Electrochemical Measurements of Salivary Amylase Activity

Department of Physics, Chemistry and Biology

Master’s Thesis

Electrochemical Measurements of Salivary

Alpha-Amylase Activity

Henrik Höckerdal

LITH-IFM-A-EX--12/2577--SE

Department of Physics, Chemistry and Biology Linköpings universitet

Master’s Thesis

LITH-IFM-A-EX--12/2577--SE

Electrochemical Measurements of Salivary

Alpha-Amylase Activity

Henrik Höckerdal

Supervisor: Fredrik Winquist

ifm, Linköpings universitet

Anthony Turner

ifm, Linköpings universitet

Examiner: Fredrik Winquist

ifm, Linköpings universitet

Avdelning, Institution

Division, Department

Physics, Chemistry, and Biology

Department of Physics and Measurement Technology Linköpings universitet

SE-581 83 Linköping, Sweden

Datum Date 2012-05-02 Språk Language  Svenska/Swedish  Engelska/English   Rapporttyp Report category  Licentiatavhandling  Examensarbete  C-uppsats  D-uppsats  Övrig rapport  

URL för elektronisk version

http://www.ifm.liu.se http://www.ep.liu.se ISBN — ISRN LITH-IFM-A-EX--12/2577--SE

Serietitel och serienummer

Title of series, numbering

ISSN

—

Titel

Title

Elektrokemiska mätningar av alfa-amylas-aktivitet i saliv Electrochemical Measurements of Salivary Alpha-Amylase Activity

Författare

Author

Henrik Höckerdal

Sammanfattning

Abstract

Stress constitutes a more and more common cause for many health disorders in modern society. Salivary α-amylase (AA), the most abundant enzyme in human whole saliva, has in recent years been found to be a good surrogate biomarker for monitoring stress levels in individuals. This work aims to form the foundation of a novel approach for measuring the activity of the enzyme in saliva samples by means of electrochemistry. The idea is to implement several enzymes along with a starch substrate and an electron mediator in a single system. This system is then to be coated onto a screen-printed electrode (SPE), which is used along with an electrical component, designed to give rise to a quantifiable, electrical signal when the starch is broken down by the AA contained in an added saliva sample. Several such enzyme systems are here qualitatively evaluated. As electron medi-ator, ferro-/ferricyanide is used. Two different enzymes, glucose oxidase (GOx) and pyrroloquinoline quinone dependent glucose dehydrogenase (PQQ-GDH), are tested for making up the saccharide oxidising part of the system. Both prove them-selves capable in terms of qualitatively giving rise to an electrical signal. But, in terms of internal quantitative comparisons between the two, no practical experi-ments are performed in this work. In some runs, the enzyme α-glucosidase (AG) is used as an intermediate for breaking down the AA/starch oligosaccharide prod-ucts into monosaccharides. This to increase the system’s electrical signal output when using GOx as oxidising agent. Regrettably, due to lack of AG enzyme, these runs do not provide any conclusive data, and so further investigations of systems including this enzyme are needed. Otherwise, all systems tested seem to work, and neither of them appear better than the others. Therefore, all of them will require further quantitative testing to determine which one is best to implement in the final design of the enzyme system to be applied onto the SPE.

Nyckelord

Abstract

Stress constitutes a more and more common cause for many health disorders in modern society. Salivary α-amylase (AA), the most abundant enzyme in human whole saliva, has in recent years been found to be a good surrogate biomarker for monitoring stress levels in individuals. This work aims to form the foundation of a novel approach for measuring the activity of the enzyme in saliva samples by means of electrochemistry. The idea is to implement several enzymes along with a starch substrate and an electron mediator in a single system. This system is then to be coated onto a screen-printed electrode (SPE), which is used along with an electrical component, designed to give rise to a quantifiable, electrical signal when the starch is broken down by the AA contained in an added saliva sample. Several such enzyme systems are here qualitatively evaluated. As electron mediator, ferro-/ferricyanide is used. Two different enzymes, glucose oxidase (GOx) and pyrroloquinoline quinone dependent glucose dehydrogenase (PQQ-GDH), are tested for making up the saccharide oxidising part of the system. Both prove themselves capable in terms of qualitatively giving rise to an electrical signal. But, in terms of internal quantitative comparisons between the two, no practical experiments are performed in this work. In some runs, the enzyme α-glucosidase (AG) is used as an intermediate for breaking down the AA/starch oligosaccharide products into monosaccharides. This to increase the system’s electrical signal output when using GOx as oxidising agent. Regrettably, due to lack of AG enzyme, these runs do not provide any conclusive data, and so further investigations of systems including this enzyme are needed. Otherwise, all systems tested seem to work, and neither of them appear better than the others. Therefore, all of them will require further quantitative testing to determine which one is best to implement in the final design of the enzyme system to be applied onto the SPE.

Sammanfattning

Stress utgör ett allt vanligare och allvarligare hälsoproblem i dagens moderna samhälle. På senare år har α-amylas, som är det vanligaste enzymet i helsaliv, visat sig korrelera med ett flertal stressfaktorer, vilket har gjort det till en intressant biomarkör för att övervaka stress hos enskilda individer. Detta arbete syftar till att utgöra grunden för en ny metod för att mäta amylasaktiviteten i salivprov, baserad på elektrokemi. Grundtanken är att designa ett enzymsystem uppbyggt av ett stärkelsesubstrat, enzymer och en elektronmediator och belägga detta på ytan av en s.k. printad elektrod. Elektroden kommer användas tillsammans med en elektronikdel. När ett salivprov innehållandes amylas sätts till elektrodytan, kommer

vi

systemet ge upphov till en mätbar elektrisk signal. Denna signal är proportionell mot amylasets aktivitet i saliven. Kvalitativa utvärderingar av ett flertal kandiderande enzymsystem att belägga på elektrodens yta utförs i detta arbete. Glukosoxidas och pyrrolokinolinkinonberoende glukosdehydrogenas är två enzym som prövas för att utgöra steget i systemet där man övergår från kemi till elektronik. Detta är möjligt då dessa båda är oxidoreduktas, vilket innebär plockar upp elektroner samtidigt som de oxiderar sitt substrat. De överflödiga elektronerna kan sedan överföras från enzymet till en elektrod via en mediator, som i detta arbete utgörs av ferro-/ferricyanid. På detta sätt uppstår den elektriska signalen som mäts. Båda dessa enzym visar sig vara kapabla till att fungera i ett sådant system, men för att avgöra vilket som passar bäst att ha i den slutgiltiga designen krävs kvantitativa jämförelser dem emellan. I en försöksserie inkluderas enzymet α-glukosidas som en intermediär mellan amylasreaktionen och glukosoxidas för att bryta ner de oligosackarider, som utgör huvudprodukten i amylas- och stärkelsereaktionen, till monosackarider. Detta för att ge mer substrat åt oxidoreduktaset och därigenom öka känsligheten hos systemet. På grund av brist på glukosoxidasenzym och bristfälligt experimentellt utförande så kan tyvärr inga slutsatser dras av dessa försök gällande glukosidasets betydelse, utan kommer kräva ytterligare undersökningar. I övrigt verkar alla prövade enzymsystem fungera rent kvalitativt, vilket innebär att inget av dem kan förkastas som icke värt att gå vidare med i framtida, kvantitativa, försök.

Acknowledgments

During my work with this thesis, I have had a lot of help and support from my supervisors, Prof. Winquist and Prof. Turner. Other staff members at IFM have also helped me with ideas and practical tips in the lab.

I also thank my student reviewer, Sofia Tjernström, for reading and reviewing my report and my presentation, and also for encouraging me while I was writing the report and preparing my presentation.

Lastly, I owe great thanks to my family, for their support and encouragement to finish the report. Especially, I thank my brother Erik for all the help given when writing the report and preparing the presentation.

Thank you!

Contents

2.2.1 The Enzyme Reaction . . . 6

2.2.2 Enzymatic Activity . . . 7

2.2.3 Measuring the Enzymatic Activity . . . 8

2.2.4 Michaelis-Menten Kinetics . . . 9

2.2.5 Cofactors & Inhibitors . . . 11

2.2.6 Enzymes, Buffers and Mediators Used in This Project . . . 12

3 Instruments & Material 17 3.1 Screen-Printed Electrodes (SPEs) . . . 17

3.2 Enzymes . . . 18

3.3 Electrical Equipment & Other Chemicals . . . 18

4 Experimental 21 4.1 Beaker . . . 22

4.2 Bare SPEs . . . 22

4.3 Glucose Test Strips . . . 22

4.4 Saliva Sample . . . 24

4.4.1 Preparation Before Saliva Sample Collection . . . 24

4.4.2 Saliva Sample Collection & Handling . . . 25

4.5 Performing the Experiments . . . 25

4.5.1 Glucose Standards . . . 25

4.5.2 GOx Experiments . . . 27

4.5.3 α-Amylase Experiments Including α-Glucosidase . . . . 28

4.5.4 α-Amylase Experiments Excluding α-Glucosidase . . . . 29

4.5.5 PQQ-GDH Experiments . . . 29

4.5.6 Saliva Experiments . . . 31

x Contents

5 Results & Discussion 33

5.1 Glucose Standards . . . 33 5.2 GOx Experiments . . . 34 5.3 α-Amylase Experiments . . . . 35 5.3.1 Including α-Glucosidase . . . . 35 5.3.2 Excluding α-Glucosidase . . . . 36 5.4 PQQ-GDH Experiments . . . 37 5.5 Saliva Experiments . . . 38 6 Conclusions 41 7 Recommendations for Future Work 43 Bibliography 45 A Preparation of Buffers, Enzymes & Substrates 49 A.1 KPO4 Buffer . . . 49

A.2 PIPES Buffer . . . 49

A.3 PIPES Buffer Containing Ferricyanide Mediator . . . 50

A.4 Starch Solutions . . . 50

A.5 PQQ-GDH Holoenzyme . . . 51

A.6 α-Amylase, α-Glucosidase & Glucose Oxidase Stock Solutions . . . 51

A.7 Preparation of Experimental Component & Test Solutions . . . 52

A.7.1 Glucose Standards Experiments . . . 52

A.7.2 GOx Experiments . . . 53

Abbreviations

AA α-amylase

AG α-glucosidase

a.u. Arbitrary Unit

FAD Flavine Adenine Dinucleotide

FDA United States Food and Drug Administration GOx Glucose Oxidase

HPLC High Performance Liquid Chromatography PQQ-GDH Pyrrolo quinoline quinone glucose dehydrogenase PO2 Oxygen Partial Pressure

PIPES Piperazine-N,N’-bis(2-ethanesulfonic acid) SPE Screen Printed Electrode

Chapter 1

Introduction

1.1

Background

To ascertain the cause of pathological symptoms in individuals has long been one of the most basic purposes of the modern health care. In the last decades, the waiting time for a doctor’s appointment has become a more and more substantial problem for many patients, and a medical disorder is often allowed to go too far before it gets properly managed for a simple treatment to be enough for the patient to get fully recovered. This, in turn, leads to unnecessary discomfort for the patient as well as more stress for the medical staff.

One approach for achieving quicker diagnostics is to make it possible for com-mon people to com-monitor their general health at home. This has lead to an increasing encouragement for the development of so-called ubiquitous biosensors. These portable devices can be brought along anywhere and used by anyone, anytime. The main benefit of ubiquitous sensors is that they allow for a fast and simple point-of-care diagnosis or method for monitoring of medical issues. The pregnancy test for qualitative diagnosis, and the glucometer used by diabetics for quantita-tive monitoring, are two examples of such devices that are already commercially available.

The ubiquitous biosensors are quite specialised and only measure the presence or concentration of one single type of analyte. This in turn means that their respective field of application is very narrow. This changes, though, if the analyte in question correlates with a wider variety of health problems, such as stress.

Today, one of the most common underlying causes for different medical discom-forts is stress, and several grave health disorders, e.g. heart problems and even some types of cancer, have been shown to relate to it [1, 2]. The stress itself can arise from a number of different reasons, both physical and psychological.

One major problem when trying to quantify stress hormones directly is that the procedure in itself is often stressful, since it is either invasive or quite circumstan-tial. This makes the interpretation of the results from such samples exceedingly difficult. Therefore, researchers from different fields, related to both psychology and physiology, have been searching for alternative, non-invasive ways for monitoring

4 Introduction

stress levels.

In recent years, (salivary) α-Amylase (AA), which is an enzyme abundant in whole saliva, has been suggested as a convenient biomarker for activation of the sympathetic nervous system, by correlating with the hormonal response in both psychological and physical stress [3].

1.2

Project Aim

This project aims to build the foundation of a novel ubiquitous biosensor for mea-suring the activity of α-amylase (AA) in human whole saliva by an electrochemical approach. For this purpose, several systems including AA along with other enzymes and an electron mediator are presented and evaluated. These evaluations are only qualitative: No quantitative qualities of the different systems are considered, only whether they work or not.

1.3

Method

Several approaches for determining the salivary AA activity in saliva are tested and qualitatively evaluated. These include glucose test strips pre-coated with GOx and an electron mediator, and bare SPE’s.

Chapter 2

Theory

2.1

Saliva

Human whole saliva is mainly composed of fluid secreted by the salivary glands and gingival crevices. It is made up of 99 % water, and is otherwise a mixture with quite complex composition. In addition to electrolytes and solvents (such as transient food and drink residues), it contains a myriad of enzymes, immunoglobulins, desquamated epithelial cells and bacteria native to the mouth. The concentrations of buffering agents, e.g. phosphate, calcium and bicarbonate, as well as protein concentrations in the saliva varies considerably, even within a single individual. The pH is normally somewhere between 6.0–7.4. These variations do not seem to follow any pattern, circadian rhythm or whether the saliva is stimulated or unstimulated. Therefore, it is difficult to compose a representative buffer when conducting in vitro experiments on saliva [4, 5, 6].

There are several types of salivary glands, all located in different places in the oral cavity. Each type differs from the others in their secreted saliva’s flow rate and composition. What they have in common, though, is that they are all governed by the sympathetic and parasympathetic nervous systems, which thereby controls the overall flow rate of saliva. These nerves respond to different stimuli, including taste, smell and hormones. When we are eating, the salivary glands increase the flow rate of saliva. Otherwise, there is always a slow, continuous flow. This “resting” saliva is used for moistening the mouth and lubricating the mucous membranes [7]. The overall composition of whole saliva has been shown to reflect the systemic concentrations of hormones, enzymes, electrolytes etc. [8]. This has made it a very interesting subject for research when it comes to developing new ubiquitous diagnostic tools for different purposes. The field has already shown promising results for aiding in the diagnosis of connective tissue disorders [9], cystic fibrosis [10] and stress [3, 11, 12].

This work focuses on the enzyme α-amylase and how it correlates to stress. Stimulated saliva has been shown to contain a higher concentration of AA than unstimulated saliva. Both physical and psychological stress factors seem to induce this change [13]. Several tools for monitoring this enzyme’s concentration in saliva

6 Theory

have already been developed [8, 14, 15]. All of these systems accomplish this by different approaches. In this work, several ideas for a novel approach are presented.

2.2

Introduction to Enzymes

Enzymes are a class of proteins with quite complex structure, that induce chemical changes in the body. Generally, they catalyse chemical reactions by lowering their activation energy, see figure 2.1.

In the process, they do not undergo any permanent chemical changes themselves, i.e. they are not used up but can start a reaction anew as soon as the one before is finished [16].

∆Espontaneous

∆Eenzyme

Time

Energy

Figure 2.1: Activation energy of one spontaneous and one enzyme catalysed reaction. In the latter, the activation energy is lowered due to the catalysing effects of the enzyme.

2.2.1

The Enzyme Reaction

Since enzymes affect a reaction by lowering its activation energy, they do not alter the spontaneous equilibrium between the substrates and products [17]. Rather, they simply accelerate the reaction to reach this equilibrium much faster.

2.2 Introduction to Enzymes 7

A simple example of how an enzyme-catalysed reaction progresses over time is illustrated in figure 2.2.

The reaction can be divided into three phases:

Phase 1 The rate of product formation accumulates over time. This phase actually

lasts less than a second, and is highly exaggerated in figure 2.2 [18].

Phase 2 Later, the amount of product increases almost linearly with time. The

rate at which reaction product is formed in this phase is often called the initial rate. This can be somewhat confusing, since phase 1 then seems to be ignored. Phase 2 is also called steady state, because the concentration of enzyme-substrate complex does not change, or changes very little, thus allowing for the somewhat linear increase in product concentration [18].

Phase 3 Finally, the substrate becomes depleted and the concentration of reaction

product levels out.

1 2 3 Time Amoun t of reaction pro duct

Figure 2.2: The three phases of an enzyme reaction. First, the rate at which product is formed accumulates (1) until it reaches the steady state, where it becomes linear (2). When the substrate finally gets depleted (or an equilibrium is reached), the

amount of product reaches a plateau (3).

2.2.2

Enzymatic Activity

The activity of an enzyme is measured in International Units (I.U or U). 1 U is defined as the amount of enzyme that catalyses the conversion of 1 µmole of substrate per minute. An enzyme’s specific activity measures the activity per mg of enzyme (U/mg) [19].

Except for the enzyme and substrate concentrations, there are many factors that affect how well an enzyme is performing. Temperature, pH, the presence of substrates like inhibitors/competing substrates or activators/cofactors, all affect the work capacity of the enzyme.

8 Theory

When determining the activity of an enzyme, the initial rate is evaluated for different substrate concentrations, but for the same enzymatic activity. When enough such data has been collected, the initial substrate concentration for each case is plotted against respective initial velocity.

When performing this type of assay, there are a few conditions that have to be fulfilled for it to be accurate [19]:

• The substrate needs to be present in excess, to ensure that the enzyme is working at maximum capacity. This means that all enzyme is present as enzyme-substrate complex (the so-called steady-state approximation, see section 2.2.4).

• No activators or inhibitors may be present.

• The pH should be held at the optimal level for the enzyme. • The temperature should be held at 25◦_C.

2.2.3

Measuring the Enzymatic Activity

When measuring the activity of an enzyme in practice, either the amount product formed or the substrate consumed, is quantified. There are many ways to perform these measurements, e.g. photometric, radiometric, electrochemical and HPLC [20]. Here, amperometrically controlled electrochemical measurements are carried out.

Amperometry

The most prominent advantages of amperometric techniques are their accuracy, pre-cision and, in particular, their sensitivity. Most of today’s conventional glucometers use amperometry for determining blood glucose levels [21].

All amperometric (as well as voltammetric) techniques involve the application of an electric potential to an electrode, while measuring the resulting current that flows through the electrochemical cell. The Faradaic current that arises when the potential is applied gives a quantitative measure of how fast a redox reaction is running at the electrode’s surface.

Every redox reaction has a more or less unique redox potential that is required to drive it in either direction. Thus, by adjusting the magnitude of the applied potential, it is to some extent possible to steer what particular reaction one wants to measure. The magnitude of the current is, besides the reaction speed, affected by other factors, such as the concentration of the redox species, the solution’s resistance, and the characteristics of the working electrode.

The flux of the analyte from the bulk of the solution to and from the electrode surface also has an impact on the current. There are three modes of this mass transport: diffusion, migration and convection. Diffusion is the spontaneous, random movement of a solute. Migration occurs when a charged particle is affected by an electric field, and convection is forced movement by e.g. stirring or applied thermal currents [22].

2.2 Introduction to Enzymes 9

Instrumentation For Amperometrical Experiments

The basic components required for an amperometric measurement setup are a potentiostat, a computer, and the electrochemical cell where the experiment is carried out, and the electrodes [22].

The potentiostat Responsible for applying, maintaining and monitoring the

potential on the electrochemical cell. It also acts as a signal transducer between the cell and the computer. Since the Faradaic current is quite small, in the range of µA–mA, the potentiostat needs to have quite low current capabilities.

The electrodes and cell The cell consists of a sample holder (beaker) containing

the analyte dissolved in a buffer, and a set of electrodes. The beaker can be of different sizes and materials depending on the current experiment. In a typical setup, there are three electrodes: a reference, a working, and a counter, or auxiliary, electrode. The current is being fed through the working electrode, where the redox reaction of interest takes place, and the counter electrode. The purpose of the counter electrode is to pass the current needed to balance the current at the working electrode.

The reference electrode’s role is to act as a reference in measuring and controlling the potential of the working electrode. Any current flowing through the reference is undesired, since it would eventually make its potential unreliable [22]. Generally, the reference and working electrodes should be placed as close together as possible in order to minimise the impact of the solution’s resistance on the signal.

A three-electrode setup is appropriate when the Faradic current is in the

µA–mA area. When using micron-sized electrodes, however, the current is

smaller, in the range pA–nA. In these situations, only two electrodes are used. Here, the reference electrode is also acting as a counter electrode. A wiring diagram of a two- and three electrode setup including a potentiostat is shown in figure 2.3.

The computer Containing software for storing and displaying the collected data.

2.2.4

Michaelis-Menten Kinetics

In the beginning of the 20th _{century, Leonor Michaelis and Maud Menten described}

enzyme processes according to equation (2.1). In their model, they postulated that the enzyme and substrate first bind reversibly to form an enzyme-substrate complex. Then, the reaction itself, where the substrate is transformed into product, occurs, and finally the enzyme and newly formed product dissociates.

S + E −−−*)−−−k1

k−1

ES−−−*)−−−k2

k−2

E + P (2.1)

E = enzyme, S = substrate, ES = enzyme-substrate complex, and P =

10 Theory

Counter

Reference Working V

A

(a) Three-electrode setup

Counter

Reference Working

A

(b) Two-electrode setup

Figure 2.3: Two- and three-electrode setup examples.

Regarding the concentration of the substances in the model, the following apply:

Total amount of enzyme, [Etot] = [E] + [ES]

ES formed = k1· [E][S]

= k1· ([Etot] − [ES]) · [S]

ES dissociated = (k−1+ k2) · [ES]

(2.2)

Note that eq (2.2) assumes that no enzyme-substrate complex is formed by any backward reactions, i.e. k−2= 0. If the substrate is present in excess, then

the enzyme should ideally be present only as enzyme-substrate complexes: As soon as an enzyme has completed transforming a substrate into a product, it immediately picks up another substrate to transform. Thus, the enzyme is working at maximum capacity. This also means that the concentration of ES is constant, and the assumption is therefore called the steady-state approximation. Assuming this is true, the rate at which enzyme-substrate complex is formed should be the same as the rate at which it dissociates:

2.2 Introduction to Enzymes 11

Divide by ([ES] · k1) and solve for [ES], yields the following expression:

[ES] = [S][Etot]

k−1+ k2

k1

+ [S]

Regarding the initial velocity (V0) and maximum velocity (the absolute maximum

velocity for very high substrate concentrations, Vmax), the following statements

apply [19]:

V0= [ES] · k2 Vmax= [Etot] · k2 (2.3)

Substituting the expressions in (2.3) and introducing the Michaelis-Menten con-stant,

KM=

k−1+ k2

k1

yields the Michaelis-Menten equation:

V0=

Vmax· [S]

[S] + KM

(2.4)

KMis equivalent to the substrate concentration where half the maximum reaction

velocity is reached, and indicates the affinity of the substrate for the enzyme. A lower value indicates a higher affinity. Typically, KMvaries between 10−2–10−9 M

[19].

In short, the following statements apply for the KM:

• A small KM indicates that the enzyme requires only a small amount of

substrate to become saturated. Hence, the maximum velocity is reached at relatively low substrate concentrations.

• A large KM indicates the need for high substrate concentrations to achieve

maximum reaction velocity.

• The substrate with the lowest KM upon which the enzyme acts as a catalyst

is frequently assumed to be the enzyme’s natural substrate, though this is not true for all enzymes.

The Michaelis-Menten model is valid only for simple situations, where no allostericity or cooperativity occurs. Depending on how the enzyme is affected by inhibitors, cooperativity, and intermediate states, the model can be extended to better fit the current situation.

2.2.5

Cofactors & Inhibitors

Many enzymes are dependent on the presence of another substrate, or cofactor, to function. Such enzymes are, in their inactive state, called apoenzymes. When bound together, the apoenzyme and cofactor form an active holoenzyme. There are three classes of cofactors:

12 Theory

Coenzymes are small, organic non-protein substrates that bind relatively loosely

to the apoenzyme.

Prosthetic Groups are also small organic non-protein compounds, but they bind

more tightly (covalently) to the apoenzyme.

Metal-Ion-Activators are simply metallic ions that bind firmly to the

apoen-zyme.

Besides cofactors, which increase the efficiency of the enzyme they bind to, there are some chemical species with the opposite effect. These are called enzyme inhibitors. Simply put, when these substances bind to an enzyme, they inhibit its catalytic effect.

The effects an inhibitor has on an enzyme can be either reversible or irreversible. A reversible inhibiting substance does not bind covalently to the enzyme, or induce any lasting conformation changes, which would result in permanent loss of activity. Instead, they only bind non-covalently, meaning that if the enzyme was separated from the inhibitor, it would regain its former activity.

A reversible inhibitor can fall into one of three major categories: competitive, non-competitive, or uncompetitive [23].

Competitive inhibitors are called so because it competes with the enzyme’s

substrate for the same binding site, the active site. Thus, increasing the substrate concentration can to some extent overcome the effect of the inhibitor.

Noncompetitive inhibitors bind to the enzyme at a site separated from the

active site. Upon binding, these induce a change in conformation of the active site, rendering the enzyme unable to bind any substrate. Since this is not a competitive inhibition, it cannot be overcome by increasing the concentration of substrate.

Uncompetitive inhibitors bind exclusively to the enzyme-substrate complex, thus

affecting the rate at which the substrate or product dissociates from the enzyme.

2.2.6

Enzymes, Buffers and Mediators Used in This Project

In this section, the enzymes, buffers, and electron mediator involved in the experi-ments performed in this work are presented.

α-Amylase (AA)

α-amylase is a product of the pancreas and of the parotid and submandibular

glands (two types of major salivary glands) in the mouth, and is the most abun-dant enzyme present in human saliva. It hydrolyses the α-1:4-glycosidic linkages along polysaccharide chains of starch. It can thus act upon both of the major components present in ordinary starch, amylose and amylopectin. Both of these are polysaccharides made up of glucose units. The structural difference between

2.2 Introduction to Enzymes 13

them is that amylose is linear, while the amylopectin is branched, see figure 2.4. Also, amylopectin is more soluble in water than amylose. The splitting of the chain can occur anywhere along the molecule, except where branching occurs, see figure 2.4. Naturally, the enzyme also cleaves the glycosidic bonds of the products resulting from the early cleavages, thus degrading the starch to finally yield mainly maltose and maltotriose. To some extent, glucose is also formed, but the hydrolysis of terminal dextrose units is much slower [7]. The optimal pH for α-amylase is around 7.0 [24].

There are many known inhibitors that can affect the activity α-amylase, which come in contact with the enzyme by ingestion or inhaling. Examples of substances that contain such inhibitors are cigarette smoke [25], proso [26], and certain types of tea [27].

Even though no AA inhibitors are present in any of the experiments carried out in this particular work, it is important to understand that for an amylase activity measuring method to be valid, appropriate standard protocols for sample collection must be followed. O O OH HO O O OH HO O O OH HO · · · O O OH HO O O OH HO · ·· O O OH HO O O OH HO O O OH HO HO HO HO HO HO HO HO HO O α1,6-bond α1,4-bond

Figure 2.4: An amylopectin polysaccharide branching site, where the AA is unable to cleave the polysaccharide. Cleaving of terminal glucose units (red) seldom occurs.

α-Glucosidase (AG)

This enzyme breaks the α-1,4-linkages between the glucose units in maltose to yield glucose [28]. In vivo, its purpose is mainly to help break down glycogen, which is our equivalent to the plants’ amylopectin, to glucose for energy. The optimal pH of α-glucosidase varies depending on origin. Generally, it is acidic and in the range of pH 5.0–6.5 [29, 30].

In this work, the AG is interesting since it could help increase the amount of glucose formed in the AA/starch reaction by cleaving the reaction’s

oligosac-14 Theory

charide products and yield monosaccharides (glucose). Supposedly, the increased concentration of glucose would induce an amplified electrical signal output in the experiments, in particular when using GOx and not PQQ-GDH, see section 4.5.3.

Glucose Oxidase (GOx)

GOx is a flavo-protein containing the prosthetic group flavine adenine dinucleotide (FAD). It catalyses the oxidation of β-D-glucose to form δ-gluconolactone and

hydrogen peroxide:

β-glucose + GOx-FAD −−→ GOx-FADH2+ δ-gluconolactone

GOx-FADH2+ O2−−→ GOx-FAD + H2O2

The optimal pH of GOx is around 5.5, with a broad range from pH 4–7 [31]. It has a particularly high specificity as well as high affinity for the β-glucose molecule. This means that the enzyme relatively fast can oxidise this particular substrate without cross-reacting with other, similar chemical species (sugars), even when it is present only in low concentrations in a complex fluid. These attributes have made GOx a well-established component for use in biosensor devices for measuring blood glucose levels.

One drawback with GOx is that its activity depends on the partial pressure of molecular oxygen (PO2). Oxygen is its natural electron acceptor, and if another

mediator (section 2.2.6) is present, it only acts as a competitor. Thus, unless the reaction is carried out in an anaerobic environment, there is always a risk that the readout of a possible glucometer device is affected by the PO2 [21, 32]. Furthermore,

the H₂O₂ formed when metabolising molecular oxygen can, to some extent, affect the activity of the enzyme [33]. In conventional blood glucometers, there is very often, if not always, a mediator involved.

Pyrrolo Quinoline Quinone Glucose Dehydrogenase (PQQ-GDH)

Just as GOx, GDH catalyses the oxidation of glucose to produce gluconolactone. The active site is located deep within its 3-dimensional structure along with the prosthetic group, PQQ. This makes the electron transfer from PQQ directly to an electrode remarkably difficult. The addition of a mediator can facilitate the transport of electrons. Several such mediators and approaches for immobilising them and the GDH onto an electrode surface have already been evaluated [34]. The optimal pH of the PQQ-GDH is around 7.0 [35].

In its catalytic activity, PQQ-GDH requires no oxygen. Therefore, no hydrogen peroxide is formed that could disturb the enzyme’s performance, as in the case of GOx [34]. Due to this oxygen insensitivity, PQQ-GDH was discussed as an optional candidate for GOx to employ in blood glucometers in the beginning of the 21th _{century [36]. Lately, however, PQQ-GDH has been shown not to present the}

same level of specificity towards glucose as GOx. Rather, it tends to cross-react with other, similar substrates, such as the disaccharides maltose and lactose [37]. These non-glucose sugars can be found in, or be the result of the metabolism, of

2.2 Introduction to Enzymes 15

certain therapeutic products or drugs that sometimes are administered to diabetics [38]. Therefore, when used to measure the blood glucose levels, there is a chance of observing falsely elevated values when using an instrument that utilises PQQ-GDH. In the U.S. alone, several diabetic patients have died after receiving too aggressive insulin doses following such false readings. As of 2009, the FDA has urged all medical personnel to avoid using any PQQ-GDH based sensors for measuring blood glucose [38].

In this particular work, however, the PQQ-GDH’s lack of specificityis actually an advantage, since a system employing this enzyme instead of GOx may not require any AG to convert the oligosaccharide products from the AA/starch enzyme reaction into monosaccharides. Rather, the PQQ-GDH is capable of acting upon these oligosaccharides directly, as well as any glucose formed.

Electron Mediator

A mediator is a substance that acts as an electron transporter between the enzyme and the electrode in an electrochemical setup. Consider, for example, the PQQ-GDH: When reduced by oxidising a glucose molecule, the enzyme needs to somehow re-oxidise to be able to oxidise the next glucose molecule. In an amperometric experiment, this happens at the working electrode’s surface. However, since the active site and PQQ, which together are responsible for the uptake of electrons from the substrate, reside so deep within the 3D-structure of the enzyme, it requires a large applied potential on the electrode to extract the electron directly. If, however, a mediator is present, the reduced PQQ-GDH can quite easily surrender the excess electrons to this instead. The reduced mediator then reoxidises at the electrode surface [34], see figure 2.5. Since the travel route of the electrons now is divided into smaller steps, the size of the applied potential on the electrode can be reduced. Not only does this save battery power in portable devices, but it also serves to reduce the risk of other possible redox reactions that require larger potentials to interfere with the measurement.

Electro de Enzyme Substrate Product MOX MRED 2e− → → 2e− 2e− →

Figure 2.5: When oxidising its substrate to form the reaction product, the enzyme takes up the excess electrons. The mediator, M, then acts as a shuttle that helps re-oxidising the enzyme by transporting these electrons to the electrode surface, where they induce an electric current that is measured.

16 Theory

Potassium ferricyanide is the most common mediator used in commercial glucometers [21]. Its redox potential is around +436 mV [39], as compared to the hydrogen peroxide formed by un-mediated GOx (+600 mV) [40].

Buffers

In enzymatic experiments, the importance of using an appropriate buffer solution is critical. It needs to be non-toxic to cells, able to buffer at the appropriate pH and otherwise interfere as little as possible with the enzymatic activity. Preferably, it should also be stable and easy to handle and store. Potassium phosphate and PIPES are two examples of buffering agents that both fulfill these criteria.

Phosphate buffer is a multi-purpose buffer that is frequently used in biological

experiments. It is cheap, easy to prepare, and stable. It can be stored for several weeks in a refrigerator. It is highly soluble in water and buffers in a wide pH range (5.8–8.0) [41]. One inconvenience with phosphate is that it tends to sequester divalent cations. In this particular work, this could constitute a problem when used in assays involving PQQ-GDH, since the PQQ-GDH holoenzyme is dependent on Ca2+-ions to maintain its structure and function [42]. But, since the cation’s binding site is situated so deep within the enzyme’s 3D structure, many PQQ-GDH assays performed today still use phosphate buffer.

PIPES buffer , or piperazine-N,N’-bis(2-ethanesulfonic acid) is one of the twelve

buffering agents developed by Good et al. in the 1960’s [43, 44]. Good focused on developing buffering agents specifically designed for use in biology and biochemistry. Since the aim of Good’s research was to design buffering agents specifically for use in biological applications, any of these would have sufficed in this work. The PIPES buffer range reaches between pH 6.1–7-5.

A list of the characteristic attributes of a Good buffer is provided in [45], and details on the PIPES buffer are presented in [43].

Chapter 3

Instruments & Material

3.1

Screen-Printed Electrodes (SPEs)

Two batches of pre-coated conventional blood glucose test strips are used in this work: Bayer Breeze 2, and ACON International On-Call Plus. Both are pre-coated with GOx enzyme and a ferro/-ferricyanide electron mediator.

The On-Call Plus strips are used along with an ACON International On-Call EZ glucometer. Since no matching glucometer was available for the Bayer Breeze 2 strips, these are coupled to the potentiostat presented below. The Bayer Breeze 2 strips are of a two-electrode type, while the On-Call Plus strips use three electrodes. Both are presented further in section 4.3.

Besides the pre-coated SPEs, two types of bare SPEs, designated “DS 110” and “DS C110”, are used. Both are three-electrode SPEs, with carbon working- and counter electrodes, and an Ag reference. They were supplied by Dropsens. No further references are given, but there is no difference in their respective function or how they are designed. Figure 3.1 shows each of the bare SPEs.

(a) DS 110 SPE (b) DS C110 SPE

Figure 3.1: The DS 110 and DS C110 SPEs.

18 Instruments & Material

In regard of experimental setup, the SPEs are almost equivalent to using a two- or three-electrode system as shown in figure 2.3, depending on how many electrodes the current SPE has. A schematic of an SPE is shown in figure 3.2.

Working electrode Reference electrode Counter electrode W.E connection R.E connection C.E connection Ceramic substrate Hydrophobic layer

Figure 3.2: A bare three-electrode SPE. All SPEs, both pre-coated and bare, have the same general design and components. The main differences are whether they are coated or not, and the number of electrodes they are designed with.

3.2

Enzymes

Two batches of AA were used. One was provided by Sigma-Aldrich, product number A7595, henceforth designated AA1, and the other by Sigma Chemical Company, US, designated AA2. AA1 came pre-dissolved in a fluid, while AA2 was in the form of a dry powder. The specific activity of AA1 was, according to the manufacturer, 250 U/ml. The specific activity of AA2 was 25 U/mg.

The AG enzyme was provided by Sigma Chemical Company, US. According to the manufacturer, the specific activity of this enzyme was 131 U/mg.

Two batches of GOx were used. One was provided by Sigma-Aldrich, henceforth designated GOx1, with a specific activity around 100–250 U/mg. The other was provided by Sekisui Diagnostics Ltd, UK, and had a specific activity of 271.7 U/mg. This batch is designated GOx2.

Note that the enzyme activity unit, U, is not the same for any of the different enzyme batches. To avoid confusing the enzymes’ respective levels of activity with each other, or any international standards, their respective activity level is presented in each enzyme’s individual arbitrary unit (a.u.) in the following chapters.

The PQQ and GDH used in this work were provided by Sekisui Diagnostics Ltd, UK.

3.3

Electrical Equipment & Other Chemicals

In all amperometric experiments not involving a commercial glucometer, an Iskra Potenciostat MA 5410 and a laptop with Q-Basic software for data acquisition and storage, were used. All experiments conducted in a beaker were carried out in a

3.3 Electrical Equipment & Other Chemicals 19

20 ml glass beaker. In these experiments, either a two-electrode or a three-electrode setup was employed. In the case of two electrodes, an Ag/AgCl-electrode was used as reference/counter, while a small piece of Pt-thread (about 2 mm in length, and 1 mm in diameter) was used as working electrode. In the three-electrode system, a spatula made of stainless steel acted as working electrode. All this equipment was available at the IFM lab.

The potassium ferricyanide mediator was provided by Sigma-Aldrich, as was the glucose and mono- and dibasic potassium phosphate used to make the phosphate buffer. Potato starch and plastic straws were bought from the local grocery store.

Chapter 4

Experimental

As previously mentioned, this work focuses mainly on the qualitative evaluation of some ideas on how to quantify the amount of AA in saliva samples by an electrochemical approach. In all experiments carried out, the concept of how to detect the enzyme remains the same, see (4.1). As stated in section 2.2, AA degrades starch into oligosaccharides and glucose. GOx or PQQ-GDH then transform the glucose into gluconolactone while being reduced. The enzyme then yields the excess electrons via a mediator to the electrode and thus give rise to a current that is measured. If AA, GOx or PQQ-GDH, an appropriate mediator, and starch are mixed together in an amperometric setup in a single system, the following occurs:

starch (polysaccharide)−−−−−−→ Oligosaccharides + Glucoseα-amylase Oligosaccharides−−−−−−−−→ Glucoseα-glucosidase

Glucose GOx (or PQQ-GDH)+M

+

−−−−−−−−−−−−−−−−−→ Gluconolactone + M−

M− Electrode−−−−−−→ M++ Signal

(4.1)

where M represents the electron mediator, e.g. molecular oxygen in the case of un-mediated use of GOx and ferricyanide in the case of PQQ-GDH.

The enzyme system in (4.1) is representative for more or less all experiments throughout this work. The main difference between the approaches lies in which oxidoreductase (GOx or PQQ-GDH) that is used to oxidise the glucose formed by the amylase/starch reaction, and whether AG, i.e. the second row in the equation, is included in the system or not.

For each experiment, the final mixture consisting of a buffer, enzymes, cofactors, and substrates upon which the actual measurements are conducted is called “test solution”.

Details regarding the preparation of all buffers, enzymes, substrates and test solutions are presented in the appendices.

22 Experimental

4.1

Beaker

All experiments carried out in a beaker involved convection by a stir bar. Generally, such migration should be eliminated or operated accurately to ensure a controlled migration [22]. But again, since this work focuses on qualitative measurements, being able to see a clear difference between a blank and an actual run is all that is taken into account.

Between each run, the beaker, stir bar, and all involved electrodes were cleaned. The Pt working electrode is also cleaned electrochemically according to the in-formation found in [46]. This to minimize the risk of having one experiment contaminating or otherwise cause disturbances in following experiments.

4.2

Bare SPEs

When used in the experiments, the bare SPEs were connected to the potentiostat via cables, and the test solution was simply applied onto the electrode surface using a pipette, see figure 4.1.

(a) DS C110 SPE connected to the potentio-stat.

(b) DS C110 SPE connected to the potentio-stat, with a sample of test fluid applied.

Figure 4.1: Photographic representation of the bare SPE setup. The Bayer Breeze 2 SPEs were connected to the potentiostat in the exact same way. The only difference is that the Bayer Breeze 2 only uses two electrodes, see figure 4.2.

4.3

Glucose Test Strips

The commercial glucometer test strips are constructed in much the same way as the bare SPEs presented in section 3.1, only these are pre-coated with GOx and a mediator (ferricyanide). To ensure accuracy in glucose measurements, they are designed to work on a very specific volume of blood, generally in the range of 0.5–3 µl. Today’s commercial strips are capable of measuring blood glucose levels in the range between 0.5–35 mM. Since the amount of mediator and glucose oxidising enzyme coated onto the strips are specified for a certain volume of blood, as well

4.3 Glucose Test Strips 23

as a limited range of glucose level, there is always a limit of how large a signal they are able to produce when used here.

When used for blood glucose measurements, the precise blood volume absorption is achieved by letting the strip absorb the sample by its end through capillary forces. Again, this volume is in the µl range. This is on the borderline to insufficient for the experimental setup employed here to yield reliable results, since there is no guarantee that a sufficient amount of each component in the test solution (i.e. mediator, enzymes and substrate) is present on the test strip when the experiment is carried out. Still, the On-Call Plus strips were not modified in any way in regard of the volume they absorb. Instead, they were simply dipped in a beaker containing the test solution and then immediately inserted into the On-Call EZ glucometer for a read-out.

Also, when using the On-Call EZ glucometer, controls with the reference fluid, see figure 4.3, were performed regularly according to the manual.

In contrast, the Bayer Breeze 2 strips were peeled before the test solution was applied, see figure 4.2. The plastic layer covering the electrodes was removed, allowing for a drop of test solution to be applied onto the electrode surface using a pipette, much in the same way as for the bare SPEs. Also, no commercial glucometer suited for these SPEs was supplied, so when used in the experiments, they were connected to the potentiostat much in the same manner as the bare SPEs, see figure 4.1. The only thing that differs between the Bayer Breeze 2 and bare SPE setups is that the Bayer Breeze strips only had two electrodes instead of three.

(a) Un-peeled test strip. (b) Peeled test strip

24 Experimental

Figure 4.3: On-Call EZ glucometer and an On-Call Plus test strip, next to a bottle of reference fluid.

4.4

Saliva Sample

In the late stages of this work, experiments were made on a sample of human saliva. There are an abundance of protocols for proper whole saliva sample collection and handling. Here, the “passive drool” method was employed following the instructions given in [47].

The collection procedure and sample handling are presented in detail in sections 4.4.1 and 4.4.2.

4.4.1

Preparation Before Saliva Sample Collection

The only saliva sample in this work was collected from a male subject, with no oral health problems.

The sample collection was carried out at 11 a.m. the day before the experiments requiring the saliva were performed. The subject was instructed not to have breakfast on the day the sample was to be collected. Also, he was to brush his teeth about 1 hour before, as well as rinse his mouth with water 15 minutes prior to the sample collection.

4.5 Performing the Experiments 25

4.4.2

Saliva Sample Collection & Handling

As stated previously, the saliva sample was collected by the so-called “passive drool” method, since this is the simplest and most versatile collection method presented in [47].

Before collecting the saliva sample, a plastic straw was cut in two pieces. The length of the piece that was used was about 5 cm. One end of it was placed in the subject’s mouth, and the saliva was simply allowed to run through the straw and was collected in an eppendorf tube. No salivary stimulants were used.

When enough saliva (∼2 ml) was collected, the sample was immediately placed in the freezer at -20◦C and stored over night.

The next day, the sample was thawed, vortexed and centrifuged for 15 minutes at 3000 RPM. By doing this, the mucins precipitate, making the saliva’s supernatant easier to pipette [47].

4.5

Performing the Experiments

In this section, all of the different experimental series are presented regarding their respective aim and performance. Preparation details on all solutions used are found in the appendices, and all results are presented and evaluated in chapter 5.

4.5.1

Glucose Standards

The aim of this series of experiments was to determine whether it was possible to use SPEs pre-coated with GOx and a mediator to detect and perhaps quantify glucose in a fluid other than blood.

Two glucose standard solutions were prepared and used in this work. Each was used together with one type of pre-coated glucose test strip: Standard 1 was used on the On-Call Plus, and standard 2 was used on the Bayer Breeze 2 test strips.

The reason for using two standards is that the test strips may be sensitive to the test solution’s ion strength, a fact that was overlooked at first. The ion strength of standard 1 is somewhat lower than standard 2, which has about the same ion strength as saline. Saline contains 0.9 % NaCl, which corresponds to a salt concentration of about 150 mM.

For both standards, the tested glucose concentrations were 20, 15, 10, and 5 mM. In the Bayer Breeze 2 runs, two test strips were used for each concentration, i.e. a total of eight strips for each experiment. Since some of the carbon electrode material was removed when peeling the test strips, the second run for each concentration was needed for validation. Also, these strips were not used along with a glucometer, but were instead coupled to the potentiostat.

When running the glucose standards on the On-Call Plus, only one test strip was used for each glucose concentration. The need for further controls was deemed unnecessary, since the glucometer was regularly controlled using the reference fluid provided, see section 4.3.

26 Experimental

Procedure

The two 20 mM glucose standards were prepared separately, each in a 20 ml glass beaker. For preparation details, see section A.7.1. A stir bar was used continuously throughout the experiments.

Specific volumes of the dilution solutions 1 and 2 were added to respective standard to lower the glucose concentrations in steps of 5 mM down to 15, 10, and finally 5 mM. For preparation details on these, see section A.7.1. For each concentration, a sample of the standards were applied onto the test strips.

The volumes of dilution solution added between each glucose concentration step for standard 1 are found in table 4.1.

Table 4.1: Added volumes of glucose standard dilution solution 1

Volume added (ml) Glucose concentration (mM)

0 20

1.1 15

2.0 10

6.1 5

Note: Since the volume needed by the test strip was so small (1 µl), this was

ignored when calculating the appropriate volume of dilution solution to be added. Volumes of dilution solution added between each glucose concentration step for standard 2 are found in table 4.2:

Table 4.2: Added volumes of glucose standard dilution solution 2

Volume added (µl) Glucose concentration (mM)

0 20

1035 15

1917 10

5652 5

Note: Here, 50 µl solution was applied to each test strip, meaning that a total

of 100 µl of fluid was removed on each level of glucose concentration. This was accounted for when calculating the amount of dilution solution to be added.

The practical procedures of the glucose standard experiments only differ in how the solutions were applied onto the SPEs and how the signal was read:

• For the On-Call Plus test strips, and the standard solution 1, the sample was applied by dipping the end of the strip in the beaker. The strip was then immediately inserted into the On-Call EZ glucometer for a read-out. • In the case of standard solution 2 and the Bayer Breeze 2 test strips, a

50 µl sample was pipetted onto the electrode surface while it was connected to the potentiostat. A potential of +440 mV was applied on the working

4.5 Performing the Experiments 27

electrode. The Q-Basic software started measuring the current 30 seconds prior to applying the sample to the test strip. Each of these runs lasted for a total of 15 minutes.

4.5.2

GOx Experiments

Here, the purpose was to see how well “free” GOx, i.e. GOx that is not coated onto an SPE, performed without the help of an electron mediator.

Both batches of GOx presented in section 3.2 were used in this series of experiments, as well as setups employing both two and three electrodes.

The “fundamental” solution presented in table A.11 functioned as a buffer that was present in the test beaker to which the acting components (i.e. the GOx and glucose) were added.

The glucose and GOx were prepared in separate containers prior to the experi-ments. The GOx stock solutions were prepared as presented in section A.6, and their respective concentrations are found in table 4.3.

Table 4.3: GOx stock soluion concentrations

Enzyme Final concentration (a.u./ml)

GOx1 35

GOx2 54

The glucose stock solution was prepared as presented in table A.12 in the appendices.

Procedure

Throughout the GOx experiments, the applied potential of the potentiostat during the measurements was +600 mV on the working electrode vs. the reference. Also, a stir bar was used continuously.

In all experiments, the starting volume in the test beaker was always 10 ml: 9 ml of fundamental solution and 1 ml of GOx stock solution.

In each run, the GOx was present in the test beaker from the start, while the concentration of glucose was increased in steps at certain time intervals by adding volumes of the glucose solution to the test beaker found in table 4.4.

The first addition of glucose occurred 30 seconds after the measurement was started. Then, the additions occurred every 5 minutes. Thus, each run lasted for a total of 1 hour and 30 seconds.

In addition to the experiments including both GOx and glucose, two blanks were made. These experiments were identical to the ordinary ones in every aspect, except no GOx was added to the “GOx Solution” in the first run, and no glucose was added in the “Glucose Solution” in the second.

28 Experimental

Table 4.4: Volumes of glucose stock solution added in the GOx experiments

Volume added (µl) Glucose concentration (mM)

0 0 50 0.1 51 0.2 51 0.3 52 0.4 52 0.5 53 0.6 217 1.0 585 2.0

4.5.3

α-Amylase Experiments Including α-Glucosidase

The initial purpose of these experiments and the ones described in section 4.5.4, was to determine how large impact the addition of AG in the enzyme system would have on the signal output of an enzyme system employing GOx, since this supposedly increases the amount of glucose formed from the AA/starch enzyme reaction by cleaving its maltose and maltotriose products into glucose. Sadly, the amount of AG enzyme available for this project was only enough for performing a first few test runs, so the experiments in section 4.5.4 were re-designed to determine whether it would be possible to detect any glucose originating from only an AA/starch reaction.

The AA experiments that involve the AG enzyme were performed on the On-Call Plus pre-coated SPEs. Thus, since they were both already present on the strips’ electrode surface, no additional GOx or mediator was included into either test solution.

The stock solutions used were composed of one batch of AA, one of AG, and a 10 w/v% starch solution. All of these solutions are presented in detail in sections A.4, and A.6. The concentrations of the enzymatic stock solutions are displayed in table 4.5.

Table 4.5: The concentration of enzyme in their respective stock solution

Enzyme batch Enzyme Concentration (a.u./ml)

AA1 25

AA2 1000

AG 100

Procedure

Before performing the actual experiments, the AA- and starch solution used was prepared in a separate beaker and stirred for 45 minutes using a stir bar. This to

4.5 Performing the Experiments 29

give the AA plenty of time to break down the starch before the AG was added. These AA/starch solutions were prepared according to table A.13 in the appendices. After 45 minutes had passed, the AA/starch solution was blended with the saturated KCl and AG stock solution, thus making up the final test solution. For details, see table A.14. Because of its small volume, the test solution was prepared in a well in a 96 well plate.

The solution was then given an additional 2 minutes before an On-Call Plus test strip was dipped in the well and inserted into the glucometer. This to give the AG some extra time to break down the maltose from the AA/starch reaction.

Apart from the actual runs, blanks were performed where the AG stock solution was replaced by 20 µl of deionised water.

4.5.4

α-Amylase Experiments Excluding α-Glucosidase

After completing the experiments in section 4.5.3, there was no AG enzyme left. Therefore, these experiments were performed to determine whether the glucose formed by an AA/starch enzyme reaction alone, would be somewhat quantifiable by GOx.

The test solution was composed from two separate components, a 2 w/v% starch stock solution, and a 500 a.u./ml AA2 stock solution. For details on these, see sections A.4 and A.6, respectively.

Procedure

When running the experiments, the two solution components were added separately onto a Bayer Breeze 2 test strip that was connected to the potentiostat. Both components were added in equal volumes, thus yielding final concentrations equal to half the concentrations in the respective stock solutions, i.e. 1 w/v% starch and 250 a.u./ml AA2. Two final volumes of test solution were evaluated, 20 µl and 50 µl.

The first component was applied 30 seconds after that the computer started measuring the current, and the second was added 1 minute after the first. This to distinguish any false signal that could have arisen from either of the two test solution components. Totally, each run lasted for 15 minutes. During the whole experiment, the applied potential on the working electrode was +440 mV.

In some runs, the dissolved AA was added first, followed by the starch solution, while the opposite applied in others.

Blanks were also performed, where the AA solution was replaced by an equal volume of deionised water. Otherwise, the blanks were performed just as the actual runs, both where the water was added first, and where the starch was added first.

4.5.5

PQQ-GDH Experiments

Two fundamental differences from the AA experiments apply here. First, GOx was replaced by PQQ-GDH as the oxidoreductase agent. Second, the pre-coated test strips were replaced by the bare SPEs. Runs were made with both the DS110 and DS C110 SPEs.

30 Experimental

The test solution used in these experiments was composed of two components, 1 and 2.

Component 1 was a mix of equal volumes of a 0.2 M PIPES and 0.1 M potassium

ferricyanide stock solution, 3 w/v% starch stock solution, and 0.6 mM PQQ-GDH stock solution (sections A.3, A.4, and A.5, respectively). The substances’ respective concentrations in this mixture component are displayed in table 4.6.

Table 4.6: Concentrations of the PQQ-GDH experiments’ mixture component 1

Substance Final concentration

PQQ-GDH 0.2 mM

PIPES 67 mM

Potassium ferricyanide 33 mM

Starch 1 w/v%

As for the ferricyanide mediator, it was required to be included in the test solution, since no mediator was pre-coated onto the SPE.

Component 2 simply constituted the addition of the AA enzyme. AA1 was the

only batch used in these experiments.

The concentration of this component differed between runs, to ascertain a semi-quantitative measurement. Either a volume of the batch was added directly (250 a.u./ml), or it was pre-diluted in deionised water in ratios 1:4 (50 a.u./ml), or 1:8 (∼28 a.u./ml).

The final test solution applied onto the Bayer Breeze 2 strips was composed of 36 µl of component 1, and 4 µl of component 2. The final concentrations of all included substances of the test solution are shown in table 4.7.

Table 4.7: Concentrations in the PQQ-GDH experiments’ test solution

Substance Final concentration

PQQ-GDH 0.18 mM PIPES 60 mM Potassium ferricyanide 30 mM Starch 0.9 w/v % AA1 25 a.u./ml 5 a.u./ml ∼2.8 a.u./ml

Blanks were performed where component 2 was replaced by deionised water, leaving the test solution free from any AA enzyme.

4.5 Performing the Experiments 31

Procedure

When starting the measurement of the current, the electrode was dry. 30 seconds after the measurement of the current was initiated, component 1 was applied onto the SPE. After another 30 seconds, component 2 was added. Each run lasted for a total of 15 minutes.

During the measurements, the potentiostat was set to +500 mV on the working electrode, as done in [36].

Since the supply of bare SPEs was limited, each that was employed here was used several times over. Between each run, they were rinsed thoroughly with water and dried with N₂.

4.5.6

Saliva Experiments

The final experiments of this work aimed to analyse human saliva. The saliva sample collection and handling is presented in detail in section 4.4.2.

In these experiments, only the DS 110 SPEs were used. As with the PQQ-GDH experiments, ferricyanide mediator was applied onto the electrode surface along with the enzymes, and the potential applied on the working electrode was +500 mV during all measurements.

The test solution for these experiments was composed of the two components presented below.

Component 1 was the same as component 1 in the PQQ-GDH experiments. For

details, see “Component 1” in section 4.5.5.

Component 2 was made up of the saliva sample. In some runs, this was

pre-diluted in a 1:1 ratio in 0.05 M PIPES buffer, see section A.2, thus halving the concentration of saliva.

The volume ratios of the two components applied onto the SPE were 1:1, with 25 µl of each. The concentration of each component in the test solution is displayed in table 4.8.

Table 4.8: Substance concentrations in the saliva experiments’ test solution. The far right column shows the altered concentrations of pure saliva and PIPES in the runs where the saliva sample was pre-diluted in the buffer.

Substance Final concentration (saliva pre-diluted in PIPES)

PQQ-GDH 0.1 mM —

PIPES 33 mM 46 mM

Potassium ferricyanide 17 mM —

Starch 0.5 w/v % —

32 Experimental

Procedure

30 seconds after the measurement of the current was initiated, component 1 was dropped onto the SPE. After another 30 seconds, component 2 was added.

Each run lasted for a total of 15 minutes.

The applied potential on the working electrode was always +500 mV when conducting these experiments.

For these experiments, there were only two DS 110 SPEs available. Therefore, each was used several times. Between each run, the SPE was rinsed thoroughly with deionised water and dried with N₂.

Chapter 5

Results & Discussion

In the following chapter, the results from the experimental procedures described in section 4.5 are presented. When studying the figures displaying graphs, only the steady currents are interesting. The narrow peaks that appear between 0 – 100 seconds in each graph are the result of adding a drop of fluid to the electrode surface, see correspondant “procedure” for each experiment in section 4.5. Thereby, they are not relevant for the results. Also, the current in each graph is presented as a.u. This is because in each experimental series, different settings were applied on the potentiostat’s sensitivity level. This implies that the results presented in each graph is only internally comparable, unless stated otherwise.

5.1

Glucose Standards

This series of experiments was performed according to section 4.5.1.

Both the On-Call Plus and the Bayer Breeze 2 test strips appear fully capable of detecting glucose in the tested standard solutions. But, the results do not appear precise enough to achieve more than a semi-quantitative level of accuracy for either type of SPE, see table 5.1, and the graph in figure 5.1.

The underlying reason for this could be that the SPEs are built for measuring glucose in blood and not in the buffer solution used here. Glucometer test strips are designed for measuring on a very specific volume in a solution (blood) with certain characteristics regarding pH, ion strength etc.

Still, being able to get a semi-quantifiable read-out of the glucose content is sufficient for this work.

34 Results & Discussion

Table 5.1: On-Call Plus test strip results for the Glucose standard 1 experimental series. The actual read-outs are not quite in accordance with the theoretical glucose concentrations. The results appear at least semi-quantifiable, though. “HI” means that the glucose concentration was above the maximum value that could be measured by the glucometer, i.e. >33.3 mM.

Theoretical Glucose concentration (mM) Test strip Readout (mM)

20 HI 15 29.3 10 20.7 5 8.3 0 100 200 300 400 500 600 700 800 900 0 0.5 1 Time – [s] Curren t – [a.u.]

Glucose standards results from Bayer Breeze 2 test strips

0 mM (blank) 5 mM

20 mM

Figure 5.1: The Bayer Breeze 2 glucose standards results show a barely semi-quantifiable result. The difference in measured current between the 5 mM and 20 mM runs is not large enough for the runs to be clearly distinguishable from each other.

5.2

GOx Experiments

These experiments were performed according to section 4.5.2.

Like the glucose standards experiments, these runs relied on the GOx enzyme for oxidising the glucose, but this time it was dissolved along with its substrate, rather than coated onto an SPE. Also, no mediator was present, so the reaction was dependent on the atmospheric oxygen as mediator.

As stated in section 4.5.2, runs were performed using both two- and three-electrode setups. But, since the two-three-electrode setup gave more distinguishable results, only these results are displayed in this paper.