dissolved in solution rather than immobilized on the electrode surface, signal

loss due to the blocking of the electrode surface by the enzyme is avoided.

Furthermore, since the gold surface is permanently modified by the sulphur-gold

interaction with thiocholine, immobilization of the enzyme on the gold surface would

prove relatively costly since the electrode has to be replaced between each

measurement. Immobilization itself is a time-consuming and costly process, so

wasting the immobilized enzyme is not desirable. In order to re-use AChE in this

configuration, it is definitely feasible to apply magnetic separation techniques to

retrieve AChE enzymes from solution between measurements. This technique was

demonstrated by Sole et al.77 where AChE was attached to magnetic microbeads by

means of a bioengineered linker that was introduced into the enzyme at a location

away from the active site. The magnetic microbeads were then retrieved through the

application of a magnetic field during wash steps. This approach provides superior

42

limits of detection in biosensors of AChE inhibitors over those that use traditional

enzyme immobilization on electrode surfaces. When Istanboulie et al.78 bound AChE

to magnetic microbeads using Ni-Histidine affinity, they obtained detection limits 100

times better than those achieved in an identical biosensor that had AChE

immobilized on the electrode surface using a traditional Azide Unit water soluble

polymer (AWP).

It is not implied here that enzyme immobilization is not important in biosensor

designs. Immobilization serves to keep the enzyme from being degraded by

proteases that may appear in solution and from natural unfolding processes that

occur over time if enzyme movement is not restricted. Rather, what is being

contested here is the benefit of enzyme immobilization on the electrode surface. The

active electrode surface area should be optimized and the enzyme can still be

protected from degradation through immobilization on other surfaces that are not

involved in transduction. Furthermore, it is not necessary and arguably not even

desirable for AChE or other enzymes to be present in solution when measurements

are being collected. Hernandez et al.29 reported a comparison of a similar biosensor

using solvated and electrode surface-immobilized AChE. Although it is sometimes

argued that the closer proximity of the enzyme’s products to the electrode surface

may improve product oxidation at the electrode surface, the experimental results of

Hernandez et al. showed poorer sensitivity overall in the design featuring the

electrode surface-immobilized enzyme. On the contrary, the enzymes may adsorb

on the electrode surface, blocking the oxidation of products in the vicinity of the

working electrode, or the enzymes may become degraded when a potential

difference is applied across the electrode. Instead, with a neat sensor design using

magnetic microbeads, the incubation steps with the sample and substrate can take

place in separate solutions. Thereafter, the substrate solution, having reacted with

the enzyme, can be injected onto the electrode surface while AChE remains

magnetically separated from this entire process. This proposed design is illustrated

Figure 2.8:

43

Figure 2.8. Proposed design and illustration for an AChE inhibitor biosensor with AChE

fixed in solution attached to magnetic microbeads. Here, in the diagrams, I is an inhibitor

molecule in the sample, S is acetylthiocholine, and P is thiocholine. After the second

magnetic separation, only the remaining substrate and product are oxidized on the electrode

surface without the AChE enzymes in solution.

With the completion of this research project, it is my perspective that the optimal

biosensor assembly for this purpose is one of a lab-on-a-chip biosensor containing a

wash solution chamber and a substrate solution chamber connected to separate

compartments containing the enzyme and the electrode using microfluidic

technology. However, this type of catalytic-mode biosensor may find competition

from alternative biosensor designs that utilize newly discovered properties of

indicator compounds, which can report indirectly on the presence of AChE inhibitors.

One such indicator compound is Thioflavin T, which was at the focus of my second

research project due to its reported interactions with the peripheral binding site of

the AChE enzyme.

44

2.2 Project #2: Electrochemical Detection of Thioflavin T’s

Interaction with the Acetylcholinesterase Peripheral Binding Site:

Application to the Detection of Acetylcholinesterase Inhibitors

Thioflavin T (ThT) is a positively charged benzothiozole fluorescent dye that is well-

known from the research of amyloid-β aggregation in Alzheimer’s disease. The dye

is incorporated into the β-sheet structure of the amyloid peptides, which alters the

ThT molecular conformation to increase measured fluorescence 79. ThT is also

known to interact with the β-sheet structure of serum albumins80. In 2001, it was

shown that ThT also has a measurable interaction with the AChE enzyme structure.

In the pioneering work of the Rosenberry group using fluorescence spectroscopy, it

was demonstrated that ThT interacts with the AChE peripheral binding site81. This

interaction does not serve to inhibit the function of the AChE enzyme. Furthermore,

Rosenberry et al.82 showed that when inhibitor molecules interacted with the AChE

active site, ThT’s fluorescence decreased. For instance, the fluorescence signal of 1

µM ThT was quenched when exposed to 76 nM AChE that was pre-treated with 0.5

- 60 mM carbachol, a common carbamate pesticide83. This indicated that binding of

inhibitors at the active site modifies ThT’s binding or conformation at the peripheral

binding site of AChE83-89. The second research project which was started near the

end of 2009 set the goal of translating Rosenberry’s research with ThT and AChE

using electrochemical techniques in place of fluorescence spectroscopy.

Figure 2.9. A representation of the chemical structure of ThT (4-(3,6-dimethyl-1,3-

benzothiazol-3-ium-2-yl)-N,N-dimethylaniline chloride).

45

Figure 2.10. An illustration of the three-electrode system used to take measurements for the

ThT-based research project. A 6 mL sample was loaded into a sample container which

could also accommodate the Ag/AgCl glass reference electrode (R), GC working electrode

(W) and the platinum counter electrode (C). The apparatus was placed on a magnetic stir

plate, which was used to stir the sample solution using a magnetic stir bar prior to

measurement collection.

Figure 2.11. DP voltammogram of a 280 nM ThT solution in PBS pH 7.4 on a GC electrode.

The peak appearing near 1.2 V is not diagnostically useful since it is unstable and is near

the upper limit of the GC electrode potential window.

46

ThT Electrochemical Profile

ThT’s electrochemical properties were not well-known when the project was started.

The analysis of ThT using DP voltammetry in a three-electrode system indicated that

ThT undergoes oxidation near 0.9 V on a glassy carbon (GC) electrode surface at

pH 7.4 in a phosphate-buffered solution (PBS). Furthermore, ThT appears to

degrade when in solution over a period of 24 h, leading to substantially decreased

oxidation peak currents unless a new stock solution is prepared every day for each

new experiment. Unlike the DEP chips that were used in the previous research

project, the GC three-electrode system was reusable and required polishing of the

GC working electrode between measurements. It was seen that the ThT oxidation

current intensity increased linearly with increasing ThT concentration between 40

and 400 nM. Higher concentrations than 400 nM are detectable but are not

recommended for three-electrode systems such as the one used for this project

since the glass surface of the Ag/AgCl reference electrode has a significant affinity

for the ThT molecules. With higher ThT concentrations, the reference electrode

quickly acquires a yellow colour as ThT binds to its surface.

Figure 2.12.Calibration plot for the concentration dependence of ThT’s oxidation peak

current with DP voltammetry.

47

Once this exposure to high concentrations of ThT occurs, the reference electrode

begins to contribute to unstable measurements and yields a completely modified

ThT calibration curve if lower concentrations of ThT are then re-tested. The fouled

reference electrode seems to decrease the overall sensitivity of electrochemical

detection – i.e. the slope of the new calibration curve is less steep than the original

calibration curve obtained before the reference electrode becomes fouled. The

affinity of ThT for the reference electrode caused many problems at the onset of this

research project, since it was necessary to find a ThT concentration that had the

highest possible oxidation peak current without causing problems to measurement

reproducibility. After a substantial amount of testing, the ThT concentration was

selected at 280 nM to achieve a reproducible and stable detection platform. In

hindsight, if this research project were to ever be repeated, I would strongly

recommend that the scientist responsible for the project use DEP chips instead of

the three-electrode system. This may allow for the use of higher ThT concentrations

without fear of contaminating the reference electrode.

Apart from a potentially problematic reference electrode, a time-dependence was

seen in the observed ThT oxidation peak current. Oxidation current increased with

increasing incubation time in a predictable manner up to 5 min. If longer incubation

times were applied, the measurement of ThT oxidation current would sporadically

change from current values similar to shorter incubations down to almost no current

at all. This may have occurred due to unexpected events taking place at the open-

circuit potential or possible fouling of the reference electrode with the longer

incubation times. For this reason, incubation was limited to 4 min at open-circuit

potential and the electrode system integrity was checked every three measurements

against a known stock concentration of ThT to ensure that it was still possible to

produce the same oxidation peak current as obtained with the very first

measurement of that solution in the day. This setup proved to be stable for the

purposes of this research project.

In addition to determining the ThT calibration curve with the GC three-electrode

system, it was observed that ThT’s oxidation signal is also pH-dependent.

48

Continuing with a concentration of 280 nM ThT, the optimal oxidation peak current

occurred between pH 7-7.4, with the current falling off on either side of this range.

Changing the pH of the solution also led to changes in the ThT oxidation potential,

with higher pH values reducing the oxidation potential towards 0.8 V, whereas more

acidic pH values increased oxidation potential towards 0.95 V.

Figure 2.13. The pH dependence of 280 nM ThT oxidation peak current between pH 4 and

9. A maximum occurs near pH 7, with the oxidation current dropping off on either side.

Interaction of ThT with AChE

With the electrochemical properties of ThT having been defined for the given

electrochemical detection platform, the interaction of ThT with AChE was probed.

Using electric eel AChE, it was seen that the ThT oxidation peak current dropped

substantially when the enzyme was present in solution. A calibration plot was made

to determine the dependence of 280 nM ThT oxidation peak current on AChE

concentration. The dependence was found to be inversely exponential, with a more-

or-less linear decrease between 0 and 12.5 nM AChE followed by an insignificant

decrease in ThT oxidation current when enzyme concentration was increased to 25

nM. Surprisingly, a particularly reproducible ThT oxidation peak current was

measured with 12.5 nM AChE.

49

Figure 2.14. Calibration plot of 280 nM ThT oxidation peak current as a function of AChE

concentration.

With enzyme economy and reproducibility in mind, 12.5 nM was chosen for the

AChE concentration used in this research project. Previous work done by

Rosenberry et al.82 suggested a 1:1 stoichiometric ratio for the interaction of ThT

and AChE. However, the electrochemical data collected during this project did not

appear to reflect the ratio determined by Rosenberry et al.82 from fluorescent data.

On average, a 12.5 nM AChE solution decreased the oxidation peak current of 280

nM ThT by almost 3-fold. Control measurements were performed to explore the

possible reasons for this observation. The ThT oxidation peak current drops

regardless of whether the measurement is taken with AChE in solution together with

ThT or if the GC working electrode is first incubated in AChE solution and then

washed and transferred into the ThT solution. This indicates that the observed

current decrease is due to events that are occurring with AChE that adsorbs at the

electrode surface. To determine whether this was simply a matter of electrode

surface fouling by the AChE enzyme, the GC electrode was studied through control

experiments designed using cyclic voltammetry and the ferri-ferrocyanide redox

couple.

50

Investigation of the Surface Fouling Hypothesis

The ferri-ferrocyanide redox couple experiences reversible redox events at the GC

electrode surface as iron is repeatedly oxidized and reduced. This was observed

with a 20 mM ferri-ferrocyanide solution. Upon incubation of the GC electrode with

AChE and subsequent washing and exposure of the electrode to the ferri-

ferrocyanide redox cycle, it was seen that AChE had aggregated on the electrode

surface. The current intensity of ferri-ferrocyanide redox began at almost the same

value as the control for the first scan, but the current progressively decreased in the

four subsequent scans. Increasing the AChE concentration yielded a lower current

intensity for the first scan and subsequent scans showed even lower currents.

However, even with an AChE concentration that was more than 50 times larger than

what was used in our DP voltammetry experiments, the ferri-ferrocyanide redox

current intensity had only dropped by ~50%. This observation suggested that

electrode surface fouling was not primarily responsible for the drop in ThT oxidation

signal observed when the working electrode was incubated with AChE.

Figure 2.15. (left) The structure of ThT and (right) the structure of BTA-1, the neutral

derivative of ThT that lacks a methyl substituent on each nitrogen and in the benzothiozole

benzene.

Electrochemical Profile and Interactions of BTA-1

Additional experiments were attempted using BTA-1 in place of ThT. BTA-1 is a very

similar compound to ThT except for BTA-1’s neutral charge and fewer methyl

substituents on the structure. BTA-1 is a more expensive compound that turns out to

be much less stable in an aqueous solution at pH 7.4 than ThT. As a result, BTA-1

solutions had to be made hourly during experiments with the compound.

Furthermore, BTA-1 appears to be less electrochemically active, so a concentration

51

of 860 nM had to be used to achieve comparable oxidation peak currents. When

BTA-1 was placed in solution with AChE and oxidized, a similar drop in oxidation

peak current was seen. However, the pre-treatment of AChE by inhibitors did not

produce any significant regenerative effects on the oxidation peak current of BTA-1

on the GC electrode surface. The lack of an effect on BTA-1’s oxidation at the

electrode surface in the presence of AChE inhibitors disproves the electrode surface

fouling hypothesis. Instead, the experimental results with BTA-1 show that the

charged state of ThT and possibly the methyl substituents contribute to some sort of

unstable interaction with AChE that BTA-1 does not participate in.

Figure 2.16. Sample DP voltammograms showing (a) the oxidation of 860 nM BTA-1, (b)

860 nM BTA-1 with 20 nM AChE and 4 ppm paraoxon, and (c) 860 nM BTA-1 with only 20

nM AChE. The differences between the oxidation peak currents of (b) and (c) were not

found to be significant over several measurements.

If surface fouling by AChE is not the major reason for the observed drop in ThT

oxidation peak current, it is possible that the enzymes that are adsorbed on the GC

electrode surface may be capturing more ThT molecules besides the one that is

known to interact with the peripheral binding site. Fluorescence spectroscopy

studies by Rosenberry et al.82 had indicated a 1:1 stoichiometric ratio for the

interaction of a single ThT molecule with the AChE peripheral binding site of a single

enzyme. This interaction is detected as an increase in the measured fluorescence

levels of ThT when the molecule undergoes a favourable conformational change

during its interaction with the

molecules may also be interacting with the enzyme without experiencing the same

type of conformational cha

electrochemical measurements but not through fluorescence measurements. In this

scenario, not only would the AChE enzymes on the electrode surface be capturing

ThT at the peripheral binding site but also other ThT

the enzyme topology. This hypothesis would need to be investigated using

fluorescence microscopy or atomic force microscopy to visualize the distribution of

ThT on the GC electrode surface after incubation with AChE.

Figure

A maximum oxidation peak current is observed at a paraoxon concentration of 5 ppm. The

bottom graph shows the sensor response at lower paraoxon concentrations. Concentrations

of par

electrode surface. (

showing three maxima ~ 10 ppb, 10 ppm, and 30 ppm respectively. (

voltammogram o

and 7 ppm paraoxon, and (c) 280 nM

A










Figure 2.17. (A)



of paraoxon approaching and exceeding 20 ppm quench the oxidation of ThT at the

electrode surface. (


voltammogram o











Calibration plots of the ThT



aoxon approaching and exceeding 20 ppm quench the oxidation of ThT at the

electrode surface. (B) Calibration plot of the same biosensor for carbachol


voltammogram of (a) 280 nM


during its interaction with the AChE


type of conformational change. This could be what was detected using







Calibration plots of the ThT




) Calibration plot of the same biosensor for carbachol


f (a) 280 nM ThT oxidation, (b) 280 nM

and 7 ppm paraoxon, and (c) 280 nM ThT oxidation with only 12.5 nM AChE.

AChE peripheral site. However,


nge. This could be what was detected using







Calibration plots of the ThT-based AChE






ThT oxidation, (b) 280 nM

ThT oxidation with only 12.5 nM AChE.

B

C

peripheral site. However,





ThT at the peripheral binding site but also other ThT molecules at other locations on




based AChE-inhibitor biosensor for paraoxon.






ThT oxidation, (b) 280 nM ThT oxidation with 12.5 nM


peripheral site. However,





molecules at other locations on




inhibitor biosensor for paraoxon.





showing three maxima ~ 10 ppb, 10 ppm, and 30 ppm respectively. (C

ThT oxidation with 12.5 nM


peripheral site. However, additional ThT












) Calibration plot of the same biosensor for carbachol – this time

C) A sample DP

ThT oxidation with 12.5 nM AChE


52

additional ThT












this time

) A sample DP

AChE

Figure 2.18.

when ThT is exposed to

AChE that was pre

disrupts the docking of the ThT molecule(s) on the enzyme.

AChE and

It was observed that the incubation of AChE with carbachol and paraoxon before

exposure to ThT produces a greater ThT oxidation p

control with untreated AChE. However, the calibra

not linear with 10

concentrations. There appears to be a complex relationship

concentration used to treat AChE and the resulting ThT oxidation peak current after

incubation with the treated AChE. For carbachol,

peak current

30 ppm. Paraoxon produced

led to values below the AChE controls at concentrations beyond 20 ppm

ppm, paraoxon had almost completely

was performed that confirmed paraoxon to decrease ThT oxidation peak current in

the absence of AChE at concentrations approaching and exceeding 20 ppm. This

suggests the possibility of a simple ThT sensor for

The same control test ap

peak current

There was a marked difference in ThT oxidation peak current obtained f

AChE controls and 10 ppb concentrations for either inhibitor. The discrimination of

these concentrations from baseline AChE controls already illustrate very impressive

Figure 2.18. An illustration of the events that may be taking place at the electrode surface

when ThT is exposed to

AChE that was pre


AChE and Pre-




not linear with 10




peak current were








peak current even with carbachol concentrations




An illustration of the events that may be taking place at the electrode surface

when ThT is exposed to (A)

AChE that was pre-treated with inhibitors. The presence of inhibitors in solution with


-Exposure to Inhibitors: Effects on ThT Oxidation




not linear with 10-30% increases in current versus AChE controls at various inhibitor




were obtained with the inhibitor’s concentration at 10 ppb, 10 ppm and








even with carbachol concentrations





AChE that has not been pre

treated with inhibitors. The presence of inhibitors in solution with


xposure to Inhibitors: Effects on ThT Oxidation




30% increases in current versus AChE controls at various inhibitor




obtained with the inhibitor’s concentration at 10 ppb, 10 ppm and

30 ppm. Paraoxon produced a maximum ThT oxidation peak current at 7 ppm






The same control test applied with carbachol showed no






AChE that has not been pre












maximum ThT oxidation peak current at 7 ppm


ppm, paraoxon had almost completely quenched the oxidation of ThT.




plied with carbachol showed no






AChE that has not been pre-treated with any inhibitors and (B)





exposure to ThT produces a greater ThT oxidation peak current compared to the

control with untreated AChE. However, the calibration plots for both inhibitors are




incubation with the treated AChE. For carbachol, three maxima for




quenched the oxidation of ThT.



suggests the possibility of a simple ThT sensor for toxic concentrations of paraoxon.

plied with carbachol showed no

even with carbachol concentrations of 100 ppm





treated with any inhibitors and (B)




eak current compared to the

tion plots for both inhibitors are


concentrations. There appears to be a complex relationship between the inhibitor


maxima for




quenched the oxidation of ThT.



toxic concentrations of paraoxon.

plied with carbachol showed no effect on ThT oxidation

100 ppm and greater






treated with inhibitors. The presence of inhibitors in solution with AChE





between the inhibitor


maxima for ThT oxidation



led to values below the AChE controls at concentrations beyond 20 ppm. At 100

quenched the oxidation of ThT. A control test




effect on ThT oxidation

and greater.

There was a marked difference in ThT oxidation peak current obtained from the



53



AChE





between the inhibitor


ThT oxidation


maximum ThT oxidation peak current at 7 ppm but

At 100

A control test




effect on ThT oxidation

rom the



54

detection capabilities, since regulatory standards require that most AChE inhibitors

be present at 20 ppb and preferably lower in foodstuffs and water8, 9. No lower

concentrations were tested during this research project since the purpose was to

establish that electrochemical monitoring of ThT oxidation in the presence of AChE

could be used to detect AChE inhibitors. Due to time and resource constraints, the

measurement of the true detection limits for this sensor configuration was left for

future experimentation. Complete calibration plots could be obtained using DEP

chips and square-wave voltammetry, which would allow for much faster and reliable

data collection.

Summary

Thus, in this research project, it was shown that the electrochemical oxidation signal

of ThT and BTA-1 decreases in the presence of AChE. Incubation of AChE with

AChE inhibitors, carbachol and paraoxon, prior to incubation with ThT but not BTA-1

led to the collection of DP voltammograms that displayed unique patterns of current

signal intensity above the baseline controls obtained with AChE but below control

measurements with ThT oxidation alone. These results imply that in addition to

fluorescent detection, ThT’s oxidation signal may be useful for the detection of

interactions between insecticides and AChE using electrochemical techniques.

Electrochemical detection of these interactions is preferable since fewer amounts of

reagents are used and greater sensitivity is achievable than with fluorescence

studies of the same phenomenon. This featured electrochemical research may be a

promising tool that can be used to probe AChE interactions with a variety of small

molecule ligands.

55

2.3 CONCLUDING REMARKS AND FUTURE DIRECTIONS

The research projects completed during 2009 and 2010 have produced interesting

results and ample opportunities and ideas for future research. First, a biosensor was

developed using gold DEP chips that could achieve competitive detection limits even

with an AChE enzyme at the biological recognition element that was 5-10 times less

sensitive to the tested inhibitors compared to the wild-type form. Second, a novel

electrochemical detection system for AChE inhibitors was demonstrated through the

monitoring of ThT oxidation peak currents in the presence of AChE exposed to

different inhibitor concentrations. The ThT oxidation peak currents increased when

the dye was incubated with inhibitor-pre-treated AChE. This project produced

unusual results with calibration plots that were not linear but still clearly distinct from

the baseline controls measured in the absence of inhibitors. Thus, such a sensor

would work as a “there or not” detector within the concentration ranges tested for

carbachol and paraoxon. It was shown that this system has the capacity to detect

concentrations of carbachol and paraoxon as low as 10 ppb and possibly lower with

further testing.

Several questions remain that can be answered in future research efforts. With the

first project featuring gold DEP chips, some questions that would be interesting to

see answered include:

1. What kind of performance would be seen in this biosensor with the use of

different mutant strains of the N. brasiliensis AChE enzymes and different

AChE inhibitors? The F345Y variant was less sensitive than the wild-type

AChE for both inhibitors that were tested. However, there exist different

mutants with much higher sensitivities and different activities. Beyond N.

brasiliensis, there exist much more advanced genetically engineered

AChE enzymes that work much faster on their substrates and it would be

interesting to study their application in this type of biosensor. The genetic

engineering of AChE enzymes for the purpose of using them in AChE

inhibitor biosensors involves a trade-off between enzyme activity and

inhibitor sensitivity. With a better understanding of protein folding, it may

56

eventually be possible to engineer complex AChE derivatives that are

specific for chosen inhibitors and are still able to rapidly convert substrate.

2. What is the meaning of the differences in thiocholine oxidation peak

shapes on the voltammograms obtained from real samples such as

solutions with milk versus Tris buffer? In DP voltammetry, peak area does

not have a theoretical analytical value. However, consistent differences in

peak shape under controlled conditions suggests that a diagnostic value

may still exist.

3. How would the biosensor perform if the AChE enzymes were linked to

magnetic beads? Istamboulie et al.78 utilized a TCNQ-modified carbon

paste electrode for their sensor featuring AChE attached to magnetic

beads. It would be interesting to see a comparison of the carbon paste

electrode to the gold DEP chips used in this particular research project.

Furthermore, the use of OPH enzyme attached to magnetic microbeads in

a step-wise sensor design would allow for the conservation of AChE

enzymes if a sample measurement shows OPH activity.

With the second project featuring ThT oxidation as an indicator of AChE

contamination by inhibitors, there is much research needed to explain unusual

observations:

1. The events taking place at the electrode surface that lead to decreased

ThT oxidation peak currents in the presence of AChE remain unresolved.

Likewise, it is also unclear why the ThT oxidation current increases if the

AChE is pre-treated with AChE inhibitors but not with sucrose or ATCh.

Fluorescence microscopy or atomic force microscopy should give

additional insight into this mystery.

2. The separation of AChE from the sample mixture after incubation with

ThT may also clarify the situation taking place at the electrode surface.

This can be compared to results with AChE immobilized on the electrode

surface.

57

3. The biosensor’s response was studied for paraoxon and carbachol. It

would be interesting to see calibration plots of the sensor’s response to

other AChE inhibitors. The existing calibration plots for paraoxon and

carbachol are complex. Explaining the shapes will require more

electrochemical data from different inhibitors together with microscopy

images.

4. ThT is a sulphur-containing compound. Thus far, a GC electrode was

used for the described research. Other types of electrode surfaces should

also be tested, particularly gold and platinum, both of which can benefit

from interacting with the sulphur group.

Without question, the research described in this document has provided me with

much insight and intrigue into the use of AChE in biological sensing. In my opinion,

the most fruitful future directions for research with AChE would be to probe the

enzyme’s interactions with other biological molecules of interest, such as amyloid-β

which has relevance in Alzheimer’s disease. It was previously discussed that AChE

is involved in complex-formation with amyloid-β and other important processes in

living organisms in addition to its traditional role as a regulator of the acetylcholine

neurotransmitter22. The use of unconventional electrochemical detection systems

with sensitive dyes such as ThT or Congo red is not very well-reported at this time

and this research niche could be effectively filled by scientists who are willing to

venture into such unconventional approaches to yield exciting innovative biosensor

designs.

58

3. EXPERIMENTAL DESCRIPTION AND SUPPORTING

MATERIAL

3.1 Project #1: DEP gold chips for the analysis of AChE inhibition

using DP voltammetry

Reagents

Except when stated otherwise, all samples were prepared using a Tris buffer

solution consisting of 20 mM Tris and 100 mM NaCl at pH 7. Paraoxon stock

solutions were prepared by diluting PS-610 Paraoxon oil (1.274 g/mL) purchased

from Sigma-Aldrich in the Tris buffer solution. Carbofuran (Sigma-Aldrich) stock

solutions required 1 part of DMSO per 50 parts of the Tris buffer to assist with

solubility. The F345Y N. brasiliensis AChE mutant was selected since its sensitivity

to carbofuran and paraoxon was comparable based on data from previous studies67.

The substrate, acetylthiocholine iodide (ATCh), was obtained from Sigma-Aldrich.

For the real samples, Highland Creek water that was collected near the University of

Toronto Scarborough and 18% cream (Sealtest Dairy Creamer™) were diluted 3-

fold with Tris buffer. This yielded 1:3 river water in Tris and 6% cream in Tris

(referred to as 6% milk).

Measurement Procedure

Initially, equal volumes (30 µL) of enzyme and either paraoxon or carbofuran

solutions were mixed together and allowed to incubate for 30 min as previously done

by Schulze et al.67. The enzyme-insecticide solutions were briefly mixed through

medium-speed vortexing every 10 min during the incubation. Upon completion of the

incubation time with the insecticide sample, another equal volume (30 µL) of

substrate solution was injected into the enzyme-insecticide solution for the substrate

incubation step. For the studies featuring constant enzyme concentration of 323

mU/mL (~5x10-10 M), the final substrate concentration for ATCh was set at 2.1 mM.

A voltammogram was collected upon completion of a subsequent 10 min incubation

period with the substrate, during which the sample was vortexed every 2 min.

59

The Eco Chemie µAutolab Type III FRA2 Potentiostat/Galvanostat, purchased from

Metrohm Autolab, was used to collect electrochemical data from disposable

electrochemical printed (DEP) chips (Fig. 1A). The DEP chips (DEP-ER-N) were

kindly donated by Professor Eiichi Tamiya from Osaka University and BioDevice

Technology, Japan. The counter electrode and the connecting wires of these DEP

chips were printed using carbon ink. The reference electrode contained Ag/AgCl

paste. The hydrophobic coating between the electrical connectors and the working

area of the chip prevented the contact of liquids with the electrical connections of the

potentiostat. The overall dimensions of the chip were 12.5 mm (Length) x 4 mm

(Width) x 0.3 mm (Thickness).

For each measurement, a 12 µL drop of sample was placed evenly over the working

electrode surface of a new DEP chip. DP voltammetry measurements were taken at

a modulation amplitude of 50 mV, step potential of 5 mV, modulation time of 50 ms,

interval time of 500 ms, and equilibration time of 5 s. The scan range was set

between – 0.1 V and 0.9 V. The data was processed using a moving average

baseline correction with minimum peak height of 0.003 prior to signal peak analysis.

The oxidation peak current of thiocholine was recorded at ~0.3 V.

Inhibition Calculations

For the evaluation of the enzyme inhibition, the TCh peak current at ~0.3 V was

used to detect inhibition by carbofuran and paraoxon. The following expression for

enzyme inhibition was applied:

where Itreatment is the insecticide-containing sample’s TCh current peak amplitude,

while Icontrol refers to the corresponding control sample in the absence of the

insecticide. The detection limit was defined as the smallest concentration of

insecticide in the treatment sample that resulted in a 20% average inhibition of

AChE (I20). Each concentration was evaluated with 3 DEP chips for the Itreatment value

60

and compared to control values (Icontrol) collected through another 3 DEP chips. A

summary of the procedure is shown in Figure 24.

Figure 3.1. The flowchart of the procedure used to collect measurements for the gold DEP

chip research project. Note that three measurements were taken on three new DEP chips

for each insecticide concentration and for each control value.

61

3.2 Project #2: Electrochemical Detection of ThT’s Interaction with

the AChE Peripheral Binding Site: Application to the Detection of

AChE Inhibitors

Reagents

All reagent solutions were prepared in a 20 mM phosphate 100 mM NaCl buffer

solution at pH 7.4 unless otherwise specified. Purified 18.2 MΩ water from the PAL

Cascada LS purification system was used to make all aqueous solutions. Thioflavin

T (T3516) and BTA-1 (B9934) were obtained from Sigma Aldrich. Electric eel AChE

(C3389), paraoxon (PS610), and carbachol (C2409) were all obtained from Sigma.

Sample Preparation Steps and DP Voltammetry

A conventional three-electrode system from CH Instruments, consisting of a CH104

GC working electrode (GCE), CH111 Ag/AgCl reference electrode, and CHI115

platinum wire counter electrode, was connected to an Eco Chemie µAutolab Type III

FRA2 Potentiostat/Galvanostat (Metrohm International). The General Purpose

Electrochemical System (GPES) software by Eco Chemie was used as the data

collection interface.

Prior to each oxidation voltammogram measurement, a 6 mL sample solution (280

nM ThT together with 12.5 nM AChE and/or variable concentrations of paraoxon or

carbachol) was prepared and lightly stirred for 1 min on a magnetic stir plate. This

was followed by the introduction of the electrode system and incubation with stirring

for an additional 4 min at the open-circuit potential (OCP). At this point, the stirrer

was turned off. The sample was then oxidized on the GCE from 0-1.2 V using

differential pulse voltammetry (DP Voltammetry) at a modulation amplitude of 50

mV, step potential of 5 mV, modulation time of 50 ms, interval time of 500 ms, and

equilibration time of 10 s. The oxidation voltammogram was processed using the

GPES moving average baseline correction tool with a selected minimum peak height

of 0.003 prior to peak analysis using the peak search function. Between

measurements, the GCE surface was renewed via polishing on a polishing pad for 1

62

min in a suspension of 0.05 micron Gamma Alumina Powder from CH Instruments

and 18.2 MΩ Cascada water. On average, each data point consists of three to four

measurements (n ≥ 3).

Calibration Plot Construction for ThT Concentration Dependence

Five concentrations of ThT between 40 and 400 nM were oxidized on the GCE, with

the peak current measured at 0.87 V to make the calibration plot for ThT

concentration dependence. A linear response was measured in this concentration

range for ThT oxidation. Measurements of higher concentrations were attempted,

but this altered the values obtained for the recorded linear range when those

concentrations were re-tested. A yellow discoloration or the reference electrode was

clearly visible after these measurements at higher ThT concentrations were

collected. Polishing and/or replacement of the GC working electrode did not have a

significant effect on the measurements. Replacement of the glass Ag/AgCl reference

electrode returned measurements to normal expected readings. As a result, higher

concentrations were avoided in subsequent experimentation steps.

Calibration Plot Construction for pH Dependence of 280 nM ThT Peak

Oxidation Current

Solutions of 280 nM ThT were prepared in buffer solutions set at pH 4, 5, 6.5, 6.8, 7,

7.4 and 9. Of these pH values, a 20 mM acetic acid / sodium acetate buffer was

used to prepare the pH 4 and 5 buffer, 20 mM PBS buffer was used for pH 6.5, 6.7,

7 and 7.4 buffer, and finally 20 mM Tris was used for the pH 9 buffer. Each buffer

solution also contained 100 mM NaCl electrolyte as with the PBS buffer mentioned

earlier. Measurements were taken as described previously (n≥3).

Calibration Plot Construction for 280 nM ThT Peak Oxidation Current

Dependence on AChE Concentration

Five AChE concentrations were tested for their effect on the 280 nM ThT peak

oxidation current. AChE stock solutions were prepared via serial 2x dilutions starting

63

from 25 nM and proceeding through 12.5 nM, 6.25 nM, 3.125 nM, and 1.563 nM

respectively. Measurements were then collected as previously described (n≥3).

Control Experiments

Impact of Paraoxon and Carbachol Concentrations on ThT Oxidation Signal

Intensity

In this experiment, 280 nM ThT was incubated for 5 min with varying concentrations

of paraoxon and carbachol before DP voltammetry measurement. No AChE was

added at any point during this control experiment. Carbachol showed no significant

effect on the ThT oxidation signal even at concentrations as high as 100 ppm.

Paraoxon showed no significant effects on the ThT signal when present at low

concentrations approaching 20 ppm. At concentrations higher than 20 ppm,

paraoxon had the effect of decreasing the intensity of the ThT signal substantially.

This explains the drop in ThT oxidation when AChE is pre-treated with higher

paraoxon concentrations. The inhibition of ThT’s oxidation signal by ppm levels of

paraoxon is worth investigating in the future for a simple detection method of toxic

levels of this insecticide.

ThT Adsorption and Oxidation Signal Intensity

The GC electrode was incubated for four min in a 280 nM ThT solution. The GC

electrode was then removed from the 280 nM ThT solution and rinsed with PBS

buffer. The GC electrode was then placed in a sample of clean PBS buffer

solution. The DP voltammogram was immediately collected to see how much of

the ThT signal was due to ThT aggregation at the GC electrode surface. The ThT

peak intensity of the DP voltammetry measurement in clean buffer solution was

80% as big as the measurement taken in ThT solution. This shows that there is a

high level of ThT adsorption taking place at the GC electrode prior to a

measurement being made. Therefore, this control experiment shows that most of

the recorded current is due to the adsorption of ThT on the electrode surface.

64

Figure 3.2. DP voltammetry oxidation signal taken in 280 nM ThT solution (a) and in

PBS buffer after rinsing the GC electrode (b).

When 12.5 nM AChE and 280 nM ThT are tested in the same way, the same

results are found. The DP voltammetry measurement taken in a clean PBS

solution after prior GC electrode incubation with ThT and AChE produces a ThT

oxidation peak that is 80% as big as the measurement recorded in solution with

ThT and AChE.

Figure 3.3. DP voltammetry oxidation signal taken in 280 nM ThT and 12.5 nM AChE

solution (a) and in PBS buffer after rinsing the GC electrode (b).

These results suggest that most of the ThT oxidation signal that is observed is

65

determined by ThT’s adsorption behaviour at the surface of the electrode during

the 4 min incubation step. The same proportion (80%) of ThT adsorbs to the

electrode surface when AChE is present in solution.

Investigation of Surface Fouling by AChE

It is possible that the observed drop in ThT oxidation signal in the presence of

AChE is due to surface fouling of the GC electrode by AChE. AChE would block

ThT’s access to the electrode surface for oxidation. Several control experiments

were designed and tested to investigate this possibility.

1. Separate GC electrode incubation steps

In this experiment, the GC electrode was incubated first in 12.5 nM AChE

solution for 4 min. The electrode was then removed and rinsed with PBS

buffer solution. This was followed by an incubation of the GC electrode for

another 4 min in 280 nM ThT solution. The DP voltammetry measurement

was then taken in the 280 nM ThT solution. This produced similar results

as seen with DP voltammetry measurements taken in a solution of 280 nM

ThT and 12.5 nM AChE.

These results indicated that the drop in ThT oxidation signal intensity does

not require AChE to be in solution with ThT. The signal decrease is

presumably seen with AChE adsorbed on the surface of the GC electrode.

The question that followed was whether AChE blocked the access of ThT

to the GC electrode surface, thus leading to the measured reduction in ThT

oxidation peak current.

2. Ferri-Ferrocyanide Cyclic Voltammograms

Cyclic voltammograms (CVs) with ferri-ferrocyanide allow for the

measurement of the Fe2+/3+ redox processes taking place at the electrode

surface. The measurements are often used to characterise modified

electrode surfaces, where a drop in ferri-ferro redox activity would indicate

66

the presence of adsorbed or bound species on the electrode surface. The

CV measurements taken in the following control experiments used 5

standard linear scans from -0.5to 1.25 V, with characteristic Fe2+/3+ redox

occuring at ~ 0.23 V and 0.27 V respectively.

i. Testing ferri-ferrocyanide accessibility to the GC electrode

surface after incubation with 12.5 nM AChE.

The GC electrode was incubated for 4 min in 12.5 nM AChE

solution. It was then washed with PBS bufer solution and placed into

20 mM ferri-ferrocyanide soluion. CVs were measured and showed

the first scan to be very close to the CV measurements in clean 20

mM ferri-ferrocyanide solution. Scans 2-5 showed a progressive

decrease in redox couple intensity, with the fifth scan showing a

20% drop in intensity compared to scan 1.

Figure 3.4. CVs of 20 mM ferri-ferrocyanide with a 4 min pre-incubation of the GC

electrode in PBS buffer solution (black) and 12.5 nM AChE solution (red).

ii. Testing ferri-ferrocyanide accessibility to the GC electrode

surface after incubation with 200 nM AChE.

The same experiment was repeated but with 200 nM AChE. The

67

same pattern was observed, but with an overall 40% drop in the

redox couple intensity after 5 scans compared to the 20% drop seen

with 12.5 nM AChE in the previous experiment..

iii. Testing ferri-ferrocyanide accessibility to the GC electrode

surface after incubation with 600 nM AChE.

When this experiment was repeated with 600 nM AChE, there was a

50% drop in the redox couple intensity after 5 scans. In addition, the

first scan now showed a 20% drop in redox couple intensity

compared to the control measurements made on a clean GC

electrode in 20 mM ferri-ferrocyanide solution.

Figure 3.5. CVs of 20 mM ferri-ferrocyanide after 4 min pre-incubations of the GC

electrode in 12.5 nM AChE (black), 200 nM AChE (red), and 600 nM AChE (blue).

Even with 600 nM AChE, the GC surface is not saturated with the enzyme at the time

of measurement.

In summary, the first CV scan showed that the ferri-ferrocyanide molecules

had easy access to the GC electrode at the experimental AChE

concentrations. Only when the AChE concentration was ~60 times higher

(600 nM vs 12.5 nM) was there a substantial change in the first CV scan

intensity when compared to clean GC electrode controls in 20 mM ferri-

68

ferrocyanide solution. Furthermore, even after 5 scans, the ferri-

ferrocyanide species could still undergo their redox processes with 50% of

the initial signal magnitudes even at 600 nM AChE. This shows that the

electrode surface is not saturated at this concentration or at the

experimental 12.5 nM AChE conditions.

These results suggest that, with our experimental conditions of 12.5 nM

AChE, the enzyme does not block a substantial portion of the electrode

area to justify the observed ~60% drop in ThT oxidation signal intensity.

Instead, the ThT molecules may be preferably interacting with AChE

before oxidizing on the surface. The native tetrameric form of AChE may

be creating a surface modification that effectively captures ThT from

solution near the electrode surface.

iv. Testing ferri-ferrocyanide accessibility to the GC electrode

surface after incubation with 100 ppm carbachol.

In this experiment, the GC electrode was incubated in 100 ppm

carbachol for 4 min. The electrode was rinsed in PBS buffer solution

and then placed into 20 mM ferri-ferrocyanide solution, where CV

measurements were taken. The incubation with carbachol showed

no effect on the redox processes of ferri-ferrocyanide in any of the 5

scans even at this high concentration, indicating that carbachol does

not adsorb significantly to the electrode surface.

69

Figure 3.6. CVs of 20 mM ferri-ferrocyanide after a 4 min pre-treatment of the GC

electrode in PBS buffer solution (black) and 100 ppm carbachol solution (red).

v. Testing ferri-ferrocyanide accessibility to the GC electrode

surface after incubation with 100 ppm carbachol-pre-treated

12.5 nM AChE.

The same procedure was followed as in (iv) to see the effects of

incubating the GC electrode in a solution of 12.5 nM AChE and 100

ppm carbachol. The CVs that were measured were not substantially

different from the CVs collected in (i) for clean 12.5 nM AChE.

70

Figure 3.7. CVs of 20 mM ferri-ferrocyanide after a 4 min pre-incubation of the GC

electrode in 12.5 nM AChE (black) and 12.5 nM AChE with 100 ppm carbachol (red).

No significant differences are observed in any of the 5 scans.

These results suggest that the changes in ThT signal intensity

observed in our studies in the presence of insecticides are not a

result of the acetylcholinesterase inhibitors somehow detaching the

enzyme from the electrode surface. The electrode surface is

essentially the same whether carbachol is present or not.

vi. Testing ferri-ferrocyanide accessibility to the GC electrode

surface after incubation with 200 nM AChE followed by another

incubation with 280 nM ThT.

This experiment featured the incubation steps described in (ii) but

also included an extra step where the GC electrode was washed

with PBS buffer solution and then incubated for another 4 min in 280

nM ThT solution. The GC electrode was then washed again with

clean PBS buffer solution and placed in 20 mM ferri-ferrocyanide

solution for CV measurements. The ferri-ferrocyanide CVs obtained

in (i) showed redox couple current intensities that were 10% greater

71

than those obtained after incubation with 280 nM ThT.

Figure 3.8. CVs of 20 mM ferri-ferrocyanide after a 4 min pre-incubation of the GC

electrode in 200 nM AChE solution (black) compared with the same pre-incubation in

200 nM AChE followed by another 4 min incubation in 280 nM ThT (red). The ThT

adsorbs to the GC electrode surface and blocks ferri-ferrocyanide redox processes by

an additional ~10% over AChE.

These results indicate that the ferri- and ferro-cyanide molecules are

still easily able to reach the electrode surface even in the presence

of adsorbed species of both 280 nM ThT and 200 nM AChE on the

GC electrode surface. Experimentally, 12.5 nM AChE was used, so

there was more opportunity for ThT to reach the GC electrode

surface.

Overall, the ferri-ferrocyanide CV control experiments suggest that surface

fouling by AChE is not the primary reason behind the measured drop in ThT

oxidation at the GC electrode surface.

Determining the Selectivity of the Biosensor for AChE Inhibitors

The ThT oxidation peak current increases when AChE is pre-treated with varying

72

concentrations of carbachol and paraoxon (the latter below 20 ppm). As a

selectivity control, ThT oxidation was monitored when AChE was pre-treated with

ATCh (the substrate used in the first research project) and sucrose. ATCh binds

to the active site of AChE, but its presence in solution at low concentrations

(<<0.1 mM) did not affect the pure 280 nM ThT oxidation or the decrease in ThT

oxidation that is seen when AChE is in solution. At higher concentrations of

ATCh, the compound’s oxidation dwarfs the oxidation signal from 280 nM ThT.

Likewise, tests with 20 mM sucrose showed no impact on ThT’s oxidation signal

in the presence and absence of 12.5 nM AChE. Thus, the ThT-based biosensor

appears to show selectivity for the presence of paraoxon and carbachol and not

for sucrose or ATCh.

73

4. Literature Cited

1. J. de Bleecker, Organophosphate and carbamate poisoning. In Neuromuscular

junction disorders, Eds: A. G. Engel, Handbook of Clinical Neurology Series 3,

Elsevier B. V. New York, N.Y., 2008, Vol. 91, pp. 401 – 432.

2. T. Satoh, Global Epidemiology of Organophosphate and Carbamate Poisonings.

In Toxicology of Organophosphate & Carbamate Compounds, Eds: R. C. Gupta,

Elsevier: Oxford, 2006, pp. 89-100.

3. B. Ballantyne, H. Salem, Occupational Toxicology and Occupational Hygiene

Aspects of Organophosphate and Carbamate Anticholinesterases with Particular

Reference to Pesticides. In Toxicology of Organophosphate & Carbamate

Compounds, Eds: R. C. Gupta, Elsevier: San Diego, 2006, pp. 567-595.

4. K. Olson, “Aum Shinrikyo: Once and Future Threat?”, 1999, Retrieved July 17,

2010 from: http://www.cdc.gov/ncidod/eid/vol5no4/olson.htm

5. C. Smulders, T.Bueters, S. Vailati, R. van Kleef, H. Vijverberg, Toxicol. Sci. 82

(2004) 545-554.

6. D. Ray, P. Richards, Toxicol. Lett. 120 (2001) 343.

7. Health Canada, List of Maximum Residue Limits Regulated Under the Pest

Control Products Act, 2009, Retrieved July 17, 2010 from: http://www.hc-

sc.gc.ca/cps-spc/alt_formats/pdf/pest/part/protect-proteger/food-nourriture/mrls-lmr-

eng.pdf

8. European Union Pesticides database, Pesticide residues for Carbofuran and

Parathion-methyl, 2009, Retrieved July 17, 2010 from:

http://ec.europa.eu/sanco_pesticides/public/index.cfm?event=substance.selection

9. Electronic Code of Federal Regulations, Title 40: Protection of Environment, Part

180 – Tolerances and Exemptions for Pesticide Chemical Residues in Food,

Subpart C – Specific Tolerances for Carbofuran (180.254) and Methyl Parathion

74

(180.121), Retrieved July 17, 2010 from: http://ecfr.gpoaccess.gov/cgi/t/text/text-

idx?c=ecfr&tpl=%2Findex.tpl

10. K. W. Hunter, D. E. Lenz, Life Sci. 30 (1982) 355.

11. A. Mulchandani, W. Chen, P. Mulchandani, J. Wang, K. R. Rogers, Biosens.

Bioelectron. 16 (2001) 225.

12. W. R. Algar, U. J. Krull, Quantum Dots for the Development of Optical

Biosensors Based on Fluorescence. In Biosensing Using Nanomaterials, Eds: A.

Merkoçi. John Wiley & Sons, USA, 2009, pp. 200-201

13. C. R. Lowe, Overview of Biosensor and Bioarray Technologies. In Handbook of

Biosensors and Biochips, Eds: R.S. Marks, D. C. Cullen, I. Karube, C. R. Lowe, H.

H. Weetal. John Wiley & Sons, USA, 2007, pp. 1-4.

14. NCBI Entrez, ACHE acetylcholinesterase (Yt blood group) [ Homo sapiens ] Gene ID: 43, Retrieved on July 16, 2010 from: http://www.ncbi.nlm.nih.gov/gene/43 15. D. Alonso, I. Dorronsoro, L. Rubio, P. Munoz, E. Garcia-Palomero, M. Del Monte, A. Bidon-Chanal, M. Orozko, F. J. Luque, A. Castro, M. Medina, A. Martinez, Bioorg. Med. Chem. 13 (2005) 6588. 16. H. A. Berman, M. M. Decker, J. Biol. Chem. 261 (1986) 10646. 17. A. Saxena, D. Maxwell, D. Quinn, Z. Radic, P. Taylor, B. Doctor, Biochem. Pharmacol. 54 (1997) 269. 18. M. Q. Yi, H. X. Lui, X. Y. Shi, P. Liang, X. W. Gao, Comp. Biochem. Physiol. Part C 143 (2006) 113. 19. G. Mastrantonio, H. G. Mack, C. Omar, D. Vedova, J. Molec. Model. 14 (2008) 813. 20. J. L. Sussman, I. Silman, Curr. Opin. Struct. Biol. 2 (1992) 721.

21. L. Dziri, S. Boussaad, S. Wang, R. M. Leblanc, J. Phys. Chem. B 101 (1997)

6741.

22. H. Soreq, S. Seidman, Nat. Rev. Neurosci. 2, 4 (2001) 294.

75

23. M. Mehrvar, C. Bis, J. Scharer, M. M. Young, J. H. Luong, Anal. Sci. 16 (2000)

677.

24. K. Kerman, M. Vestergaard, E. Tamiya, Electrochemical DNA biosensors:

Protocols for intercalator-based detection of hybridization in solution and at the

surface, In Biosensors and Biodetection: Methods and Protocols Vol. 2:

Electrochemical and Mechanical detectors, lateral flow and ligands for biosensors,

Eds: A. Rasooly, K. E. Herold. Humana Press, USA, 2009.

25. B. di Sioudi, C. E. Miller, K. Lai, J. K. Grimsley, J. R. Wild, Chem. Biol. Interact.

119 (1999) 211.

26. J. Stojan, V. Marcel, S. Estrada-Mondaca, A.Klaebe, P. Masson, D. Fournier,

FEBS Lett 440 (1998) 85.

27. R. Kaushik, C. A. Rosenfeld, L. G. Sultatos, Toxicol. Appl. Pharmacol. 221

(2007) 243.

28. S. Zhang, H. Zhao, R. John, Biosens. Bioelectron.16 (2001) 1119.

29. S. Hernandez, I. Palchetti, M. Masscini, Int. J. Environ. Anal. Chem. 78 (2000)

263.

30. D. E. Lenz, A. A. Brimfield, K. W. Hunter, H. P. Benschop, L. P. A. de Jong, C.

van Dijk, T. R. Clow, Fundam. Appl. Toxicol.4 (1984) S156.

31. S. Liu, L. Yuan, X. Yue, Z. Zheng, Z. Tang, Adv. Powder Technol. 19, 5 (2008) 419. 32. P. Zhou, Y. Lu, J. Zhu, J. Hong, B. Li, J. Zhou, D. Gong, A. Montoya, J. Agric.

Food Chem. 52 (2004) 4355.

33. K. Rogers, E. Poziomek, Chemosphere 33 (1996) 1151.

34. B. White, H. Harmon, Biosens. Bioelectron. 20 (2005) 1977.

35. N. Nagatani, A. Takeuchi, M. A. Hossain, T. Yuhi, T. Endo, K. Kerman, Y.

Takamura, E. Tamiya, Food Control 18 (2007) 914.

76

36. V. Vamvakaki, D. Fournier, N. A. Chaniotakis, Biosens. Bioelectron. 21 (2005)

384.

37. T. J. Dale, J. Rebek, Angew. Chem. Int. Ed. 121 (2009) 7990.

38. Costa-Fernandez, R. Pereiro, A. Sanz-Medel, TrAC, Trends Anal. Chem. 25

(2006) 207.

39. C. A. Constantine, K. M. Gatta’s-Asfura, S. V. Mello, G. Crespo, V. Rastogi, T.

C. Cheng, J. J. DeFrank, R. M. Leblanc, J. Phys. Chem. B 107 (2003) 13762.

40. V. Pardo-Yissar, E. Katz, J. Wasserman, I. Willner, J. Am. Chem. Soc. 125

(2003) 622

41. A. Virel, L. Saa, V. Pavlov, Anal. Chem. 81 (2009) 268.

42. Y. Xiao, V. Pavlov, B. Shlyahovsky, I. Willner, Chemistry 11 (2005) 2698.

43. A. Ischii, S. Takeda, S. Hattori, K. Sueoka, K. Mukasa, Colloids Surf., A 313-314

(2008) 456.

44. N. Kim, I. Park, D. Kim, Biosens. Bioelectron. 22 (2007) 1593.

45. A. Brecht, G. Gauglitz, Biosens. Bioelectron. 10 (1995) 923.

46. Rajan, S. Chand, B. D. Gupta, Sens. Actuators B, 123 (2007) 661.

47. E. Mauriz, A. Calle, L. M. Lechuga, J. Quintana, A. Montoya, J.J. Manclus, Anal.

Chim. Acta 561 (2006) 40.

48. L. Pogacnik, M. Franko, Biosens. Bioelectron. 18 (2003) 1.

49. K. Mawatari, Y. Naganuma, K. Shimoide, Anal. Chem. 77 (2005) 687.

50. M.D. Luque De Castro, M. C. Herrera, Biosens. Bioelectron. 18 (2003) 279.

51. T. T. Bachmann, B. Leca, F. Vilatte, J. L. Marty, D. Fournier, R. Schmid,

Biosens. Bioelectron. 15 (2000) 193.

77

52. J. Wang, Analytical Electrochemistry, third ed., Wiley-VCH, 2000. Chapter 1, pp.

3 – 13.

53. A. J. Bard, L. R. Faulkner, Electrochemical Methods: Fundamentals and

Applications, second ed., Wiley, 2000.

54. I. Palchetti, S. Laschi, M. Mascini, Electrochemical Biosensor Technology:

Application to Pesticide Detection, In Biosensors and Biodetection: Methods and

Protocols Vol. 2: Electrochemical and Mechanical detectors, lateral flow and ligands

for biosensors, Eds: A. Rasooly, K. E. Herold. Humana Press, USA, 2009.

55. J. Wang, Analytical Electrochemistry, third ed., Wiley-VCH, 2000. Chapter 4, pp.

113 – 114.

56. W. E. van der Linden, J. W. Dieker, Anal. Chim. Acta 119 (1980) 1.

57. T. Shigemitsu, G. Matsumoto, S. Tsukahara, Med. & Bio. Eng. & Comput. 17

(1979) 465.

58. A. B. Florou, M. I. Prodromidis, S. M. Tzouwara-Karayanni, M. I. Karayannis,

Anal. Chim. Acta 423 (2000) 107.

59. M. Chikae, K. Idegami, K. Kerman, N. Nagatani, M. Ishikawa, Y. Takamura, E.

Tamiya, Electrochem. Comm. 8 (2006) 1375.

60. W. N. Dahl, A. M. R. Holm, K. H. Schroder, Anal. Chim. Acta 93 (1977) 91.

61. A. Mulchandani, P. Mulchandani, I. Kaneva, W. Chen, Anal. Chem. 70 (1998)

4140.

62. P. Mulchandani, A. Mulchandani, I. Kaneva, W. Chen, Biosens. Bioelectron. 14

(1999) 77.

63. E. Bakker, E. Pretch, Trends Anal. Chem. 27 (2008) 7.

64. S. Suwansa-ard, P. Kanatharana, P. Asawateratanakul, C. Limsakul, B.

Wongkittisuksa, P. Thavarungkul, Biosens. Bioelectron. 21 (2005) 445.

78

65. M. Prodromidis, Electrochim. Acta 55 (2010) 4227.

66. J. Kulys, E. J. D’Costa, Biosens. Bioelectron. 6 (1991) 109.

67. H. Schulze, S. B. Muench, F. Villatte, R. Schmid, T. Bachmann, Anal. Chem. 77

(2005) 5823.

68. S. Laschi, D. Ogonczyk, I. Palchetti, M. Mascini, Enzyme Microb. Technol. 40

(2007) 485.

69. M. Pohanka, D. Jun, K. Kuca, Sensors 8 (2008) 5303.

70. I. Marinov, Y. Ivanov, K. Gabrovska, T. Godjevargova, J. Mol. Catal. B: Enzym.

62 (2010) 66.

71. J. Qiu, J. Chen, Q. Ma, Y. Miao, Chemosphere 77 (2009) 129.

72. A. Arvinte, L. Rotariu, C. Bala, Development of a pesticide biosensor using

carbon-based electrode systems. In Chemicals as Intentional and Accidental Global

Environmental Threats, NATO Security through Science Series, Springer:

Netherlands, 2006, pp. 337-343.

73. N. Mionetto, J. L. Marty, I. Karube, Biosens. Bioelectron. 9 (1994) 463.

74. J. Cartaud, S. Bon, J. Massouli, J. Cell Biol. 77 (1978) 315.

75. M. A. Grafius, D. B. Millar, Biochim. Biophys. Acta 110 (1965) 540.

76. H. J. Nolte, T. L. Rosenberry, E. Neumann, Biochemistry 19 (1980) 3704.

77. S. Sole, A. Merkoci, S. Alegret, Trends Anal. Chem. 20 (2001) 102.

78. G. Istamboulie, S. Andreescu, J. L. Marty, T. Noguer, Biosens. Bioelectron.23, 4

(2007) 506.

79. A. J. Veloso, V. Hung, G. Sindhu, A. Constantinof, K. Kerman, Anal. Chem. 81

(2009) 9410.

79

80. P. Sena, S. Fatimaa, B. Ahmada, R. H. Khan, Spectrochim. Acta, Part A 74

(2009) 94.

81. G. V. De Ferrari, W. D. Mallender, N. C. Inestrosa, T. L. Rosenberry, J. Biol.

Chem. 276 (2001) 23282.

82. T. L. Rosenberry, J. L. Johnson, B. Cusack, J. L. Thomas, S. Emani and K. S.

Venkatasubban, Chem. -Biol. Interact.157 (2005) 181.

83. T. L. Rosenberry, L. K. Sonoda, S. E. Dekat, B. Cusack and J. L. Johnson,

Chem.-Biol. Interact. 175 (2008) 235.

84. S. A. Kardos and L. G. Sultatos, Toxicol. Sci. 58 (2000 118.

85. S. A. Kousba, L. G. Sultatos, T. S. Poet, C. Timchalk, Toxicol. Sci. 80 (2004)

239.

86. C. A. Rosenfeld, L. G. Sultatos, Toxicol. Sci. 90 (2006) 460.

87. R. Kaushik, C. A. Rosenfeld, L. G. Sultatos, Toxicol. Appl. Pharmacol. 221

(2007) 243.

88. L. G. Sultatos, R. Kaushik, Toxicol. Appl. Pharmacol. 230 (2008) 390.

89. M.Harel, L. K. Sonoda, I. Silman, J. L. Sussman, T. L. Rosenberry, J. Am.

Chem. Soc. 130 (2008) 7856.

Dounin Vladimir v 201011 MSc Thesis

Documents

Transcript of Dounin Vladimir v 201011 MSc Thesis