Molecular and Nanoscale Strategies for Enhanced Biosensing · 2018-11-19 · Molecular and Nanoscale Strategies for Enhanced Biosensing Wendi Zhou Doctor of Philosophy ... readout

Molecular and Nanoscale Strategies for Enhanced Biosensing

by

Wendi Zhou

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Department of Electrical and Computer Engineering University of Toronto

© Copyright by Wendi Zhou 2018

ii

Molecular and Nanoscale Strategies for Enhanced Biosensing

Wendi Zhou

Doctor of Philosophy

Department of Electrical and Computer Engineering

University of Toronto

2018

Abstract

As personalized medicine advances, the need continues to grow for new sensors in applications

ranging from clinical diagnostics to therapeutic development. Biosensors are powerful tools and

their evolution will play an important role in healthcare by enabling the detection of specific

analytes that are clinically significant.

The central aim of this thesis is to formulate new sensing strategies to detect clinically relevant

biomolecules in a sensitive and specific manner. Many target analytes are present at extremely

low concentrations while remaining relevant. The utility of a biosensor is limited by its

sensitivity; but also, as we investigate herein, its ability to maintain its function in a

heterogeneous environment. Here, we develop a novel sensing method for low concentration

analysis of soluble signaling factors by coupling DNA hybridization engineering and antibody-

capturing chemistry with an electrochemical-based reporting method. This technique – when

combined with a chip-based approach – allows for both sensitive and rapid detection.

Through the use of nanostructured electrodes, we showcase our technologies in applications that

include protein quantification and low-level light detection. We design a device that analyzes

multiple small signaling proteins in stem cell cultures with 10 pg·mL-1 sensitivity and validate

our results by comparing them to the gold standard enzyme-linked immunosorbent assay, with

iii

more suitable integration capability and in-line monitoring along with smaller sample sizes and

decreased process times. This platform is improved by further developing a reusable,

multiplexed one-step assay and sensor, simplifying and speeding up the steps required to obtain

readout. Finally, we report on nanogap-separated fractal electrodes as a method for the formation

of nanoscale features, exhibited by the fabrication of quantum dot photodetectors with nearly a

one hundredfold improvement in performance over that of conventional devices.

Overall, these strategies present important steps toward the development of more sensitive and

faster sensors. By demonstrating low concentration analysis in complex environments and

unprecedented performance with novel methods, we forge a pathway to more cost-effective and

powerful sensing.

iv

Acknowledgments

I would like to thank Professor Edward H. Sargent and Professor Shana O. Kelley for the

opportunity to work in their labs. Their support and guidance were invaluable over the course of

this work and they have continuously challenged me to think creatively, encouraged me to find

new solutions, and pushed me to be a better engineer.

I would like to extend a thank you to Professor Stewart Aitchison as my committee member for

his helpful suggestions.

This work would not be possible without all the members past and present of the Sargent and

Kelley Labs. I would like to particularly thank Dr. Valerio Adinolfi, Dr. Justin Besant, Dr. Alex

Ip, Dr. André Labelle, Dr. Brian Lam, Dr. Mahla Poudineh, Dr. Andrew Sage, and Dr. Brandon

Sutherland for introducing me to and educating me on many aspects of the lab. Thank you to

Barbara Alexander, Jeannie Ing, Damir Kopilovic, Elenita Palmiano, Dr. Mark Pereira, Dr. Ali

Seifitokaldani, Remigiusz Wolowiec, and Alexander Zaragoza for help in and outside of the lab.

I would also like to thank Dr. Jagotamoy Das, Dr. Mahmoud Labib, Dr. Reza Mohamadi, and

Dr. Sahar Mahshid for sharing their knowledge and expertise. I am grateful to Dr. Sharif Ahmed,

Dr. Ian Burgess, Bill Duong, Dr. Fengjia Fan, Surath Gomis, Dr. Brenda Green, Dr. Sae Rin

Jean, Andrew Johnston, Dr. Leyla Kermanshah, Dr. Laili Mahmoudian, Dr. Sara Mahshid, Mona

Mukhopadhyay, David Philpott, Dr. Tina Saberi Safaei, Dr. Ying Wan, Daniel Wang, Dr.

Guangli Wang, Dr. Simon Wisnovsky, Fan Xia, Xiaolong Yang, and Dr. Libing Zhang, and Dr.

Yige Zhou for their help and support. I would especially like to thank Peter Aldridge, Jenise

Chen, Dr. Eric Lei, Wenhan Liu, Adam Mepham, Carine Nemr, Tanja Sack, and Dr. Sarah Smith

for their helpful insights and advice, particularly in teaching me about biology and chemistry

and, most importantly, in making lab life fun.

I am thankful for my friends, with special appreciation for my NDRC and Iron Dragons families

to whom I owe a whole host of memorable experiences. A big thank you goes to Dr. Chris Wong

for his support throughout the years. Last but not least, I would most like to thank my parents,

Dr. Guanghan Wang and Dr. Bai Zhou, for their continued support in all things I do.

v

Table of Contents

Acknowledgments.......................................................................................................................... iv

Table of Contents .............................................................................................................................v

List of Figures ................................................................................................................................ ix

List of Abbreviations ................................................................................................................... xiii

Chapter 1 ..........................................................................................................................................1

1 Introduction .................................................................................................................................1

1.1 The Development of Advanced Biosensors .........................................................................1

1.2 Electrochemical Biosensors .................................................................................................2

1.3 Scope of Thesis ....................................................................................................................4

Chapter 2 ..........................................................................................................................................6

2 Background .................................................................................................................................6

2.1 Electrochemistry Principles and Techniques .......................................................................6

2.2 Protein Detection ...............................................................................................................11

2.3 Nanostructured Electrodes and Nanogaps .........................................................................16

2.3.1 Nanostructured Electrodes .....................................................................................16

2.3.2 Nanogaps................................................................................................................18

Chapter 3 ........................................................................................................................................21

3 Steric Hindrance Assay for Secreted Factors in Stem Cell Culture ..........................................21

3.1 Introduction ........................................................................................................................21

3.2 Background ........................................................................................................................22

3.3 Results and Discussion ......................................................................................................25

3.3.1 Assay and Sensor Chip ..........................................................................................25

3.3.2 DNA-Protein Conjugates and Chip Preparation ....................................................25

3.3.3 Sensor Characterization .........................................................................................27

vi

3.3.4 Determining Dynamic Range and Sensitivity........................................................27

3.3.5 Electrochemical Detection in Cell Culture Media Samples ..................................28

3.4 Methods..............................................................................................................................29

3.4.1 Materials ................................................................................................................29

3.4.2 Chip Fabrication.....................................................................................................30

3.4.3 Sensor Fabrication .................................................................................................30

3.4.4 Culture Media ........................................................................................................30

3.4.5 Conjugation of Protein with DNA .........................................................................30

3.4.6 Preparation of Capture Probe-Modified Chip ........................................................31

3.4.7 Detection in Buffer and Media ..............................................................................31

3.4.8 Detection in Stem Cell Culture Samples ...............................................................31

3.4.9 Electrochemical Measurements .............................................................................31

3.4.10 Binding Curves ......................................................................................................32

3.5 Conclusions ........................................................................................................................33

Chapter 4 ........................................................................................................................................35

4 A Reusable One-Step Electrochemical DNA-Based Sensor for the Quantitative Detection

of Soluble Signaling Proteins ....................................................................................................35

4.1 Introduction ........................................................................................................................35

4.2 Background ........................................................................................................................36


4.3.1 Assay and Sensor Chip ..........................................................................................37

4.3.2 Sensor Characterization and Sensitivity ................................................................39

4.3.3 Multiplexing ...........................................................................................................40

4.3.4 Sensor Regeneration ..............................................................................................41

4.3.5 On-Chip Resistive Heater ......................................................................................43

4.4 Methods..............................................................................................................................45

vii

4.4.1 Materials ................................................................................................................45

4.4.2 Chip Fabrication.....................................................................................................46

4.4.3 Sensor Fabrication .................................................................................................46

4.4.4 Culture Media ........................................................................................................46

4.4.5 Conjugation of Antibodies with DNA ...................................................................46

4.4.6 Preparation of Capture Complex ...........................................................................46

4.4.7 Detection in Media .................................................................................................47

4.4.8 Electrochemical Measurements .............................................................................47

4.4.9 Sensor Regeneration ..............................................................................................47

4.5 Conclusions ........................................................................................................................47

Chapter 5 ........................................................................................................................................49

5 Programmable Definition of Nanogap Electronic Devices Using Self-Inhibited Reagent

Depletion ...................................................................................................................................49

5.1 Introduction ........................................................................................................................49

5.2 Background ........................................................................................................................50


5.3.1 SIRD Method Overview ........................................................................................51

5.3.2 Solution-Processed Photoconductors .....................................................................53

5.4 Methods..............................................................................................................................56

5.4.1 SIRD Device Fabrication .......................................................................................56

5.4.2 SIRD Electrodeposition .........................................................................................56

5.4.3 Fabrication of CQD Photoconductor Devices .......................................................57

5.4.4 EQE Measurements ...............................................................................................57

5.4.5 Responsivity and Irradiance Measurements ..........................................................57

5.4.6 Noise Current and Detectivity Measurements .......................................................58

5.5 Conclusions ........................................................................................................................58

viii

Chapter 6 ........................................................................................................................................60

6 Conclusions and Future Work ...................................................................................................60

6.1 Summary ............................................................................................................................60

6.2 Future Work .......................................................................................................................61

References ......................................................................................................................................64

ix

List of Figures

Figure 1.1. Schematic showing the basic components of a biosensor. ........................................... 2

Figure 2.1. Schematic for a typical three-electrode electrochemical system consisting of the

working electrode (WE), reference electrode (RE), and counter electrode (CE). .......................... 7

Figure 2.2. Cyclic voltammetry. (A) Sample curve depicting linear potential ramp. (B) Sample

curve depicting current with redox reaction peaks. ........................................................................ 9

Figure 2.3. Differential pulse voltammetry. (A) Sample curve depicting potential ramp with a

superimposed pulse. (B) Sample curve depicting current versus the potential. ............................. 9

Figure 2.4. Sample curve depicting the potential step for chronoamperometry. .......................... 10

Figure 2.5. Square wave voltammetry. (A) Sample curve depicting potential staircase ramp with

superimposed square wave pulse. (B) Sample curve depicting current versus the potential. ...... 11

Figure 2.6. Schematic for tandem mass spectrometry. A sample is injected, ionized, and

accelerated. The ions are separated according to mass and charge through electromagnetic

deflection. They are analyzed first by mass analyzer 1 (MS1), selectively fragmented, and then

analyzed by mass analyzer 2 (MS2), which generates the spectra. .............................................. 12

Figure 2.7. Schematic for surface plasmon resonance. Incident light on the metal film is

reflected, collected, and analyzed. At a specific incident angle, the plasmons resonate with light,

thus resulting in light absorption at that angle resulting in a dark line in the reflected light beam.

The angular position of the dark line moves as a binding event or molecular conformational

change occurs. ............................................................................................................................... 13

Figure 2.8. Schematic demonstrating the change in efficiency at which the redox reporter bound

to the reporting protein reaches the electrode surface upon target binding[47]

. Reprinted (adapted)

with permission from [47]

. Copyright 2017 American Chemical Society. .................................... 14

Figure 2.9. Schematic illustrating the bio-barcode assay. (Upper) Barcode DNA-functionalized

gold nanoparticles are conjugated to antibodies. (Lower) The magnetic antibody-functionalized

x

microparticles and the antibody-DNA-gold nanoparticles form a sandwich around the target

antigen and the barcode DNA is detected using a scanometric assay[48]

. ..................................... 15

Figure 2.10. Scanning electron microscope image of a nanostructured microelectrode. ............. 17

Figure 2.11. Schematic illustrating the blocking assay, where protein is captured using antibody-

functionalized NMEs, preventing [Fe(CN)6]3-/4-

from reaching the electrode surface, reducing the

measured current[12]

. Reprinted (adapted) with permission from [12]

. Copyright 2011American

Chemical Society. ......................................................................................................................... 18

Figure 2.12. (A) Schematic and (B) Scanning electron micrograph of a nanogap sensor[62]

©

2016 IEEE. .................................................................................................................................... 19

Figure 3.1. Schematic showing the influence of secreted factors in stem cell culture and an

overview of the chip-based electrochemical detection scheme. (A) Simplified schematic of the

interactions between soluble factors and cell subpopulations. As the concentrations of signaling

proteins increase, mature cells (pink) accumulate as HSCs (blue) tend toward differentiation. As

the concentration of secreted factors is decreased through media dilution, their impact is reduced,

promoting the proliferation and self-renewal of HSCs. (B) Chip layout. Contacts are formed

from circular apertures in a layer of SU-8 covering a gold pattern on chip surface. (C) Schematic

representation of ASHHA on NMEs. Samples containing the target protein are preincubated with

the blocking antibody. The samples are then mixed with signaling DNA strands labeled with

both the electrochemical reporter and the recognition element before on-chip incubation. ......... 24

Figure 3.2. (A) Signal changes upon conjugation of the RANTES, MDC, and LAP recognition

complexes and their respective antibodies to the DNA signaling strands. (B) Electrochemical

signals obtained as a function of time and concentration for the RANTES assay. ...................... 26

Figure 3.3. Square wave voltammetry-derived currents detected at electrodes with varying

concentrations of RANTES in (A) buffer and (B) cell culture media. ......................................... 28

Figure 3.4. Electrochemical detection directly in cell culture media samples. (A) At-line protein

monitoring schematic. Measured concentrations obtained for specific detection of (B) RANTES,

(C) LAP, and (D) MDC in culture samples compared against ELISA measurements from the

xi

same samples. Currents obtained from electrochemical measurements for each factor were

normalized according to calibration curves and converted to concentrations. ............................. 29

Figure 3.5. Signaling reporter validation. Electrochemical signals obtained from the capture

strand-functionalized electrodes after 30 min incubation, (1) in the absence of DNA signaling

strands, and in the presence of DNA signaling strands tagged with (2) biotin, (3) the recognition

complex, and (4) the recognition complex bound to the blocking antibody. ............................... 32

Figure 3.6. Binding curves as measured by the sensor for (A) MDC, (B) LAP, and (C) RANTES

in buffer. (D) Representative binding curve as measured by a standard ELISA for RANTES in

buffer. ............................................................................................................................................ 33

Figure 4.1. Schematic illustrating chip-based detection scheme. (A) Chip layout. Gold contacts

are formed through the apertures in the SU-8 layer and are electroplated to create NMEs. Chip is

divided into individual sections for multiplexing, separated by lines drawn by a hydrophobic pen.

(B) Chip layout with individually addressed heaters surrounding each electrode. (C) Schematic

representation of the one-step proximity principle assay. Two thiolated DNA strands (P1 and P4)

are tethered to the surface while two complementary strands (P2 and P3, respectively) act as both

the capture complex and the signaling mechanism. Upon addition of protein, a decrease in

measured electrochemical current is seen. .................................................................................... 38


concentrations of RANTES in cell culture media......................................................................... 39

Figure 4.3. Electrochemical currents detected at the electrodes for (A) Sample positive for

RANTES, (B) Sample positive for MDC, and (C) Sample positive for both targets. .................. 41

Figure 4.4. Currents detected at electrodes before heating, after heating, and after DNA

rehybridization for the capture complex. ...................................................................................... 43

Figure 4.5. Relationship between on-chip heater temperature and applied bias. ......................... 44

Figure 4.6. Electrochemical currents detected at electrodes before and after sensor regeneration

over five cycles using the on-chip heater. ..................................................................................... 45

xii

Figure 5.1. Self-inhibiting reagent depletion electroplating platform. (A) Lithographic template

fabrication. Electrical leads with a 2 μm separation distance are passivated with a layer of SU-8

photoresist. (B) 10-μm apertures (denoted with an arrow) are imaged into the SU-8 passivation

layer above the 2 μm separation. (C) Imaging after SIRD electroplating electrodes in parallel

with 2 mM HAuCl4 + 50 mM HCl at 200 mV for 3 min. (D, E) SEM top-down imaging after

SIRD electroplating. (F) Focused ion beam cross-sectional imaging after SIRD electroplating.

(Scale bar, 1 μm). .......................................................................................................................... 52

Figure 5.2. Electroplating SIRD electrodes showing overlapping electrodes that remain

electrically isolated. (Scale bar = 1 µm). ...................................................................................... 53

Figure 5.3. SIRD solution-processed photoconductors. PbS colloidal quantum dots are dip coated

on surface of SIRD devices and infiltrate narrow region creating nanoscale photoconductor

junctions. (A) Image of SIRD photoconductors with corresponding SEM images. (B, C) SEM

images. (Scale bar, 1 μm).............................................................................................................. 54

Figure 5.4. SIRD photoconductor characterization. (A) Spectral external quantum efficiency. (B)

Responsivity with respect to applied bias. (C) Noise current with respect to applied bias. (D)

Detectivity with respect to applied bias. ....................................................................................... 56

xiii

List of Abbreviations

DNA: deoxyribonucleic acid

RNA: ribonucleic acid

WE: working electrode

RE: reference electrode

CE: counter electrode

ELISA: enzyme-linked immunosorbent assay

NME: nanostructure microelectrode

PNA: peptide nucleic acid

HSC: hematopoietic stem cell

ASHHA: amplified steric hindrance hybridization assay

RANTES: regulated on activation, normal T cell expressed and secreted

MDC: macrophage-derived chemokine

TGF-β1: transforming growth factor-β1

LAP latency-associated peptide

MCH: 6-mercaptohexanol

PBS: phosphate buffered saline

TPO: thrombopoietin

SIRD: self-inhibited reagent depletion

SEM: scanning electron microscopy

xiv

CQD: colloidal quantum dots

NEP: noise-equivalent power

EQE: external quantum efficiency

LED: light-emitting diode

1

Chapter 1

1 Introduction

1.1 The Development of Advanced Biosensors

Biosensors are used for a wide variety of applications, including medical diagnostics, therapeutic

development, and general healthcare monitoring. The first enzyme electrodes were designed by

Leland C. Clark and colleagues in 1962[1]

, and research has since enabled the development of

sophisticated and advanced devices to detect a variety of biological analytes. With applications

ranging from glucose monitors to food analysis to cancer diagnosis, biosensors form powerful

tools that enable the sensitive and rapid detection of important analytes. Recent advances in

biosensing technology have improved the sensitivity and speed of these devices, allowing for the

detection of a multitude of targets such as proteins, cells, and nucleic acids.

In its simplest form, a biosensor is an analytical device designed to identify or quantify a

biological molecule of interest and create a measurable signal output. As one might expect, its

name implies that a biosensor is a combination of a biological element and a sensor element. A

typical device consists specifically of a bio-recognition site, a biotransducer component, and a

system for readout. The bio-recognition site interacts with a biological sample such as blood, cell

cultures, or environmental samples to detect the target of interest using designed biological

receptor elements such as antibodies, proteins, or nucleic acids. The biotransduction component

converts this interaction into a measurable signal generally through optical, gravimetric, or

electrochemical means, which can be detected and quantified by the readout system [2]

. The basic

components of a biosensor are illustrated in Figure 1.1.

Biosensors have been designed using a variety of readout techniques. A common example of a

biosensor is a home pregnancy test, which is based on lateral flow and the detection of human

chorionic gonadotropin in urine[3]

. In this case, a visual readout provides an immediate diagnosis.

Optical sensors[4]

provide direct visual readout, but often have poor sensitivity. Other methods,

such as electronic[5]

, optical[4]

, and electrochemical[6]

, calorimetric biosensors[7]

, and

2

piezoelectric biosensors[8]

are more sensitive and more easily designed to detect multiple

analytes at once, but can be more expensive and require more complex equipment.

Figure 1.1. Schematic showing the basic components of a biosensor.

The most commercially successful biosensor is the glucose monitor, which measures the blood

glucose level through the electrochemical detection of an enzymatic reaction[9]

. Glucose oxidase

converts glucose to glucono delta-lactone and hydrogen peroxide, which is then measured by

amperometric sensing[9]

. These sensors are relatively inexpensive and can be used by patients

with minimal training, enabling regular at-home monitoring of blood glucose levels. In this

work, we place particular emphasis on electrochemical biosensors due to their potential for

excellent sensitivity, low cost, and ease of miniaturization.

1.2 Electrochemical Biosensors

Electrochemical biosensors have several benefits and are of special interest for use in

applications where high levels of sensitivity are required[10]

. Electrochemical sensors function

through the detection of oxidation or reduction reactions at or near the surface of an electrode. A

potentiostat is used to detect an electronic signal and can take the form of a benchtop or portable,

handheld device. In contrast, though optical sensors have a long history of use, instrumentation

needed for sensitive and quantitative detection, such as fluorescence microscopes, is often bulky

and expensive, limiting their in-line or at-line monitoring applications.

Electrochemical biosensors have been designed for the detection of many different biological

molecules, including proteins[11-13]

, DNA[14-15]

, and small molecules[16-18]

, with recognition

3

elements such as antibodies, nucleic acids, enzymes, and aptamers[19]

. In many cases, the

receptor component is designed to be selective for the target and sample preparation is often

minimal. In some instances, the specific binding of an analyte to a receptor or an enzymatic

reaction in the presence of a small molecule can directly be detected electrochemically.

However, there is often no direct electrochemical signal and a system is designed to convert a

biological recognition event into an electrochemical output. Various reporter methods include

using covalent[10]

and non-covalent reporters[20]

, in addition to enzymatic labels[21]

and

conformational changes to DNA structures[22]

.

Electrochemical biosensors are, by comparison, less extensively characterized. While

electrochemical sensors have great potential for high sensitivity due to their ability to measure

small changes in charge, they must continue to be developed to analyze analytes for more

exacting applications as detection limits are challenged further and further. Other disadvantages

of electrochemical biosensors include multiple and complex assay steps and the necessary

addition of several reagents. These issues need to be addressed for more convenient and

widespread use, such as improving reaction speed or time to obtain readout through the

development of one-step assays.

Novel materials and device and system architectures could provide avenues to overcome these

limitations. Nanomaterials and nanostructures in particular have found utility in improving the

performance of electrochemical biosensors and great strides have been made with regards to

sensitivity and speed from their contributions to the field[23]

. Electrode materials and structures

play an important role in sensing applications, and nanomaterials have been shown to act as

highly effective immobilization matrices[24]

. Efficient immobilization of molecules onto

electrode surfaces has been demonstrated with a variety of nanostructured materials with

different shapes, sizes, and compositions, such as graphite nanoparticles[25]

, palladium

nanoparticles[26]

, electrodes made from nanomaterials (NiFe2O4[27]

, Tm2O3[28]

, and Cu2O[29]

), and

nanostructured silica-phytic acid materials with diverse morphologies[30]

. New sensors need to be

developed, taking advantage of such electrode materials and structures. As a new generation of

nano-inspired biosensors emerges, robust characterization and investigation into how to best

exploit nanotechnology remains a challenge.

4

In addition to new electrode materials and fabrication techniques, new measurement assays for

specific analyte detection must be developed in order to fully embrace the advantages of

electrochemical sensors. Emerging sensors are challenged to breach the current limits of

detection and function in complex environments. Furthermore, lengthy assay times and multiple

reagent addition steps necessitate the development of more rapid and user-friendly procedures.

In this work, we endeavour to confront the shortcomings of currently available sensors and to

improve on existing methods with new solutions. We aim to tackle issues of sensitivity for

challenging molecules and increase assay speed while decreasing complexity of use. We display

enhanced efficacies through the development of novel assays and harness nanostructuring to

lower detection limits and increase efficiency. To exhibit our findings, we apply our sensors to

the detection of soluble signaling proteins in stem cell culture samples. Finally, we demonstrate

improved performance for a solution-processed quantum dot photodetector through the use of

nanoscale features.

1.3 Scope of Thesis

The objective of this thesis is to develop creative approaches to quantitative biosensing. We will

explore methods for overcoming limitations in sensitivity, speed, and specificity by developing

novel assays and combining them with a chip-based platform for electrochemical sensing. We

will also investigate new nanofabrication techniques and their impact on sensor performance.

Our methods are showcased with a variety of sensors such as photodetectors and with clinical

applications such as monitoring signaling proteins in stem cell culture.

In Chapter 2, we present an overview of state-of-art of biosensing as related to protein detection

as well as a summary of the electrochemical techniques used in this thesis. We also describe

nanostructured electrodes and fabrication techniques used to create nanoscale features.

Our goal in Chapter 3 was to explore methods into improving sensitivity for the quantification of

small proteins. As previous sensors were unable to achieve the desired sensitivity for targets of

this size in complex environments, we hypothesized that the addition of a large blocking

antibody could mitigate the challenge of small protein detection. We thus investigated steric

hindrance assays and techniques to amplify the effects to obtain low detection limits. We

describe here the results of a combination of DNA hybridization engineering with antibody-

5

capturing chemistry in an amplified steric hindrance hybridization assay on nanostructured

microelectrodes, representing the first universal nonsandwich technique that allows for pg·mL-1

quantification of small proteins in complex media without signal loss.

While we were able to achieve good levels of sensitivity, we then had a new goal of developing

simple and rapid sensing to reduce the number of steps, time to readout, and complexity of the

assay. We investigated one-step assays, particularly proximity-based assays, which have the

benefit having the potential for a built-in capture surface without the need for additional

reagents. We further probed methods for sensor regeneration to increase the utility of our device.

While chemical regeneration is the most common technique, we sought to implement a system

that did not require additional reagents and came upon the solution of heating, or DNA melting.

Thus, in Chapter 4, we demonstrate a one-step assay for rapid multiplexed protein detection on a

reusable sensor. Using a proximity-based assay along with a toehold latch and competitive

binding between an antibody-protein pair and DNA hybridization, quantification is achieved

quickly and with the use of an on-chip resistive heater, the sensor can be regenerated over

multiple cycles.

In Chapter 5, we investigate nanostructuring and nanoscale features as related to sensors. As

nanoscale features can enhance sensitivity and other sensor characteristics, we sought to explore

low-cost fabrication techniques that can achieve the same or better performance and

controllability compared to established methods. Low costs and simple, scalable techniques are

highly attractive for multiple applications. We describe here a novel technique for the

programmable definition of nanogaps separating fractal electrodes, showcased through quantum

dot photodetectors with low voltage responsivities a hundred times higher than previously

reported devices.

Finally, Chapter 6 provides a summary of this work and offers insight into potential future

directions for biosensing and improvements still to be made.

6

Chapter 2

2 Background

2.1 Electrochemistry Principles and Techniques

Electrochemistry is a branch of physical chemistry whose early studies included Luigi Galvani’s

first establishing a link between chemical reactions and electricity in 1791[31]

. Galvani performed

experiments on frogs and in one instance, accidentally touched an exposed sciatic nerve with a

charged metal scalpel, which resulted in sparks and the dead frog’s leg kicking. This observation

led Galvani to make the connection between electricity and life.

Following Galvani’s discoveries, Alessandro Volta conducted his own studies, which contributed

to the eventual invention of the voltaic pile, an early version of the battery capable of producing

a continuous electric current[32]

. This invention was the first in a series of other discoveries such

as electrolysis, and was one of the most important innovations leading to the development of the

field of electrochemistry. Today, the field has broadened and extended to many other

applications and devices, such as biosensors.

Electrochemistry is the study of typically heterogeneous chemical reactions in which electrons

are transferred between species, whereby one substance loses an electron and is oxidized, while a

second substance gains an electron and is reduced. Electrochemical systems generally consist of

three electrodes: a working electrode (WE), a reference electrode (RE), and a counter electrode

(CE), seen in Figure 2.1. The working electrode is where the electrochemical reaction takes place

and acts as the charge transfer interface. This electrode is often the site for biological binding

events. The reference electrode provides a stable reference point by which to measure the

potential of the working electrode. The most commonly used type of reference electrode is

silver/silver chloride, as it is inexpensive and non-toxic. The counter electrode, also known as the

auxiliary electrode, is used to measure the current through the working electrode. It completes

the circuit along with the working electrode, thus ensuring a stable potential for the reference

electrode through which current does not pass. The counter electrode generally has a larger area

than the working electrode so as not to limit the current passing through. Typical materials for

7

counter electrodes are inert conductors such as gold, platinum, or carbon. In some cases, a two-

electrode system is used, where a known current or potential is applied between the working

electrode and a second electrode that replaces both the reference and counter electrodes.

However, this method is less precise due to the ohmic drop in solution and is generally used with

energy storage or conversion devices such as batteries or photovoltaic panels. A two-electrode

system may also be used when the ohmic drop and current are small, such as in the case of

microelectrodes.

Figure 2.1. Schematic for a typical three-electrode electrochemical system consisting of the

working electrode (WE), reference electrode (RE), and counter electrode (CE).

A potentiostat is an electronic instrument used to control the electrochemical system and

measure signals between electrodes. It measures and controls the potential difference between

the working and reference electrodes using a control amplifier in addition to measuring the

current flow between the working and counter electrodes via the ohmic drop across a series

8

resistor. Various electrochemical techniques are classified according to the type of measurement

that is being made and include amperometric/voltammetric, impedimetric, conductometric and

capacitive, and field-effect transistor-based sensors.

Voltammetric and amperometric sensors measure a change in current due to an electrochemical

oxidation or reduction while applying a potential to the working electrode versus the reference

electrode. These sensors allow for good selectivity as the oxidation or reduction potential used is

usually specific to the target. We offer a brief overview here of a few of the most common

techniques, namely cyclic voltammetry, differential pulse voltammetry, chronoamperometry, and

square wave voltammetry.

Cyclic voltammetry is a frequently used method that measures current in an electrochemical cell

while the potential changes, typically sweeping from an initial voltage and increasing (or

decreasing) linearly with time until it reaches a defined endpoint before decreasing (or

increasing) at the same rate back to the initial voltage. A redox reaction will occur as the voltage

approaches the redox potential and the rate of this reaction will increase as the potential is

increased, which will in turn result in an increase in the current. Once the electroactive species is

consumed and the rate of the reaction exceeds that of the electroactive species diffusive

replenishment, the current will then decrease. In the reverse scan, the opposite reaction will

occur with a similar peak in the opposite direction. An example of sample curves depicting the

varying potential and resulting current with time are shown in Figure 2.2. Cyclic voltammetry is

a useful technique that offers information on analyte concentration, redox potential, and reaction

rates through analysis of the peak current and width. Information can also be provided regarding

the electrode surface area and electron transport efficiency as the surface adsorbents go through a

redox reaction. This information can be obtained by calculating the area under the current peak,

adjusting the scan rate, and observing the peak shape.

9

Figure 2.2. Cyclic voltammetry. (A) Sample curve depicting linear potential ramp. (B) Sample

curve depicting current with redox reaction peaks.

A second technique, differential pulse voltammetry, is similar to cyclic voltammetry, where the

potential is linearly increased (or decreased), with the addition of a small superimposed pulse.

The current is measured immediately before the pulse and immediately before the end of the

pulse, with the difference between them plotted against the varying potential. This technique

offers the advantage of low background currents and high sensitivity due to the background

current being removed by the subtraction of the two pre- and post-pulse current values, resulting

in a differential current being reported. Sample curves illustrating the voltage pulse and resulting

current peak are shown in Figure 2.3.

Figure 2.3. Differential pulse voltammetry. (A) Sample curve depicting potential ramp with a

superimposed pulse. (B) Sample curve depicting current versus the potential.

A B

A B

10

Chronoamperometry is an electrochemical technique where the potential is maintained at a

constant value and the current is measured as a function of time. The potential may be modeled

after a step function (seen in Figure 2.4), where it is initially at a constant value where no redox

reaction occurs, and is then suddenly stepped to a value at which a redox reaction may occur. For

short time periods after charging is complete, the current can be modeled by the Cottrell equation

(eq. 2.1),

𝑖(𝑡) = 𝑛𝐹𝐴𝐶√𝐷

√𝜋𝑡 (eq. 2.1)

where n is the number of electrons, F is Faraday's constant, A is the area of the (planar)

electrode, C is the concentration, D is the diffusion coefficient, and t is time. This technique is

useful for measuring the electroactive area of the electrode or the diffusion coefficient for the

electroactive species in question, using the Cottrell equation. It is also used for real-time

monitoring, especially once a system reaches steady-state such as for spherical or

microelectrodes, as a change in electroactive moiety would result in a subsequent change in

current. An additional application is in electroplating electrodes.

Figure 2.4. Sample curve depicting the potential step for chronoamperometry.

In this work, square wave voltammetry is the main technique used for electrochemical

measurements due to its high level of sensitivity and speed. It is a pulse method similar to

differential pulse voltammetry and consists of a linear potential sweep in the form of a square

wave pulse superimposed onto a staircase potential. The current is obtained by measuring the

difference between the forward and reverse potential pulses before the potential direction is

reversed. This sampling technique results in minimal capacitive current. This technique is a

11

preferred choice as very low detection limits may be obtained, making it an excellent option for

the detection of analytes with low concentrations. A sample waveform of the potential and

current for this method are shown in Figure 2.5.

Figure 2.5. Square wave voltammetry. (A) Sample curve depicting potential staircase ramp

with superimposed square wave pulse. (B) Sample curve depicting current versus the potential.

2.2 Protein Detection

Due to their low limits of detection, electrochemical biosensors have been used to detect a

multitude of analytes. In this thesis, we will highlight our work on developing novel sensors to

detect proteins, which are biologically relevant to many applications, including tumor markers,

clinical diagnostic testing, drug dosing, and as signaling factors. We focus on an application to

monitor soluble signaling proteins in stem cell cultures, which are heterogeneous and complex

environments. Though present in low quantities (around the pg·mL-1

level), these proteins play

an influential role in cell expansion and differentiation[33-35]

and thus control over these factors

through frequent or continuous monitoring would allow for improved cell expansion. Ideally, a

biosensor for protein detection would provide a rapid response and be highly sensitive and

specific for the protein of interest with the ability to perform integrated or automated monitoring,

making this a difficult challenge.

Currently, the gold standard method for protein quantification is the enzyme linked

immunosorbent assay (ELISA). In this method, proteins are immobilized and specifically labeled

via antibodies in sandwich-type reactions and are subsequently quantified through

spectrophotometry[36]

. Commercial ELISA kits are widely available and are sensitive, with low

A B

12

detection limits in the pg·mL-1 range. However, ELISAs usually require several hours for

operation due to multiple steps including many washes, as well as large sample volumes and

auxiliary equipment, and so are more suitable for off-line measurements and require trained

personnel.

There are currently a few techniques that are capable of directly performing protein analysis

without the need for additional reagents. Mass spectrometry combined with immunoassay

techniques, along with selected reaction monitoring, is a highly sensitive protein analysis

technique. In mass spectrometry, chemical species are ionized and these ions are separated

according to their mass to charge ratio and then quantified with a mechanism capable of

detecting ions. Selected reaction monitoring is used in tandem mass spectrometry, in which

multiple steps of mass spectrometry are performed, seen in Figure 2.6. An ion with a specific

mass is selected in the first stage of tandem mass spectrometry and in the second stage, an ion

product from the precursor fragmentation reaction is selected. Unfortunately, these techniques

require complex equipment and thus as with ELISAs, they have limited utility for in-line

measurements[36-39]

. Another direct technique, Raman spectroscopy, may also be used to identify

proteins via their spectral signatures using lasers. A sample is irradiated with a laser, resulting in

a small amount of Raman scattering, which is then detected as a Raman spectrum. However, this

technique has not yet been used for quantification[40]

.

Figure 2.6. Schematic for tandem mass spectrometry. A sample is injected, ionized, and

accelerated. The ions are separated according to mass and charge through electromagnetic

deflection. They are analyzed first by mass analyzer 1 (MS1), selectively fragmented, and then

analyzed by mass analyzer 2 (MS2), which generates the spectra.

13

Another method used for the multiplexed detection of many proteins is the usage of microarrays,

such as in surface plasmon resonance[41]

(Figure 2.7). In this technique, proteins are captured

onto a surface via a ligand, such as an antibody, which has been immobilized on to a substrate

coated in a metal film, known as the sensor chip. The presence of the proteins generates an

optical response, which is then measured by a detector. Though on-line measurements are able to

be made quickly, detection limits are around the ng·mL-1 level and are therefore unsuitable for

applications with lower protein concentrations.

Figure 2.7. Schematic for surface plasmon resonance. Incident light on the metal film is

reflected, collected, and analyzed. At a specific incident angle, the plasmons resonate with light,

thus resulting in light absorption at that angle resulting in a dark line in the reflected light beam.

The angular position of the dark line moves as a binding event or molecular conformational

change occurs.

Aptamers have been developed to bind tightly to specific proteins, and have been used to

generate an electrical or optical response when bound to the protein of interest. Aptamer binding

has faster reaction kinetics compared to antibody binding, which reduces the time needed for

traditional immunoassays[42]

. However, the widespread availability of aptamer-based protein

14

detection kits is limited due to the extensive and expensive development required to generate

aptamers for individual proteins.

Electrochemical sensors are an attractive option for direct protein detection due to their

sensitivity and potential for miniaturization[43]

. Many such sensors have already been

demonstrated with a variety of assays and detection platforms. Several studies have made use of

the immobilization of proteins on gold electrodes functionalized with target-specific antibodies

combined with electrochemical reporters, but demonstrated long assay times[44-46]

. Another group

described a reagentless one-step assay based on a redox-reporter-modified protein anchored to an

electrode[47]

. Upon specifically binding to the target antigen, the efficiency at which the reporter

reaches the electrode surface is altered, resulting in a change in the measured redox current

(Figure 2.8). However, the sensitivity of this assay only reached the ng·mL-1

level.

Figure 2.8. Schematic demonstrating the change in efficiency at which the redox reporter

bound to the reporting protein reaches the electrode surface upon target binding[47]

.

Reprinted (adapted) with permission from [47]

. Copyright 2017 American Chemical Society.

Nanomaterials and nanostructures have been garnering especial interest in recent years for

sensing applications, particularly for their benefits in improving sensitivity[12,23]

. One example is

a bio-barcode approach, where 330 fg·mL-1

detection of prostate specific antigen was achieved

using a specific DNA sequence as a reporter[48]

. In this assay, shown in Figure 2.9, magnetic

microparticle probes functionalized with antibodies are mixed with the target antigen, washed

free of excess serum, and combined with gold nanoparticles functionalized with antibody-

conjugated DNA barcode probes in a sandwich assay. After magnetic separation and wash steps,

15

the DNA barcodes are detected using the scanometric assay. While this technique is extremely

sensitive, it involves the use of multiple wash steps and significant lead time.

Figure 2.9. Schematic illustrating the bio-barcode assay. (Upper) Barcode DNA-

functionalized gold nanoparticles are conjugated to antibodies. (Lower) The magnetic antibody-

functionalized microparticles and the antibody-DNA-gold nanoparticles form a sandwich around

the target antigen and the barcode DNA is detected using a scanometric assay[48]

.

Silicon nanowire based field-effect transistor sensors have been described for the detection (92

pg·mL-1

) of cardiac troponin I using immobilized antibodies on the silicon nanowire surfaces, but

this method was only validated in buffered solutions[49]

. Another technique for electrochemical

sensing is electrode modification, such as with gold nanoparticle-coated glassy carbon electrodes

for the detection of prostate specific antigen in the range of 2 pg·mL-1

to 10 ng·mL-1

using a

sandwich assay and antibodies labeled with the nanocomposite of ferrocene monocarboxylic acid

16

hybridized graphene oxide[50]

. While sensitive, its multiple steps and wash cycles and thus time

required resemble those of an ELISA.

2.3 Nanostructured Electrodes and Nanogaps

2.3.1 Nanostructured Electrodes

For the detection of pg·mL-1

levels of analytes, ultrasensitive sensors must be developed. There

have been many advances, notably nano-inspired approaches using nanomaterials and

nanostructured surfaces. With a combination of micro- and nanomaterials, these sensing schemes

have made significant improvements in both sensitivity and speed[12]

. In this work, we focus on

nanostructured microelectrodes and nanogap-based sensors.

As electrodes shrink in size, the diffusional regime changes from linear to radial[51]

. A higher

flux of analytes is able to reach the electrode, which in turn allows for increased signal to noise

ratios and lower limits of detection[52]

. There is, however, a limit to the miniaturization of the

working electrode, as the interactions between the analyte and the electrode surface may not

occur within a reasonable timeframe with too small of a surface. This issue has been addressed

by the Kelley Laboratory, who worked to increase sensitivity by increasing the surface area of

their microelectrodes through nanostructuring[53]

.

Whereas classical electrodes are gold disk electrodes, the Kelley group developed non-planar

metal electrodes electroplated with metal alloys or pure noble metals such as gold or platinum[54]

.

Under different electrodeposition conditions, various sizes and morphologies could be obtained

to form nanostructured microelectrodes (NMEs)[55]

. With these microelectrodes, the metal

surface becomes textured at the nanoscale with the structures extended three-dimensionally

outward into the solution. Using chips pre-patterned with 5 µm gold apertures with the rest of the

electrode covered by an insulating layer, three-dimensional NMEs are fabricated through

electrodeposition using a metal salt, seen in Figure 2.10. Other methods such as thermal

wrinkling[56]

, in-situ formation of the nanostructured materials[57-58]

, and hierarchical growth[55]

have also been used to form nanostructures. This nanostructuring has the effect of increasing the

surface area of the working electrode and consequently the signal intensity and sensitivity, as

well as speed of detection by increasing the probability of interaction between an analyte and a

17

receptor at the electrode surface. The use of NMEs has resulted in detection of biomolecules at

concentrations as low as attomolar[54]

.

Figure 2.10. Scanning electron microscope image of a nanostructured microelectrode.

Once nanostructuring is complete, electrodes can be functionalized using thiolated probe

molecules. Due to the rough electrode surface, probes such as thiolated DNA capture molecules

may have an increased range of immobilization deflection angles compared to planar surfaces,

increasing their accessibility and hybridization efficiency[59]

.

Several chip-based platforms utilizing nanostructured electrodes have been reported that can

achieve multiplexed sensing of analytes with improved sensitivity compared to that of planar

surfaces[60]

. Combined with a simple electrochemical readout, these sensor chips can detect

DNA, RNA, protein, and small molecules qualitatively within 30 minutes with high specificity in

diverse solutions, from human serum to whole blood[12]

. These chips can be further integrated

into microfluidic devices for automated sample processing and detection.

A variety of assays have been developed using this platform and various electrochemical

reporters have been used. One such assay specifically detects cancer antigen 125, a large 200

kDa protein, with a detection limit of 100 pg·mL-1

using [Fe(CN)6]3-/4-

as the electrochemical

reporter (Figure 2.11)[12]

. A capture antibody was assembled on the surface of the electrode and

the presence of the target analyte results in the inhibition of the interfacial electron transfer, as

[Fe(CN)6]3-/4-

can no longer interact with the electrode surface. However, this assay is unable to

18

detect small proteins as their small size is insufficient to block the electrochemical reporter from

reaching the electrode surface.

Figure 2.11. Schematic illustrating the blocking assay, where protein is captured using

antibody-functionalized NMEs, preventing [Fe(CN)6]3-/4-

from reaching the electrode

surface, reducing the measured current[12]

. Reprinted (adapted) with permission from [12]

.

Copyright 2011American Chemical Society.

Another assay developed to detect a number of analytes including small molecules employs a

capture aptamer nucleic acid probe attached to the surface of an electrode, bound to a

neutralizing complementary PNA molecule[61]

. The aptamer preferentially binds to the target as

the neutralizer contains base-pair mismatches, after which a change in the charge at the sensor

surface is detected due to the dissociation of the aptamer-neutralizer complex. This assay can

detect small molecules, but the detection limit of 1 μg·mL-1

is quite high for our application.

Nanostructured microelectrodes have significant potential for ultrasensitive biosensors and,

combined with an electrochemical readout and a creative assay, present an appealing choice for

the purposes of this work.

2.3.2 Nanogaps

Nanogap sensors offer another attractive option for sensing. A nanogap is formed when two

electrodes are separated by submicron distances. Benefits of nanogaps include reducing the

interdiffusion time, low power consumption and reagent volumes, trapping biomolecules, and the

ability to create large electric fields, and in the case of an electrochemical sensor, enabling

analytes to quickly cross the gap and switch between the oxidized and reduced states. The

amperometric response is amplified, thereby increasing the signal.

19

Several sensors have been reported that take advantage of nanogaps to enhance sensitivity. Using

coplanar nanogap electrodes (Figure 2.12) and a conductive linker, 10 pg·mL-1

detection of

cardiac troponin T was achieved[62]

. Conductive linkers were immobilized onto the nanogap

surface between the electrodes and crosslinked with antibodies. As the target antigen was bound

to the antibodies, the measured conductance decreased according to the concentration of antigen

in solution.

Figure 2.12. (A) Schematic and (B) Scanning electron micrograph of a nanogap sensor[62]

© 2016 IEEE.

In another case, nanogaps were used to trap nanoparticles for analysis using dielectrophoresis[63]

.

By shrinking the distance between electrodes to sub-10 nm, strong trapping forces were created

at low biases due to the strong electric field gradients and without the typical challenges

associated with high voltages such as heat generation, bubble formation, and unwanted surface

electrochemical reactions.

A third group fabricated electrochemical nanogap devices based on signal amplification by redox

cycling[64]

. In this work, a current amplification factor of 2.5 was achieved in redox cycling dual

mode compared to single mode, and detection of 5 μM of Fc(MeOH)2 was performed with a

volume of 20 aL, demonstrating extremely low sample volumes.

One type of nanogap is sharp pointed electrodes with interelectrode spacing to match that of the

target analyte. Sensors with this format can result in fringing effects and poor redox

amplification. The second type is better employed for sensing, where there exists both submicron

spacing between electrodes and large electrode surface area[65]

. While most early nanogap

20

devices form coplanar horizontal point-like junctions, many current fabrication techniques also

allow for the formation of larger electrode surface areas, which, combined with the narrow gaps

lead to improved device performance[65]

.

Nanogaps can be formed by a variety of different methods. Conventional nanofabrication

includes techniques such as atomic layer deposition [66-67]

, dip-pen nanolithography[68]

, electron

beam lithography[69-70]

, molecular lithography[71]

, focused ion beam milling[72]

, molecular beam

epitaxy[73]

, nanosphere lithography[74]

, interference lithography[75-76]

, block copolymer

lithography[77-79]

, and galvanic displacement[80-81]

. Unfortunately, the high costs, time

consumption, or lack of control associated with fabrication for these methods act as deterrents

against more widespread use. Thus, new simpler and cheaper fabrication methods are in demand

for this technology to have more applicability.

21

Chapter 3

3 Steric Hindrance Assay for Secreted Factors in Stem Cell Culture

3.1 Introduction

Biosensors often display great sensitivity in controlled conditions, but can suffer from loss or

drift in signal in complex environments, such as in media or solutions from real samples.

Additionally, many targets are small in size and difficult to detect, particularly at low

concentrations, using size-based techniques such as blocking assays. In this chapter, we

investigate methods into improving sensitivity for the detection of small proteins. We discuss

here the development of an electrochemical biosensor based on a novel amplified steric

hindrance assay using nanostructured microelectrodes for the purpose of monitoring soluble

signaling proteins in stem cell culture. As stem cell culture media is heterogeneous and complex,

specific and sensitive quantification is challenging.

We made use of a steric hindrance assay with amplified effects to obtain the necessary

sensitivity. This amplification is performed via careful size-based DNA hybridization

engineering and the use of nanostructured microelectrodes developed by the Kelley Laboratory.

In order to maintain specificity, we used an antibody-based competition scheme to capture the

targets of interest.

This work was an effort to accomplish low-level detection of small signaling proteins influential

in the expansion and differentiation of hematopoietic stem cells (HSCs) grown in culture. We

endeavoured to find a sensing strategy that would allow for possible future integration with a

bioreactor while also minimizing sample volumes and achieving the necessary sensitivity. We

were able to attain 10 pg·mL-1

quantification comparable to that of the current gold standard

while improving on assay time and requiring very little sample volume.

This chapter contains materials from the manuscript:

22

Reprinted with permission from Zhou, W., Mahshid, S.S., Wang, W., Vallée-Bélisle, A.,

Zandstra, P.W., Sargent, E.H., Kelley, S.O., “Steric hindrance assay for secreted factors in stem

cell culture.” ACS Sensors, 2017, 2, 4, 495-500. Copyright 2017 American Chemical Society.

Disclosure of work within this manuscript: W.Z., S.S.M, E.H.S., and S.O.K. designed the

experiments. A. V-B. provided assistance with experimental design. W.Z. performed all

experiments unless otherwise specified and interpreted the results with assistance from E.H.S.,

and S.O.K. W.W. and P.W.Z. designed the experiments regarding the fed-batch bioreactor

culture system and ELISA validation. W.W. provided the cell culture media and performed the

fed-batch bioreactor culture system experiment and ELISA validation. W.Z., S.S.M., E.H.S., and

S.O.K. wrote the manuscript.

3.2 Background

HSC transplantation is used as a clinical therapy for hematological pathologies including blood

cancers and immune system disorders[82-83]

. Umbilical cord blood is an appealing source of

HSCSs[82,84]

, but its clinical use is limited by the low cell numbers available[85]

, prompting the

need for ex vivo expansion. Expansion is made especially difficult by the accumulation of

endogenously produced signaling protein factors secreted from off-target cell populations, which

promote unwanted differentiation[33-35]

. Sensitive and specific detection of the various secreted

proteins that regulate HSC expansion would enable control over their concentrations, improving

ex vivo HSC growth.

Strategies to promote HSC expansion include attempts to minimize the influence of mature

cells[35,86-88]

, through the supplementation of additional factors[89-92]

and regular media exchange

to slow the accumulation of secreted proteins. Even at low concentrations, these factors have a

strong impact on cell fate decisions[93-95]

, with signals from mature blood cells leading to a net

negative effect on HSC expansion (Figure 3.1)[33,35]

. The quantification of signaling factors

would allow for the development of process control strategies to monitor and regulate the

concentrations of these proteins. An integrated sensor to provide sensitive, real-time feedback on

secreted proteins would thus be highly attractive as it would reduce the impact of secreted factors

on HSC differentiation and enhancing HSC expansion.

23

The enzyme-linked immunosorbent assay (ELISA) is the current gold standard method for

protein quantification, but the long process times involved, along with the labels and equipment

required, make this method less convenient for in-line monitoring applications. Several other

techniques exist for protein analysis, such as selected reaction monitoring[37]

, Raman

spectroscopy[96]

, and surface plasmon resonance[97]

, but these methods either do not lend

themselves easily to integration due to the complex equipment required or do not have

sufficiently low limits of detection. Other sensors are based on improving reaction kinetics to

reduce process times and make use of assays employing aptamers[42]

or microbeads[98]

, but are

limited in their widespread use due to the considerable development involved, high costs, and

variability. Specific pg·mL-1

protein quantification in complex media using techniques amenable

to automation is difficult to achieve and poses a significant challenge.

Electrochemical sensors are an attractive option for protein monitoring due to their versatility,

integration capability, and excellent sensitivies/low limits of detection[13,18,54,99-111]

. In particular,

chip-based platforms that make use of modified surfaces such as nanostructured microelectrodes

(NMEs) can achieve multiplexed immunosensing of several factors and have improved

sensitivity compared to that of planar surfaces. While several blocking assays or sandwich assays

have been developed for the analysis of proteins using antibody-modified sensors[12,61,112-113]

, the

detection of low molecular weight proteins at low concentrations has remained difficult. Given

that most secreted factors involved in stem cell differentiation in culture are small proteins with

100 amino acids or fewer, new assay configurations are needed to target the important

application of stem cell culture engineering.

Here, we describe a novel method for the quantification of signaling proteins in primary stem

cell cultures using a sensitive on-chip detection strategy. Drawing inspiration from the design of

a recently developed assay that uses steric hindrance effects to detect large proteins, namely,

antibodies and streptavidin[13]

, we report on a powerful approach to the analysis of small secreted

proteins. Only by combining the use of size-controlled DNA hybridization engineering on three-

dimensional gold NMEs with an alternative competitive antibody attachment scheme, we were

able to improve on the original assay to enhance steric hindrance effects in a new amplified steric

hindrance hybridization assay (ASHHA) for the detection of small proteins. We present a highly

specific protein capturing system and engineer a wide dynamic range from 10 pg·mL-1

to 10

ng·mL-1

for a number of targets that are important for stem cell expansion.

24

Figure 3.1. Schematic showing the influence of secreted factors in stem cell culture and an

overview of the chip-based electrochemical detection scheme. (A) Simplified schematic of the

interactions between soluble factors and cell subpopulations. As the concentrations of signaling

proteins increase, mature cells (pink) accumulate as HSCs (blue) tend toward differentiation. As

the concentration of secreted factors is decreased through media dilution, their impact is reduced,

promoting the proliferation and self-renewal of HSCs. (B) Chip layout. Contacts are formed

from circular apertures in a layer of SU-8 covering a gold pattern on chip surface. (C) Schematic

representation of ASHHA on NMEs. Samples containing the target protein are preincubated with

the blocking antibody. The samples are then mixed with signaling DNA strands labeled with

both the electrochemical reporter and the recognition element before on-chip incubation.

25

3.3 Results and Discussion

3.3.1 Assay and Sensor Chip

The glass microchips used as the sensor platform and the protein detection strategy are illustrated

in Figure 3.1. Gold contacts covered in SU-8 with 5 μm apertures formed the templates for the

100 μm electroplated NMEs. The surface of the NMEs is functionalized with immobilized

thiolated capture DNA strands to form a high density DNA monolayer.

As shown in Figure 1, the detection of small secreted proteins is accomplished by monitoring the

competitive binding of a blocking antibody. The target protein is attached to a strand of DNA,

which is also labeled with the redox-active reporter methylene blue; this is referred to as the

signaling DNA. If the target protein is present in solution at a high concentration, the blocking

antibody will not bind to the conjugated target on the signaling DNA, and this molecular species

is therefore free to bind to the electrode surface, producing a significant level of electrochemical

current. If the target protein is not present, the blocking antibody binds to the signaling DNA,

suppresses the hybridization of the DNA at the sensor surface because of steric hindrance, and

decreases the level of electrochemical current observed. For intermediate levels of protein, the

amount of current would then be proportional to the concentration of target protein. This

approach links increases in current with increased concentrations of protein in solution

irrespective of the size of the target. We take advantage of the competition chemistry in the

ASHHA approach and, through the incorporation of the NME platform and the inclusion of a

large antibody, enable the sensitive detection of small secreted proteins at low concentrations.

3.3.2 DNA-Protein Conjugates and Chip Preparation

We selected three analytes to demonstrate the effectiveness of this approach: (1) regulated on

activation, normal T cell expressed and secreted (RANTES); (2) macrophage-derived chemokine

(MDC); and (3) transforming growth factor-β1 (TGF-β1). All three are secreted factors

deleterious to HSC expansion[114-115]

. These factors are significant modulators in stem cell

culture and therefore monitoring and controlling the concentrations of these proteins is

important. As TGF-β1 is only present in culture in latent form and cannot be measured through

antibody detection without an activation step, we used the latency-associated peptide (LAP) as a

surrogate. LAP binds to TGF-β1 to form the Small Latent Complex[116-117]

and its concentration

26

correlates very well to that of TGF-β1[98]

. We generated target-DNA complexes corresponding to

all three proteins and monitored the steric effects resulting from the binding of the blocking

antibody to the DNA signaling reporter by measuring the differences among electrochemical

signals obtained upon hybridization (Figure 3.2A). The DNA-protein conjugates were incubated

on-chip at a concentration of 30 nM and the signal was measured after 40 min. Square wave

voltammetry was used to scan the sensors for the detection of methylene blue, with the reduction

potential peak located at -0.25 V versus Ag/AgCl.

Figure 3.2. (A) Signal changes upon conjugation of the RANTES, MDC, and LAP

recognition complexes and their respective antibodies to the DNA signaling strands. (B)

Electrochemical signals obtained as a function of time and concentration for the RANTES

assay.

27

3.3.3 Sensor Characterization

As shown in Figure 3.2, the presence of antibodies against RANTES, MDC, or LAP all caused

large changes in the electrochemical current in assays featuring corresponding DNA-protein

signaling conjugates. Addition of the blocking antibodies caused signal changes of 58%, 59%,

and 75% in terms of gain reduction for RANTES, MDC, and LAP, respectively, confirming the

ability of the sensor to detect each of the three target proteins (Figure 3.2A). The largest protein,

LAP produced the largest change in signal, but the smaller proteins also produced measurable

signal changes. We also investigated the time and concentration dependence of the assay for

RANTES (Figure 3.2B), and observed that discernible signal changes could be detected as early

as 10 min.

3.3.4 Determining Dynamic Range and Sensitivity

To determine the dynamic range of this sensor for the detection of signaling proteins, a

concentration series was performed using RANTES and a human RANTES antibody (Figure

3.3). A sample of each concentration of RANTES was mixed with the blocking RANTES

antibody and then incubated with the RANTES-bound signaling strands. The solution was

pipetted onto chips containing NMEs functionalized with capture strands and each electrode was

scanned after 30 min. In buffer solution, the sensors were able to detect concentrations ranging

from 10 pg·mL-1

to 10 ng·mL-1

(Figure 3.3A). This range of detection sensitivity has practical

significance as these proteins are typically present in culture at concentrations between 10

pg·mL-1

to 1 ng·mL-1[118]

. The change in current was measured for each concentration of protein,

and the greatest reduction in current was observed for the lowest protein concentrations.

In order to demonstrate the utility of ASHHA for in-line monitoring of signaling proteins,

detection was also performed in cell culture media, with RANTES spiked into media (Figure

3.3B). The assay is shown to be sensitive even in media, with no signal drift or change in current

compared to results obtained from using buffer. Furthermore, it is specific enough to function

selectively in a heterogeneous solution.

28


concentrations of RANTES in (A) buffer and (B) cell culture media.

3.3.5 Electrochemical Detection in Cell Culture Media Samples

Electrochemical sensors have the advantage of being versatile, and can be easily integrated into a

culture system with the potential for multiplexing. We highlighted our approach by using these

sensors to analyze samples drawn from a HSC fed-batch bioreactor culture system that was made

to dilute soluble factors (Figure 3.4A). Fresh media was added to the culture each day to dilute

the solution, lowering the concentration of the secreted factors and thus minimizing their impact

on HSC expansion. Samples were taken from the culture every 4 days over the course of 12

days, with the sample from day 0 contained no secreted factors and treated as the baseline signal

control sample. The electrochemical current levels of each of the signaling proteins of interest

(RANTES, MDC, and TGF-β1) were measured from these samples, compared to measured

calibration curves, and converted to concentrations, the results of which were validated through

29

comparison with measurements from an ELISA (Figure 3.4)[118]

. The results obtained using the

electrochemical strategy compared very favorably with the ELISA method, indicating that this

new approach displays comparable accuracy relative to the gold standard.

Figure 3.4. Electrochemical detection directly in cell culture media samples. (A) At-line

protein monitoring schematic. Measured concentrations obtained for specific detection of (B)

RANTES, (C) LAP, and (D) MDC in culture samples compared against ELISA measurements

from the same samples. Currents obtained from electrochemical measurements for each factor

were normalized according to calibration curves and converted to concentrations.

3.4 Methods

3.4.1 Materials

The DNA probes were purchased from Biosearch Technologies Inc., Novato, CA. The sequence

of the capture strand used in this work is GGA ATG AAG TCG ATG GAC CTT ACC TGC

CTT GT, with a thiolated 5' terminal. The sequence of the signaling strand is ACA AGG CAG

GTA AGG TCC ATC GAC TTC ATT CC with methylene blue at the 3' terminal and biotin at

30

the 5' terminal. RANTES, MDC, TGF-β1, and their respective antibodies were purchased from

R&D Systems (Minneaopolis, USA). Unless otherwise specified, all other chemicals were

purchased from Sigma-Aldrich and used without any further purification. 1x phosphate buffered

saline (PBS) with a pH of 7.4 was used throughout the experiments.

3.4.2 Chip Fabrication

Chips were fabricated using glass substrates from Telic Company (Valencia, USA) that were

precoated with 5 nm Cr, 50 nm Au, and AZ1600 positive photoresist. Twenty electrodes were

patterned using standard constant lithography while the Cr and Au were etched using their

respective etchants, after which the positive photoresist was removed. A layer of SU-8 2002 was

spin-coated at 5000 rpm, 30 s and patterned using contact lithography to create 5 μm apertures.

Chips were then cleaned with acetone, rinsed with isopropyl alcohol and deionized water, and

dried with a flow of air.

3.4.3 Sensor Fabrication

NMEs were fabricated by electroplating at room temperature in a solution of 20 mM HAuCl4

and 0.5 M HCl at a constant potential of -400 mV for 60 s. Electrodeposition was performed

using an Epsilon potentiostat using a standard three-electrode system with an Ag/AgCl reference

electrode and a Pt counter electrode.

3.4.4 Culture Media

The stem cell culture media used in this work consisted of serum-free IMDM media from Gibco

(Rockville, USA), 20% BIT serum substitute from StemCell Technologies (Vancouver, Canada),

and 1% Glutamax (Gibco). Media was further supplemented with 100 ng·mL-1

stem cell factor,

100 ng·mL-1

F1t3 ligand, 50 ng·mL-1

thrombopoietin, and 1 μg·mL-1

low-density lipoproteins.

3.4.5 Conjugation of Protein with DNA

The conjugation of the proteins (RANTES, MDC, and LAP) to streptavidin was performed using

the Lighting-Link Streptavidin Antibody Labeling Kit from Novus Biologicals (Oakville,

Ontario). These protein-streptavidin complexes were then incubated with the biotinylated

signaling probes for minimum 30 minutes at a 1:3 molar ratio of protein to DNA to form as tock

solution.

31

3.4.6 Preparation of Capture Probe-Modified Chip

A 50 μL immobilization solution containing the 100 nM capture probe solution was incubated on

the chip overnight at 4°C. After the removal of excess solution and washing the chip with PBS,

50 μL of 1.37 mM MCH solution was added to the chip as a backfilling step as well as to block

potential remaining active sites from nonspecific adsorption. Excess solution was further

removed and the chip washed with PBS after 3 h, after which the chip was ready for use and

either kept at room temperature in a humid environment for short-term (less than a day) or at 4°C

for long-term use.

3.4.7 Detection in Buffer and Media

Varying concentrations of RANTES from 10 pg·mL-1

to 10 ng·mL-1

spiked in PBS were

incubated with 5 nM specific antibody for one hour after which the solution was added to a 30

nM signaling probe solution that had previously been conjugated with RANTES. After an hour,

100 μL of this solution was added to the chip for electrochemical detection. For detection in

media, PBS was replaced with stem cell culture media.

3.4.8 Detection in Stem Cell Culture Samples

A solution of 5 nM specific antibody for RANTES, MDC, or LAP was incubated with separate

samples drawn from the culture (from days 0, 4, 8, and 12) for an hour. The samples were then

incubated for an hour with 30 nM signaling probe solution that had been conjugated with the

specific target to be measured. Finally, 100 μL of the sample was added to the chip for

electrochemical detection.

3.4.9 Electrochemical Measurements

All electrochemical measurements were performed using an EmStatMUX potentiostat

multiplexer (Palmsens Instruments, Netherlands) and a standard three-electrode setup with an

Ag/AgCl reference electrode and a Pt counter electrode. For protein detection, the chip was

scanned using square wave voltammetry

Documents

Molecular and Nanoscale Strategies for Enhanced Biosensing · 2018-11-19 · Molecular and Nanoscale Strategies for Enhanced Biosensing Wendi Zhou Doctor of Philosophy ... readout