Improved Proximity Assays for Protein Quantitation 
 
by 
 
Jiaming Hu 
 
 
 
 
A dissertation submitted to the Graduate Faculty of 
Auburn University 
in partial fulfillment of the 
requirements for the Degree of 
Doctor of Philosophy  
 
Auburn, Alabama 
August 3, 2013 
 
 
 
 
Keywords: electrochemical proximity assay, biosensing, 
point-of-care diagnostics, secretion sampling 
 
 
Copyright 2013 by Jiaming Hu 
 
 
Approved by 
 
Christopher J. Easley, Chair, Assistant Professor of Chemistry 
Curtis Shannon, Professor of Chemistry 
Vincenzo Cammarata, Associate Professor of Chemistry 
Wei Zhan, Associate Professor 
 
 
 
 
 
ii 
 
 
Abstract 
 
 
The work presented in this dissertation is focused on the development of proximity 
assay formats for biosensor applications in medical diagnostics, biological research, and 
point-of-care testing.  An electrochemical proximity assay (ECPA) was developed for 
direct detection of a wide variety of target proteins.  The detection limit of antibody-
based reusable ECPA for insulin was 20 fM, which is lower than commercially available 
ELISAs, with a dynamic range through 6.25 nM, more than two orders of magnitude 
wider than the ELISAs.  These improvements come with the added benefit of direct 
electrochemical readout. The measurement of hormone secretion from endocrine tissue 
via reusable ECPA and characterization by surface plasmon resonance (SPR) permit 
ECPA to be utilized in secretion sampling or high-throughput screening applications. 
Chapter 1 introduces basic concepts and common types of biosensors, the origin of 
the concept of proximity assay which plays a crucial role in ECPA, the reason of 
choosing DNA for the development of ECPA model systems, and the properties of 
apamers and antibodies as bioreceptors in biosensor technologies.  In chapter 2, a simple 
and rapid automated microchip electrophoresis method was developed for measurement 
of dissociation constants of DNA aptamers against proteins and small molecules.  
Chapter 3 is mainly focused on the DNA model systems used for the development of 
asymmetric proximity ligation assay (PLA) and ECPA.  Predictions from these model 
systems greatly promoted improvement of assays for detection of human thrombin, 
iii 
 
 
insulin and other biomolecules.  The electrochemical proximity assay is presented in 
Chapter 4, where the first generation of ECPA is composed of two thrombin 
aptamers/insulin antibody-oligonucleotide conjugates which form a cooperative complex 
only in the presence of target molecules, moving a methylene blue (MB)-conjugated 
oligonucleotide close to a gold electrode.  Without washing steps, electrical current is 
increased in proportion to the concentration of a specific target protein.  A faster, easier 
and reusable ECPA is then introduced in Chapter 5, where regeneration of the base DNA 
monolayer was accomplished enzymatically.  Uracils were incorporated into selected 
probe strands, and these strands were enzymatically cleaved using a uracil-DNA excision 
mix.  This allows measurement time to be reduced 30-fold for aptamer-based assay, from 
90 min to 3 min, without significant loss of sensitivity.  The measurement of hormones in 
blood serum and from secretions by pancreatic islets shows that ECPA  could be a 
powerful tool for fundamental biochemistry, clinical analysis, or high-throughput 
screening applications.  These works are not meant to be final, and proposals to include 
ECPA into an electronic device or to integrate ECPA with other detection technologies 
are outlined in Chapter 6. 
 
 
 
 
 
 
 
 
 
 
 
 
 
iv 
 
 
 
 
 
 
 
 
Acknowledgments 
 
 
When I crossed the Pacific from China to the United States and entered into Auburn 
University, fate tied me and this quiet campus tightly together.  In the past five years, I 
have been happy to spend this part of my youth at Auburn University. 
I have a grateful heart toward all my teachers, classmates, friends and family.  I must 
admit that I stared at the word ?Acknowledgement? on my computer screen for a long 
time without writing any words, and I don't know whether "Acknowledgement" can 
express the deep gratitude in my heart.  First of all, I would like to thank my advisor, Dr. 
Christopher J. Easley, a scholar full of wit, an elder of kindness, and a friend of the heart.  
He uses strict solid research to explain his academic viewpoints, a warm heart to teach 
me the ways of getting along with people, and a tolerant mind to remind me how to face 
life with a smile.  His unique personality charms, infects, and inspires me.  I would like to 
thank Dr. Curtis Shannon for his funny, open-minded wisdom that has deeply affected 
me, and acknowledge that this two years? collaboration has given me a much deeper 
understanding of electrochemistry.  I would like to thank Dr. Vince Cammarata for his 
carefulness, rigorous persistence, and keen insight, which has let me see a scholar?s 
infinite love in the academic study.  I would like to thank Dr. Wei Zhan, whose 
experience of life and humble attitude has brought influence and guidance to my study 
v 
 
 
and my life.  And also I would like to acknowledge Dr. David Stanbury for generously 
providing instrumentation for some of my experiments. 
In particular, I would like to offer my sincere gratitude for all members of the Easley 
group (past and present) (Dr. Joonyul Kim, Dr. William Ashby, Jessica Crumbley, 
Kennon Deal, Cheryl DeJournette, Leah Godwin, Abdul Sajib, Jean Negou, 
Subramaniam Somasundaram, Xiangpeng Li, Zac Keenum, Lauren Hoepfner, Louis 
Jackson, Haley Medlen, Ferdous Torabinejad, Amanda Kelley, Ricky Scheuerle, Rebecca 
Sollie, Meagan Pilkerton, and Terrance Weeden), I feel very honored to be a member of 
this big family which has made me feel full of strength in my continuing road of study.  
And special thanks go to Yajiao Yu and Ferdous Torabinejad, who have accompanied me 
in finishing my last project, as well as Tanyu Wang who graduated last year, with their 
help instrumental in the success of the ECPA project. 
At last, I would like to acknowledge my family for their generous support.  They 
have been one of the most important sources of encouragement and love.  And I would 
like to express my immeasurable appreciation to my wife, Na Chen, for her unconditional 
love and constant support during the years of my graduate study.  I dedicate this 
dissertation to the ones I love and respect. 
 
 
 
 
 
 
 
 
 
 
vi 
 
 
 
 
 
Hu, J. and Easley, C. J. Analyst, 2011, 136, 3461-3468. - Reproduced by permission of 
The Royal Society of Chemistry. 
 
Reprinted with permission from Hu, J.; Wang, T.; Kim, J.; Shannon, C.; Easley, C. J. J. 
Am. Chem. Soc. 2012, 134, 7066-7072. Copyright ? (2013) American Chemical Society. 
 
Reprinted with permission from Kim, J.; Hu, J.; Sollie, R. S.; Easley, C. J. Anal. Chem. 
2010, 82, 6976-6982. Copyright ? (2013) American Chemical Society. 
 
Adapted by permission from Macmillan Publishers Ltd: Nature biotechnology 
(Fredriksson, S.; Gullberg, M.; Jarvius, J.; Olsson, C.; Pietras, K.; G?stafsd?ttir, S. M.; 
?stman, A.; Landegren, U. Nature Biotechnol. 2002, 20, 473-477.), copyright? (2013) 
 
Reprinted with permission from Heyduk, E.; Heyduk, T. Anal. Chem. 2005, 77, 1147-
1156. Copyright ? (2013) American Chemical Society. 
 
Reprinted with permission from Gullberg, M.; Gustafsdottir, S.; Schallmeiner, E.; Jarvius, 
J.; BjarnegCard, M.; Betsholtz, C.; Landegren, U.; Simon, F.; Gold, L. PNAS 2004, 101, 
8420-8424. Copyright ? (2013) National Academy of Sciences, USA. 
 
vii 
 
 
 
 
 
 
 
 
Table of Contents 
 
 
Abstract ............................................................................................................................. ii 
Acknowledgments............................................................................................................ iv  
List of Figures ................................................................................................................. xii  
List of Tables .................................................................................................................. xv  
Chapter 1:  Introduction ................................................................................................ 1 
        1.1     Motivation for Research ...................................................................... 1 
                   1.2     Common used Protein-based Detection Methods ................................ 2 
                   1.3     DNA as a Bioengineering Material.................................................... 12 
                   1.4     Basic Concepts of Electrochemical Biosensors ................................. 17 
                   1.5     Aptamers and Antibodies as Bioreceptors ......................................... 20 
                   1.6     Additional Comments ........................................................................ 21 
                   1.7     References .......................................................................................... 23 
Chapter 2:  Dissociation Constants (Kd) Measurement of DNA Aptamers ............. 25 
                   2.1     Introduction ........................................................................................ 25 
                   2.2     Reagents and Experimental Methods................................................. 28 
                              2.2.1     Reagents and Materials ........................................................ 28 
                              2.2.2     Automated Microchip Electrophoresis ................................ 29 
                              2.2.3     Dissociation Constant Determination .................................. 30 
                   2.3     Validation of Method by Measurement of Aptamer-Protein 
viii 
 
 
Dissociation Constants ..................................................................... 33 
                   2.4     Measurement of Aptamer-Small Molecule  
Dissociation Constants ..................................................................... 36 
                   2.5     Method and Dissociation Constant Comparisons .............................. 40 
                   2.6     Feasibility for Kd Monitoring during Aptamer Selection .................. 43 
                   2.7     Conclusions ........................................................................................ 46 
                   2.8     References .......................................................................................... 47 
Chapter 3:  DNA Models for Development of Proximity Assays ............................. 49 
                   3.1     DNA Models for Asymmetric Proximity Ligation Assays................ 49 
                              3.1.1     Background of Proximity Ligation Assay (PLA) ................ 50 
                              3.1.2     Improvement of Sensitivity and Dynamic Range  
in PLAs by Asymmetric Connector Hybridization ............. 51 
                              3.1.3     Reagents and Experimental Methods ................................... 54 
                                            3.1.3.1   Reagents and Materials ......................................... 54 
                                            3.1.3.2   Fluorescence Measurements for  
Model Titrations .................................................. 55 
3.1.3.3   Synthesis of Loop strands ..................................... 56 
3.1.3.4   Proximity Ligation Assays .................................... 57 
                                            3.1.3.5   Connector Kd Calculations .................................... 58 
3.1.3.6   DNA Melting Curve Measurement ....................... 59 
                              3.1.4     Signal and Background in Proximity Ligation .................... 61 
                              3.1.5     Design of Experimental Model of  
Proximity Hybridization ..................................................... 62 
                              3.1.6     Complex Formation with Asymmetric Connectors ............. 64 
                              3.1.7     Background Reduction in the Experimental Model............. 65 
ix 
 
 
                              3.1.8     Confirming the Experimental Model in  
Asymmetric PLA ................................................................ 71 
                                            3.1.8.1   Design of Asymmetric PLA.................................. 71 
                                            3.1.8.2   Background Reduction Using  
Asymmetric PLA ................................................. 72 
                                            3.1.8.3   Dynamic Range and Sensitivity  
Enhancement with Asymmetric PLA .................. 73 
                   3.2     DNA-Based Experimental Model of Electrochemical  
                              Proximity Assay (ECPA) .................................................................. 77 
                              3.2.1     Background .......................................................................... 77 
                              3.2.2     Model Systems of ECPA ..................................................... 80 
                   3.3     Conclusions ........................................................................................ 83 
                   3.4     References .......................................................................................... 84 
Chapter 4:  The Electrochemical Proximity Assay (ECPA) ..................................... 85 
                   4.1     Introduction ........................................................................................ 85 
                   4.2     Reagents and Experimental Methods................................................. 88 
                              4.2.1     Reagents and Materials ........................................................ 88 
                              4.2.2     Preparation of the Electrode and  
DNA Monolayer Assembly ................................................ 89 
                              4.2.3     ECPA Probe Assembly and  
Electrochemical Measurements .......................................... 91 
                              4.2.4     Preparation of Antibody-Oligonucleotide Conjugates ......... 94 
                   4.3     Signal and Background in ECPA ....................................................... 95 
                   4.4     DNA-Based Experimental Model of ECPA ...................................... 98 
                   4.5     Aptamer-Based ECPA ..................................................................... 102 
                   4.6     Antibody-Based ECPA .................................................................... 106 
x 
 
 
                   4.7     Conclusions ...................................................................................... 111 
                   4.8     References ........................................................................................ 112 
Chapter 5:  The Reusable Electrochemical Proximity Assay (ECPA) ................... 115 
                   5.1     Abstract of ECPA ............................................................................ 115 
                   5.2     Introduction ...................................................................................... 116 
                   5.3     Reagents and Experimental Methods............................................... 119 
                              5.3.1     Reagents and Materials ...................................................... 119 
                              5.3.2     Preparation of Sensor Surface  ........................................... 120 
                              5.3.3     ECPA Probe Assembly and  
Electrochemical Measurements  ....................................... 122 
                   5.4     Principle of reusable ECPA ............................................................. 124 
                   5.5     DNA-Based Experimental Model of Reusable ECPA..................... 127 
                   5.6     Regeneration Comparisons between Uracil Enzyme 
and Several General Methods ......................................................... 129 
                   5.7     Aptamer-Based Reusable ECPA ...................................................... 130 
                   5.8     Antibody-Based Reusable ECPA .................................................... 133 
                   5.9     Temporal Measurement of Hormone Secretion from 
Endocrine Tissue ............................................................................. 135 
                   5.10    Measurement of Insulin from Murine Blood Serum....................... 138 
                   5.11     Characterization by Surface Plasmon Resonance .......................... 139 
                   5.12    Conclusions ..................................................................................... 142 
                   5.13    References ....................................................................................... 143 
Chapter 6:  Future Directions .................................................................................... 145 
                   6.1     Future Portable Electronic Device ................................................... 145 
                   6.2     Prospect of Detection Technology ................................................... 147 
xi 
 
 
 
List of Figures 
 
 
Figure 1-1  ......................................................................................................................... 4 
Figure 1-2  ......................................................................................................................... 7 
Figure 1-3  ......................................................................................................................... 8 
Figure 1-4  ....................................................................................................................... 10 
Figure 1-5  ....................................................................................................................... 11 
Figure 1-6  ....................................................................................................................... 12 
Figure 1-7  ....................................................................................................................... 13 
Figure 1-8  ....................................................................................................................... 14 
Figure 1-9  ....................................................................................................................... 16 
Figure 1-10  ..................................................................................................................... 18 
Figure 1-11  ..................................................................................................................... 19 
Figure 2-1  ....................................................................................................................... 32 
Figure 2-2  ....................................................................................................................... 34 
Figure 2-3  ....................................................................................................................... 39 
Figure 2-4  ....................................................................................................................... 45 
Figure 3-1  ....................................................................................................................... 51 
Figure 3-2  ....................................................................................................................... 53 
Figure 3-3  ....................................................................................................................... 57 
xii 
 
 
Figure 3-4  ....................................................................................................................... 60 
Figure 3-5  ....................................................................................................................... 63 
Figure 3-6  ....................................................................................................................... 67 
Figure 3-7  ....................................................................................................................... 70 
Figure 3-8  ....................................................................................................................... 75 
Figure 3-9  ....................................................................................................................... 78 
Figure 3-10  ..................................................................................................................... 79 
Figure 3-11  ..................................................................................................................... 81 
Figure 3-12  ..................................................................................................................... 82 
Figure 4-1  ....................................................................................................................... 95 
Figure 4-2  ....................................................................................................................... 97 
Figure 4-3  ....................................................................................................................... 98 
Figure 4-4  ..................................................................................................................... 100 
Figure 4-5  ..................................................................................................................... 102 
Figure 4-6  ..................................................................................................................... 105 
Figure 4-7  ..................................................................................................................... 108 
Figure 5-1  ..................................................................................................................... 122 
Figure 5-2  ..................................................................................................................... 126 
Figure 5-3  ..................................................................................................................... 128 
Figure 5-4  ..................................................................................................................... 129 
Figure 5-5  ..................................................................................................................... 132 
Figure 5-6  ..................................................................................................................... 134 
Figure 5-7  ..................................................................................................................... 137 
xiii 
 
 
Figure 5-8  ..................................................................................................................... 139 
Figure 5-9  ..................................................................................................................... 141 
Figure 6-1  ..................................................................................................................... 146 
Figure 6-2  ..................................................................................................................... 147 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
xiv 
 
 
 
List of Tables 
 
 
Table 1-1  ........................................................................................................................ 15 
Table 2-1  ........................................................................................................................ 29 
Table 2-2  ........................................................................................................................ 36 
Table 3-1  ........................................................................................................................ 55 
Table 3-2  ........................................................................................................................ 59 
Table 4-1  ........................................................................................................................ 89 
Table 4-2  ...................................................................................................................... 110 
Table 5-1  ...................................................................................................................... 120 
 
 
 
 
 
 
 
 
 
 
 
                                                                                
1 
 
 
CHAPTER 1  
Introduction 
 
1.1 Motivation for Research 
The term ?proteome? became popular after the draft of human genome was done in 
2003. Although it has long been known that protein levels are of fundamental importance 
in biological function, technologies for protein separation, detection and quantitation 
have seen accelerated development with research into the human ?proteome.? 
For protein separation, 2-dimentional gel electrophoresis was the core technology 
used in 1980s. Even today, it is still a very common separation method in clinic and 
research laboratories. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-
PAGE) and liquid chromatography (LC) have both progressed in technological 
development. For protein identification, mass spectrometry (MS) has been a popular 
choice, and the technology of GC-MS and LC-MS realized the integration of protein 
separation and detection.1-3 However, these analytical methods require expensive, bulky, 
and complicated instruments. Another useful method for protein quantitation is surface 
plasmon resonance (SPR), the advantages of which are not requiring markers or dyes, 
reaction process can be monitored in real-time and fast. However, this optical method is 
sensitive to the environment, such as light and temperature conditions, and also sensitive 
to the mobile phase composition, such as ion concentration, and contaminants.  
2 
 
 
In recent years, the frequent exposure of global food safety problems and outbreak 
of new diseases require more accurate food quality control methods and new diagnostics. 
Those medical diagnostics and food safety depend on methods that can detect and 
quantify disease-related proteins and contaminants, often at very low concentrations. In 
this light, development of improved methods with high-throughput, high sensitivity, and 
high accuracy is a main direction for protein detection research. 
 
1.2 Commonly used Protein-Based Detection Methods  
Western Blotting  
The name ?Western Blot?, like ?Northern Blot?, came mainly from scientists? sense 
of humor.  In 1975, Edwin Southern introduced a powerful DNA transfer and probing 
technique,4  which was named ?Southern Blot? after its inventor.  In 1977, George Stark 
and his co-workers successfully transferred cellular RNA to chemically activated 
cellulose paper.5  This technique was name ?Northern Blot? ? as a pun on Southern?s 
name.  In 1979, George Stark developed an early protein blotting technique, which relied 
on capillary transfer from gel to cellulose paper.6 After that, Harry Towbin developed a 
faster and simpler technique for electroblotting proteins to nitrocellulose membranes,7  
which is still widely used today.  In 1981, W. Neal Burnette8 first gave a name to Stark 
and Towbin?s technique ? ?Western Blot?.  At this point, the ?Western Blot? technique is 
about 32 years old, yet it has changed very little since Towbin?s modifications.  In 2008, 
Mathias Mann published a paper entitled ?Can Proteomic Retire the Western Blot??.9  In 
his view, mass spectrometry can cover most cellular proteins in a quantitative 
experiment10 which contains the ones detected by western blotting. However, with the 
3 
 
 
improvement of gel electrophoresis, membrane technology and increased commercial 
antibodies, especially the appearance of higher affinity antibodies, the dynamic range of 
western blotting has been extended.  Western blotting has become a simpler, diverse 
analysis method, and it is a very useful quantitative tool. Because of its cheap equipment 
and mature technology, the Western Blot will be used for protein detection for a long 
time, and it already has been applied to detection of HIV, bovine spongiform 
encephalopathy (BSE), lyme disease, hepatitis and other diseases.11-13  
More specifically, Western blotting is a technique of protein separation and 
detection by integrating gel electrophoresis, electrophoretic transfer, and antibody-based 
probes. It uses gel electrophoresis to separate native proteins by 3-D structure or 
denatured proteins by the length of the polypeptide. The proteins are then transferred to a 
solid support, where they are stained with antibodies specific to the target protein, 
detecting the target proteins from the protein mixture. Thus, the method qualitatively or 
quantitatively determines target proteins expressed in cell or tissue. There are various 
protocols for Western blotting, but more or less, it has several basic steps as shown in 
Figure 1-1A. 1) Sample preparation; as we know, proteins are composed of a variety of 
amino acids, which have amino, carboxyl groups and charged R groups, so different 
proteins have different charges. If we want the protein mixture to move towards the same 
direction, it has to be treated. The commonly used reagent is sodium dodecylsulfate 
(SDS), which is an anionic detergent that denatures secondary and nondisulfide-linked 
tertiary structures, thus unfolding and linearizing the proteins completely, and it coats 
proteins with many negative charges. This way, the proteins can be separated in an 
electric field strictly by their length or size, respectively. 2) Electrophoresis; the 
4 
 
 
 
 
 
Figure 1-1: Western blot. A) Western blot flow chart. B) Enhanced 
chemiluminescence detection for blotting analysis. 
 
denatured proteins are placed into the lanes of the stacking site of a polyacrylamide gel, 
and put into an electric field, so they will move towards the positive pole at different rates. 
3) Blotting: transfer to solid support; after proteins separated, they are transferred to a 
solid membrane to ease labeling and detection steps. 4) Blocking; the used membranes 
must be treated to block nonspecific binding between probes and the material. 5) 
Hybridization; adding the antibody-based probes to the blocked membrane.  In Western 
blot, the secondary antibody can be added after the unbound primary antibody is removed 
by washing the memebrane. 6) Detection; an enzyme linked onto the secondary antibody 
will catalyze the production of a chemiluminescent or strongly absorbing agent in 
proportion to the amount of bound protein (Figure 1-1B).  
 
 
 
5 
 
 
Enzyme-linked immunosorbent assay (ELISA) 
In 1971, Engvall et al. used cellulose and polystyrene tubes as the solid phase carrier 
for antigen or antibody adsorption, and they established an enzyme-linked 
immunosorbent assay, referred to as ELISA. Since then, the move to well-plate ELISAs 
has greatly improved the experimental throughput.14 ELISA has high flexibility. In the 
actual operation, people can perform various kinds of ELISA reactions, according to their 
own needs and materials on hand.  Also, the reagents and supplies for ELISA 
experiments have been commercialized, and the equipment has been more-or-less 
standardized. Automation has further improved ELISA, which has become the core 
method for disease diagnostics in clinical laboratories. Many authoritative international 
organizations list it as one of the priority analysis techniques.15  
ELISA is an enzyme immunoassay in the stationary phase. For this, there are three 
important requirements. First, proteins must be adsorbed onto the stationary phase 
without changing its activity; second, the antigen or antibody must still maintain its 
activity after crosslinking with an enzyme; and third, after cross-linking, enzymes must 
efficiently react with the substrate to produce a colored or luminescent substance or 
generate light in the reaction.  
Public opinions on classification of ELISAs are divergent. Different literatures have 
different opinions about the classification of ELISA from the principle and operation. 
Here, ELISA is divided into the following three categories, then for each class is 
explained in detail: (1) direct ELISA; (2) indirect ELISA; (3) sandwich ELISA. (Figure 
1-2) Direct ELISA can be regarded as the simplest type of ELISA. In direct ELISA, the 
antigen is adsorbed to a plastic plate, then an excess of other protein (using bovine serum 
6 
 
 
albumin) is added to block all the other binding sites. Then an enzyme is linked to an 
antibody in a separate reaction. The enzyme-antibody conjugate is allowed to adsorb to 
the antigen, then excess enzyme-antibody conjugate is washed off, leaving enzyme-
antibody bound to antigen in proportion to the antigen concentration. By adding the 
enzyme?s substrate, the enzyme is detected and thus the antigen. The advantage of direct 
ELISA is that it can be performed faster, because only one antibody is being used and 
fewer washing steps are required. This can be used to test specific antibody-to-antigen 
reactions, and helps to eliminate cross-reactivity between other antibodies. But, in direct 
ELISA, the primary antibody must be labeled individually, which can be costly, time-
consuming, and inflexible when performing multiple experiments.  Additionally, the 
signal is less amplified, which can be seen as a disadvantage. With a slight modification, 
indirect ELISA is a two-step method that uses a primary antibody and labeled secondary 
antibody. The primary antibody is incubated with the antigen followed by the incubation 
of the secondary antibody and corresponding washing steps. Although containing more 
steps, this allows more flexibility in operation with different analytes. However, both of 
these methods require immobilization of the antigen. 
One of the most common types of ELISA is the ?sandwich ELISA?. The "sandwich" 
type measures the amount of antigen between two layers of antibodies (i.e. capture and 
detection antibody). The antigen to be measured must contain at least two antigenic sites 
capable of binding to an antibody, since at least two antibodies act in the sandwich.  
 
 
 
7 
 
 
 
 
 
 
Figure 1-2:   Common ELISA formats.  
 
 
 
Figure 1-3 describes the protocol of sandwich ELISA: 1) The plate is coated with a 
suitable capture antibody, and the remaining protein-binding sites on plate are blocked; 2) 
sample is added to the plate, and any antigen present is bound by the capture antibody; 3) 
a suitable detection antibody is added to the plate and also binds to any antigen present in 
well; 4) an enzyme-linked secondary antibody is added, and it binds to the detection 
antibody; 5) a substrate is added and is converted by the enzyme to a detectable form. 
The sandwich ELISA has become one of the more popular versions due to its flexibility 
and propensity for automation.  However, a disadvantage to all forms of ELISA is the 
need for multiple binding, coating, and washing steps, which also requires significant 
8 
 
 
 
 
 
Figure 1-3:   Protocol of sandwich ELISA. 
Image modified from http://en.wikipedia.org/wiki/File:ELISA-sandwich.svg 
 
expertise from the user. These requirements basically limit ELISAs to clinical 
laboratories rather than point-of-care or on-site measurements. 
 
 
Proximity Assay Concept 
Advancements in science and technology tend to make complex things more simple 
over time. Users are never satisfied with the existing technologies. Although Western 
blots and ELISAs are very good methods for protein detection, the disadvantages are 
obvious. The operation process in both methods is complicated, and also needs multiple 
wash steps. So keeping the advantages of Western blots and ELISAs while avoiding the 
disadvantages becomes a challenge for assay development. In fact, Western blots and 
ELISAs can be considered the prototype of proximity assays. 
9 
 
 
Proximity assays have two characteristics: 1) two probes binding target molecule at 
different binding sites; and 2) signal based on some proximity-driven connection between 
the two probes. In our systems, taking thrombin as the target molecule for example, two 
probes can be aptamers or DNA conjugated antibodies binding thrombin at different 
sites. When pairs of probes bind thrombin, the free ends of their DNA sequence 
extensions are brought sufficiently close (into ?proximity?) to hybridize together to a 
subsequently added connector oligonucleotide. Alternatively, the signal DNA molecule 
binding to one probe can be brought close to another probe to hybridize together directly.  
This coupling of dual-probe to target binding for protein detection was used in both 
Western blotting and ELISAs, and then developed into sandwich ELISAs. The proximity 
assay concept was first adopted in a report by Landegren?s group. In 2002, their team 
developed a method for in vitro analysis of proteins and other macromolecules and 
termed the method proximity ligation assay (PLA).16,36  By inspection of Figure 1-4, 
without much thought, one should realize that this assay is similar to the sandwich 
ELISA, where two probes bind the target molecule at different binding sites. In fact, the 
proximity ligation assay fundamentally depends on the simultaneous and proximate 
recognition of target molecules by pairs of probes. When the probes bind the target 
molecule, the ends of their DNA sequence extensions are brought close?into 
proximity?to hybridize to an added connector oligonucleotide. Next, the free ends  are 
joined by DNA ligation, and the ligated products can be replicated by the polymerase 
chain reaction (PCR), giving an amplifiable detection signal in proportion to the original 
target protein?s concentration.  
10 
 
 
 
 
 
Figure 1-4:   The principal steps of the proximity ligation assay (PLA). 
Step 1, incubation of sample with proximity probe pair; step 2, addition of 
all components required for ligation and detection by quantitative PCR; 
step 3, quantitative PCR.36 
      Reprinted with permission from Gullberg, M.; Gustafsdottir, S.; Schallmeiner, E.; Jarvius, J.; 
BjarnegCard, M.; Betsholtz, C.; Landegren, U.; Simon, F.; Gold, L. PNAS 2004, 101, 8420-8424. 
Copyright ? (2013) National Academy of Sciences, USA. 
 
Comparing to sandwich ELISA, based on Landegren?s results, PLA has comparable 
sensitivity and detection limit for thrombin, and also it has a wider dynamic range. More 
importantly, this assay is homogeneous, does not need washing steps, making it a kind of 
single tube immunoassay. And in PLA, the sample preparation time is less than one hour, 
and the automated qPCR assay can be finished in two hours, so the technique is much 
simpler and faster than sandwich ELISA.  
 
11 
 
 
The assay obtains an amplifiable 
detection signal with a qPCR 
nuclease assay. A fluorogenic probe 
is frequently used in the nuclease 
assay, which contains an 
oligonucleotide to which a 
fluorogenic reporter dye and quencher 
dye are attached. (Figure 1-5) During 
the annealing phase in PCR, the 
labeled probe anneals to the target of 
interest between the forward primer 
and reverse primer sites. In the 
extension phase, the primers bind the 
DNA template, and a new DNA 
sequence is synthesized by DNA polymerase complementary to DNA template. The 
labeled probe is cleaved during extension phase, and the reporter dye will be separated 
from quencher dye, releasing highly fluorescent reporters. 
In 2005, Tomasz Heyduk and his group developed another assay based on the 
proximity assay concept, termed ?molecular pincers?.17 As shown in Figure 1-6, two 
different aptamers bind different sites of a protein target, and at the end of each aptamer, 
short fluorophore-labeled oligonucleotides (which were attached to the aptamers by non-
DNA linker) are assembled. In the absence of the target protein, the two aptamer half-
sites would not associate, since the number of complementary oligonucleotides was too 
 
 
Figure 1-5: Nuclease assay (Taqman). 
a) annealing phase; b) extension phase 
I; c) extension phase II; d) end of PCR 
cycle. 
http://upload.wikimedia.org/wikipedia/en/0/07/Ta
qman.png 
a 
b 
c 
d 
12 
 
 
small to promote efficient annealing. When the half-sites of aptamers are bound to the 
target protein, it would bring the two signaling oligonucleotides close to each other, into 
relative proximity, increasing their local concentrations. This results in the annealing of 
the signaling oligonucleotides, which brings the fluorophore and the quencher (or FRET 
pair) into proximity, resulting in a change of 
fluorescence signal. 
The development of proximity assays 
successfully overcomes some limitations of 
Western blots and ELISAs. It makes three 
breakthroughs: first, the assays are 
homogeneous, such that no wash steps are 
needed and measurement time and complexity 
can be greatly reduced. Second, the binding 
probes can be aptamers, not just antibodies. 
Third, the reaction can be done in a smaller 
tube, saving significant amounts of reagents. 
In PLA, these benefits were realized without 
loss of sensitivity compared to ELISA. 
 
1.3 DNA as a Bioengineering Material 
Due to in the following chapters, multiple DNA sequences are used in the model 
systems for assay development. For this reason, it is necessary to mention the basic 
concepts of DNA and its interactions. 
 
 
 
Figure 1-6: Molecular beacons 
for detection proteins. Molecular 
beacons for detecting proteins 
based on aptamers directed to 
two different epitopes of the 
protein.17 
Reprinted with permission from Heyduk, E.; 
Heyduk, T. Anal. Chem. 2005, 77, 1147-
1156. Copyright ? (2013) American 
Chemical Society. 
 
13 
 
 
DNA biotechnology, arising in the 1980s, is defined bytechnical developments, 
which are improving at a rate as fast or faster than improvements in computing 
technology. ?Carlson?s Law? shows that the prices of DNA synthesis and sequencing are 
dropping exponentially in time (Figure 1-7). 
 
 
 
The use of biological objects in ways different from their naturally intended use is 
not a new concept. One could argue that this idea traces back to ancient times, when 
people used animal bones as a rigid material for axes and other tools. Similarly, materials 
 
 
 
Figure 1-7:   Carlson?s Law. The price of DNA oligonucleotide and gene 
synthesis has been dropping exponentially over the past 25 years.  
Image by Robert Carlson, http://www.synthesis.cc/carlson_cost%20per_base_ 
june_2012.png 
14 
 
 
have been constantly refined through advancements in science and technology and the 
progression of time. Similarly, DNA and other nucleic acids are likely not the best 
materials with which we can make biosensors or nanoscale tools.  At some point in the 
future, it is likely that DNA will be replaced by synthetic molecules with superior 
properties. However, DNA is material that can be cheaply and widely used today, and it 
does have well-understood interaction properties.  DNA can form defined secondary and 
tertiary structures with specific properties and can bind with molecules to allow biosensor 
studies and diagnostic research.  
 
Most people understand that DNA plays an important role in inheritance (?I have 
my dad?s nose and my mom?s hair.?). Figure 1-8 presents a molecular view of DNA, 
showing that it has three components: a sugar, a nitrogen-containing base, and a 
phosphoric acid. Described by the Watson - Crick structural model, a DNA molecule 
consists of two polynucleotide strands coiled around each other in helical ?twisted ladder? 
 
 
 
Figure 1-8:   The structure of DNA. 
http://upload.wikimedia.org/wikipedia/commons/d/d8/Benzopyrene_DNA_adduct_1JDG.
png 
 
15 
 
 
structure. However, there are many more structural intricacies in DNA, not just a double 
helix. 
In Dr. David Yu Zhang?s thesis,18  he explained the biophysical properties of DNA 
in very vivid way by analogy to Velcro and magnets. In DNA, there are four nucleotide 
bases: adenine, cytosine, guanine, and thymine (A, C, G and T) which can hybridize to 
form a stable duplex (double helix) bound together by hydrogen bonds between 
complementary base pairs (A-T and C-G). Table 1-1,18 quoted from Zhang?s thesis, 
shows the names and definitions of frequently used concepts in DNA nanotechnology. 
 
 
 
 
 
 
 
 
Table 1-1: Common terms and their synonyms and definitions.18 
 
16 
 
 
 
  
Figure 1-9:  DNA strand displacement overview. Panel shows one 
example of this reaction. Single stranded DNA molecule X reacts with 
multi-stranded DNA complex S to release strand Y and complex W. This 
reaction is facilitated by the ?toehold? domains 2 and 2 : The hybridization 
of these single-stranded toeholds colocalizes S and X, and allows the 1 
domain to branch migrate. Branch migration is the random walk process 
in which one domain displaces another of identical sequence in binding 
to a complementary domain, via a series of single nucleotide dissociation 
and hybridization steps. At the completion of branch migration, complex 
W is formed and strand Y is released.18 
Image modified from David Yu Zhang?s Thesis: Dynamic DNA Strand Displacement 
Circuits, California Institute of Technology, Pasadena, California, 2010. 
 
 
Also covered in Zhang?s thesis are DNA strand displacement reactions. When one 
single-stranded nucleic acid molecule hybridizes to a partially double-stranded DNA, it 
can cause strand displacement. This reaction starts at the single-stranded ?toehold? 
domains and progresses through a branch migration process.18  This way, the originally 
bound DNA strand is released. An overview of the DNA strand displacement process is 
given in Figure 1-9. In Chapter 2 of this dissertation, a competitor oligonucleotide was 
used to bind with ATP aptamer to unfold its hairpin structure in the measurement of 
17 
 
 
aptamer-small molecule dissociation constants. In Chapter 4, the standard 
electrochemical proximity assay (ECPA) adopted a short DNA competitor strategy, 
where the DNA competitor has 9 bases complimentary to thiolated-DNA, and a one-base 
toehold from thiolated-DNA can bind with methylene blue labeled DNA (MB-DNA) 
permitting strand displacement with the help of the proximity effect.   
 
1.4 Basic concepts of Electrochemical Biosensors 
Detection systems frequently used in medical diagnostics are mainly based on 
optical methods and electrochemical methods.19-26 With optical approaches, light sources 
and optical sensors are typically expensive, and measurements are sensitive to local light 
and temperature. Comparing to the optical methods, electrochemical detection can offer 
greater signal stability and consistency, high sensitivity, simple instrumentation, and 
great compatibility with miniaturization technologies21,22. From the beginning of this 
century, electrochemical biosensors were used in many aspects in diagnostics.27,28  In 
addition, electrochemical biosensors are easier to be developed into portable devices with 
microfluidic technologies.29  These are the main reasons that we chose to combine 
electrochemical detection with the proximity assay concept for assay development 
Electrochemical biosensors measure the changes in the electrical properties of a 
sample solution based on the chemical reactions which cause production or consumption 
of ions or electrons. Typically, an electrochemical biosensor comprises of 1) a 
bioreceptor specifically binding to the analyte; 2) a transducer which transduces the 
change taking place on the interface with bioreceptor to the electronic system; and 3) an 
electronic system to transfer this electrical signal to computer software that converts it to 
18 
 
 
a meaningful physical parameter presented to the operator. (Figure 1-10) The 
electrochemical biosensors, based on the measured parameters, can be classified as 
conductimetric, potentiometric, and amperometric. Usually, conductimetric biosensors 
measure electrical conductance or resistance.21,30,31  Potentiometric biosensors measure 
the potential of electrochemical cells in the absence of appreciable currents.32-34  
Amperometric biosensors measure current.21  Generally, if the current is measured at a 
constant potential, it is referred to as amperometry, and if the current is measured in a 
controlled range of potential, it is referred to as voltammetry.30  
 
 
 
 
 
 
Figure 1-10:   A schematic representation of biosensors 
 
19 
 
 
In the following 
electrochemical proximity 
assay (ECPA) and reusable 
ECPA, methylene blue (MB) is 
the reporter molecule, and 
square-wave voltammetry 
(SWV) is used to measure the 
current change caused by 
electrons transferring between 
MB and gold electrode. SWV 
is a kind of pulse polarography, 
which can offer great speed 
and high sensitivity. The great 
sensitivity of SWV can be 
attributed to two reasons. First, 
it can enhance the faradaic 
current; second, it can decrease 
the nonfaradaic charging 
current. In SWV, the current is 
measured twice during each 
cycle, once at the end of the 
forward pulse, and again at the 
end of reverse pulse. (Figure 1-
 
  
 
Figure 1-11:   Applied excitation in square-
wave voltammetry. The staircase signal in (A) 
is added to the symmetrical square wave in 
(B) to give the square-wave excitation signal 
in (C). 
 
20 
 
 
11C) Taking our experiment as an example, let us consider the events that occur on the 
electrode surface as the potential is suddenly increased by 50 mV. If a MB molecule is 
present near the electrode surface, there will be a surge of current that lowers the 
concentration of reduced MB molecules to that demanded by the new potential. When the 
potential pulse is first applied to the gold electrode, a surge in the nonfaradaic current 
(charging current) also occurs. However, the charging current decays exponentially with 
time. By measuring the current at the end of each pulse, the charging current is greatly 
reduced, and signal-to-noise ratio is larger, thus enhancing sensitivity. Figure 1-11 shows 
the excitation signal in SWV, which is obtained by super-imposing a symmetrical square 
wave onto the stair case signal. The magnitude of each pulse, Esw, is one-half of the peak-
to-peak amplitude of the signal. The duration of each pulse is one-half the staircase 
period, ?/2. The frequency of the signal is the reciprocal of the staircase period: f = 1/?. 
And the scan rate for a SWV experiment can be defined as:  
                 ( )S c a n  R a te  ( m v /s e c )  =  
 ( s e c )s t e p
E m V
?
                             (1) 
The currents are measured during the last few microseconds of each pulse and the current 
measured on two pulses of the same step is recorded as: 
                                                      12-I i i??                                                          (2) 
 
1.5 Aptamers and Antibodies as Bioreceptors  
Biomolecules are very commonly used as bioreceptors for biosensors in detection 
and quantitation of target proteins in clinical and biomedical research.21,22,35  Bioreceptors 
21 
 
 
are the key to specificity for biosensor technologies. Aptamers and antibodies are two 
bioreceptors used commonly, due to the development of SELEX (systemic evolution of 
ligands by exponential enrichment) and monoclonal technology. Both methods have 
reduced the selection time and price and give these types of probes diversity and 
specificity.  
Aptamers are short single-stranded oligonucleotides, either RNA or DNA, that fold 
into secondary or tertiary structures to specifically bind to their targets (e.g. small 
molecules, peptides and proteins) with high affinity and specificity. Antibodies are 
proteins consisting of varied amino acids and have very specific binding capabilities for 
the target of interest. 
Particularly, aptamers have several advantages compared to antibodies: first, 
aptamers can be selected in vitro for any given target, from small molecules to large 
proteins and even cells. Second, aptamers can be synthesized with high reproducibility 
and purity from commercial sources. Third, aptamers are usually highly chemically stable, 
and once denatured under high temperature, they can reform their specific structure 
reversibly when the temperature is decreasing. Third, aptamers often undergo significant 
conformational changes upon target binding. However, the binding affinities of aptamers 
to target molecules are not as strong as that of antibodies, and the existing number of 
aptamers is still much lower than that of antibodies.  
 
1.6 Additional Comments  
In Chapter 2 of this dissertation, two different types of aptamers are chosen, which 
specifically bind to small molecules (ATP) or large proteins (thrombin, human IgE). 
22 
 
 
There, we introduce a method to measure the dissociation constant (Kd) of DNA aptamers 
to these target molecules. Because the dual-antibody recognition concept has served as a 
guide to various alternative strategies in the past few years,36-39  in Chapter 3, a proximity 
ligation assay (PLA) is introduced and optimized using DNA model systems. As noted 
before, PLA is a simpler and more sensitive assay format for protein detection and 
quantitation comparing to standard sandwich ELISA. We show in Chapter 3 that with 
careful engineering of DNA strands, experimental model systems can be developed to 
accurately describe trends in PLA and to ease optimization of assay conditions.  In 
Chapters 4 and 5, the electrochemical proximity assay (ECPA) is introduced, which 
cleverly combines the proximity assay concept with electrochemical detection. Both 
aptamers and antibody-oligonucleotide conjugates are used as assay probes in ECPA. 
Compared to antibodies, nucleic acids are easily modified with a variety of reporter 
molecules, such as methylene blue and ferrocene. The DNA ?tails? of the pairs of probes 
have complementary bases to the DNA sequences forming a DNA monolayer on the 
electrode surface and to methylene blue labeled DNA. So in the presence of target 
molecules, the antibody part of the probes would specifically bind to the target, and the 
DNA hybridization would bring the methylene blue molecule close to the electrode 
surface, giving an electric current signal. Finally, general conclusions from this work, 
along with future possibilities, are presented in Chapter 6. 
 
 
 
 
23 
 
 
 
1.7 References 
 
(1)      Salm, P.; Taylor, P. J.; Roberts, D.; Silva, J. D. J. Chromatogr. 2009, 877, 568. 
(2)      Silva, C. J.; Onisko, B. C.; Dynin, I.; Erickson, M. L.; Requena, J. R.; Carter, J. M.   
Anal. Chem. 2011, 83, 1609-1615.      
(3)      Silva, C. L.; Passos, M.; Camara, J. S. Talanta 2012, 89, 360-368.           
(4)      Southern, E. J Mol Biol 1975, 98, 503?517.      
(5)      Alwine, J.; Kemp, D.; Stark, G. Proc Natl Acad Sci USA 1977, 74, 5350?5354.      
(6)      Renart, J.; Reiser, J.; Stark, G. Proc Natl Acad Sci USA 1979, 76, 3116?3120.      
(7)      Towbin, H.; Staehelin, T.; Gordon, J. Proc Natl Acad Sci USA 1979, 76, 4350?
4354.      
(8)      Burnette, W. Anal Biochem 1981, 112, 195?203.      
(9)      Mann, M. J. Proteome Research. 2008, 7, 3065 
(10) Cox, J.; Mann, M. Cell 2007, 130, 395-398. 
(11) Papadoupulos-Eleopulos, E.; Turner, F.; Papadimitriou, M. 
Bio/Technology 1993, 11, 696-707.  
(12) Porcario, C.; et al. BMC Research Notes, 2011, 4, 376. 
(13) Li, D.; et al. J. Addiction Medicine, 2012, 6, 311-317. 
(14) Engvall, E.; Perlmann, P. Immunochemistry 1971, 8, 871-874. 
(15) Conaway,  J. J AOAC Int. 1991, 74, 715 -717 
(16) Fredriksson, S.; Gullberg, M.; Jarvius, J.; Olsson, C.; Pietras, K.; G?stafsd?ttir, S. 
M.; ?stman, A.; Landegren, U. Nature Biotechnol. 2002, 20, 473-477. 
(17) Heyduk, E.; Heyduk, T. Anal. Chem. 2005, 77, 1147-1156. 
(18) Zhang, D. Y. Thesis: Dynamic DNA strand displacement circuits, California 
Institute of Technology, 2010. 
(19) Bau, L.; Tecilla, P.; Mancin, F. Nanoscale 2011, 3, 121-133. 
(20) Dey, D.; Goswami, T. J. Biomed Biotechnol. 2011, 348218, 7. 
(21) Grieshaber, D.; MacKenzie, R.; Voros, J.; Reimhult, E. Sensors 2008, 8, 1400-
1458. 
(22) Kimmel, D. W.; LeBlanc, G.; Meschievitz, M. E.; Cliffel, D. E. Anal. 
(23) Lu, L.-M.; Zhang, X.-B.; Kong, R.-M.; Yang, B.; Tan, W. J. Am. Chem. Soc. 
2011, 133, 11686-11691. 
(24) Reches, M.; Mirica, K. A.; Dasgupta, R.; Dickey, M. D.; Butte, M. J.; Whitesides, 
G. M. ACS Appl. Mater. Inter. 2010, 2, 1722-1728. 
(25) Shang, J.; Cheng, F.; Dubey, M.; Kaplan, J. M.; Rawal, M.; Jiang, X.; Newburg, 
D. S.; Sullivan, P. A.; Andrade, R. B.; Ratner, D. M. Langmuir 2012, 28, 3338-3344. 
(26) Yang, Y.-C.; Lu, H.-H.; Wang, W.-T.; Liau, I. Anal. Chem. 2011, 83, 8267-8272. 
(27) Pemberton, R. M.; Hart, J. P.; Foulkes, J. A. Electrochimica Acta 1998, 43, 3567-
3574. 
(28) Velasco-Garcia, M. N.; Mottram, T. Trends in Biotechnology 2001, 19, 433. 
(29) Aytur, T.; Foley, J.; Anwar, M.; Boser, B.; Harris, E.; Beatty, P. R. Journal of 
Immunological Methods 2006, 314, 21-29. 
24 
 
 
(30) Chaubey, A.; Malhotra, B. D. Biosens. Bioelectron. 2002, 17, 441-456. 
(31) Blanco-Lopez, M. C.; Lobo-Catanon, M. J.; Miranda-Ordieres, A. J.; Tunon-
Blanco, P. Trends Anal. Chem. 2004, 23, 36-48.  
(32) Eggins, B. Analytical Techniques in the Sciences. John Wiley & Sons, West 
Sussex, 2002. 
(33) Chaubey, A.; Malhotra, B. D. Biosensors & Bioelectronics 2002, 17, 441?456. 
(34) D?Orazio, P. Clinica Chimica Acta 2003, 334, 41?69. 
(35) Ispas, C. R.; crivat, g.; Andreescu, S. Anal. Lett. 2012, 45, 168-186. 
(36) Gullberg, M.; et al. Proc. Natl. Acad. Sci. USA 2004, 101, 8420-8424. 
(37) Heyduk, E.; Dummit, B.; Chang, Y. H.; Heyduk, T. Anal. Chem. 2008, 80, 5152-
5159. 
(38) Kim, D.; Daniel, W. L.; Mirkin, C. A. Anal. Chem. 2009, 81, 9183-9187. 
(39) Konry, T.; Hayman, R. B.; Walt, D. R. Anal. Chem. 2009, 81, 5777-5782. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25 
 
 
 
CHAPTER 2  
Dissociation Constants (Kd) Measurement of DNA Aptamers 
 
2.1     Introduction 
Aptamers are RNA or single stranded DNA molecules that have been isolated 
from random-sequence nucleic acid libraries by in vitro selection 1, 2 and which fold into 
unique tertiary structures, allowing them to bind targets with high affinity and specificity.  
As such, aptamers have attracted attention for their potential as pharmaceutical lead 
compounds or diagnostic agents.  To date, numerous aptamers have been selected against 
a broad range of targets, such as small molecules 3, 4, proteins 5, 6, and even entire cells 7, 8.  
Due to their similar affinity and selectivity, aptamers draw obvious comparisons with 
antibodies 9, yet aptamers exhibit a number of advantages over antibodies as molecular 
recognition elements, especially for diagnosis.  First, the relative ease of in vitro aptamer 
isolation eliminates the need for animal hosts, chemical modification of nucleic acids is 
highly flexible and well-developed, and aptamer binding affinity can be tailored or 
manipulated.  Furthermore, aptamers can be selected against virtually any molecule, even 
those that are toxic or do not stimulate an immune response.  Lastly, aptamers will refold 
to regain functionality after exposure to denaturing conditions such as extreme pH or 
26 
 
 
temperature.  Due to these advantages over antibodies, aptamers have now infiltrated 
much of analytical chemistry 4. 
DNA aptamers against thrombin, human IgE, and ATP were among the first 
selected, thus they have been thoroughly characterized 10-12.  Commonly used thrombin 
aptamers include two G-quartet conformations that selectively bind to ? -thrombin: a 29-
nucleotide ligand, 60-18(29) (a modified version is named ?THRaptA? herein) 10, and a 
15-nucleotide aptamer, known as G15D (?THRaptB? herein) 10.  For human IgE, often 
used is a 37-nucleotide aptamer, D17.4 (?IgEapt? herein,) 12, and a commonly used ATP 
aptamer is DH25.42 (?ATPapt? herein) 13.  In the original reports, these aptamers were 
isolated from pools of random-sequence, single-stranded DNA molecules by in vitro 
selection, and all have been characterized to bind their targets in solution with 
dissociation constants in the nanomolar or micromolar ranges.  
Dissociation constant (Kd) measurements of aptamers against their target 
molecules have been achieved previously using techniques such as filter binding 10, 12, 14, 
15, fiber-optic microarrays 16, affinity capillary electrophoresis (ACE) 11, capillary 
electrophoresis (CE) 17, 18, nonequilibrium capillary electrophoresis of equilibrium 
mixtures (NECEEM) 19, micro free flow electrophoresis (?FFE) 20, microchip CE 21, 
optical biosensors using fluorescence anisotropy 22, and affinity chromatography 13, 23.  A 
common characteristic of many of these approaches is the requirement of a separation 
step, namely rinsing or washing.  For example, washing of unbound DNA and elution of 
DNA complexes is required during selection cycles in filter binding, washing of beads is 
needed during DNA immobilization in fiber-optic microarrays, and rinsing steps are 
needed during aptamer coupling in the optical biosensor method.  The approaches that do 
27 
 
 
not require a washing step prior to measurement (ACE, CE, NECEEM, ?FFE) are 
advantageous due to their low reagent consumption, but these methods are not applicable 
to measuring dissociation constants of aptamer-small molecule binding.  Measurable 
binding with these methods generally requires a large shift in size or charge-to-size ratio.  
Thus, there is a need for a method that combines low reagent consumption, homogeneous 
measurement (no separation or washing required), and the ability to measure binding of 
small and large molecules. 
In this chapter, we describe two approaches based on automated microchip 
electrophoresis for aptamer-target effective dissociation constant (Kd,eff) measurement.  
These approaches, which allow Kd,eff measurements on aptamers against large or small 
molecules, are both based on a microchip version of the DNA mobility shift assay.  The 
methods offer typical advantages of the microfluidic platform, including minimized 
reagent use, reduced cost, and reduced analysis time.  Additionally, there is no need for 
labor-intensive, tedious washing steps.  The first approach applies to large molecule 
(protein) targets, where non-denaturing microchip gel electrophoresis separations of 
DNA could be used to resolve unbound from target-bound aptamers.  This approach is 
similar to previously reported CE methods 17, 18, yet requires even less starting material 
for measurement.  The second approach adds a level of flexibility that is novel to this 
work, in which an aptamer/competitor strategy was used to extend the technique to small 
molecules targets such as ATP.  Finally, we provide evidence that our approach could be 
used during aptamer selection (e.g. SELEX) by accurately measuring aptamer Kd,eff in the 
presence of a large background of randomized DNA sequences. 
 
28 
 
 
 
 
2.2 Reagents and Experimental Methods 
2.2.1 Reagents and Materials 
All solutions were prepared with deionized, ultra filtered water (Fisher Scientific). 
Human thrombin, IgE, and ATP were obtained from Sigma Aldrich. All oligonucleotides 
were obtained from Integrated DNA Technologies (IDT; Coralville, Iowa), with purity 
and yield confirmed by mass spectrometry and HPLC, respectively. Nucleotide 
sequences of DNA aptamers used in experiments are given in Table 2-2. Buffer reagents 
were purchased from VWR. The binding buffer used for THRaptA-thrombin or 
THRaptB-thrombin incubations consisted of 50 mM Tris-HCl, 100 mM NaCl, and 1 mM 
MgCl2 at pH 7.5; the IgE aptamer buffer consisted of 1x PBS (138 mM NaCl, 2.7 mM 
KCl, 8.1 mM Na2HPO4, 1.1 mM KH2PO4, at pH 7.4) with 1 mM MgCl2; and the ATP 
aptamer binding buffer consisted of 10 mM Tris?HCl, 150 mM NaCl, and 5 mM MgCl2 
at pH 7.5. 
29 
 
 
 
2.2.2 Automated Microchip Electrophoresis 
In order to form the aptamers? functional secondary and tertiary structure, solutions 
of THRaptA, THRaptB, and IgEapt were heated to 95 ?C , then cooled slowly to room 
temperature (22 ?C ) at speed of ?3 ?C  min-1 using a Mastercycler-EP gradient thermal 
cycler (Eppendorf). Microchip electrophoresis was performed on an Experion automated 
electrophoresis system (Bio-Rad). According to the manufacturer?s protocol, the 
microchip used in the experiment was an Experion DNA 1K chip, in which 11 sample 
wells and 1 DNA sizing ladder well are included. Titration sets (11 points) were carried 
out in triplicate. Fixed amounts of THRaptA, THRaptB, or IgEapt strands were incubated 
 
Table 2-1. Oligonucleotide sequences used for Kd,eff measurements of aptamer binding 
to large or small molecules by automated microchip electrophoresis.    
 
 
30 
 
 
with variable amounts of thrombin or IgE for 2 h at room temperature, in a total volume 
of 5 mL, prior to loading into the chip. 1 mL aliquots of these reaction mixtures 
(samples), or standards, or ladder were loaded into each sample well, and the 
manufacturer?s protocol was followed for injection, separation, and fluorescence-based 
detection using DNA intercalating dyes. The overall analysis time was ~30 min for 
separation of all 11 samples and the sizing ladder, inferring that each separation 
(including heating and transition times) required approximately 2.5 min. The intercalating 
dye used in the Experion system binds to both double- and single-stranded DNA, with 
approximately 5-fold higher sensitivity toward double-stranded DNA. Since aptamer 
structure is comprised of both double- and single stranded DNA segments, these dyes are 
appropriate for detection of folded aptamers in solution. As discussed in detail below, the 
presence of intercalating dyes may affect aptamer?target binding affinities to some 
extent. Note that the Experion DNA 1K loading buffer is of low ionic strength, but its 
composition and the intercalating dye identity remain proprietary. 
2.2.3 Dissociation Constant Determination 
The effective dissociation constants (Kd,eff) were determined using nonlinear least-
squares fitting to the Hill equation (eqn (1)), with the Hill coefficients (n) set as variable. 
,
[]( - )
( ) [ ]
n
i f i nn
d e ff
TS S S S
KT?? ?                             (1) 
Where S is the fluorescence signal measured as a function of [T], the titrant concentration 
(target is typically used as the titrant); Si and Sf are the initial and final signals, 
respectively; and Kd,eff represents the effective dissociation constant of the complex of 
interest. Intensities of peaks of interest in microchip electropherograms were fit to the 
31 
 
 
Hill equation after titration experiments. This simple model, which has found widespread 
use in biochemistry, physiology, and pharmacology to analyze binding equilibria,53-55 
provides flexibility in measuring systems of unknown stoichiometry. Without 
information on stoichiometry, this approach results in an effective dissociation constant 
(Kd,eff) measurement rather than a strict dissociation constant (Kd) with known 
stoichiometry. Kd,eff measurement is proper in the current work, since in most cases little 
is known about the stoichiometry of aptamer-target interactions. For large molecule?
aptamer Kd,eff measurements (Figure 2-1A), we chose to measure the decrease in the free 
aptamer peak.11,17,20 In some cases, it should be possible to simplify the Kd,eff 
measurements using the ratios of bound and free aptamers to determine the average Kd,eff 
value.21 However, this methodology would be less flexible than using the free aptamer 
peak, since it would only apply to special cases where the complex peak could be 
detected. Furthermore, the response of aptamer-target complexes in our system will vary 
greatly depending on the extent to which target displaces DNA intercalating dyes. For 
small molecule-aptamer Kd,eff measurements, an aptamer-competitor scheme (Figure 2-6B 
and C) was used to extract binding constants, as discussed in detail below. 
 
 
 
 
 
 
 
32 
 
 
 
 
 
 
 
 
Figure 2-1:   Schematic representations of aptamer?target binding equilibria 
during Kd,eff measurements. (A) Protein?aptamer Kd,eff measurements were 
straightforward, where the decrease in the amount of free aptamer could be 
followed with microchip electrophoresis as the concentration of target (T) was 
varied. Small molecule aptamer Kd,eff measurements relied upon (B) aptamer?
competitor equilibria in the absence of target as well as (C) the complete 
system in the presence of target. 
 
33 
 
 
2.3 Validation of Method by Measurement of Aptamer-Protein Dissociation 
Constants 
As depicted by the first scheme in Figure 2-1A, aptamer-protein dissociation 
constants (Kd,eff) were measured using the microchip ssDNA mobility shift assay. Upon 
binding to a protein or large molecule target (T), the electrophoretic mobility of the 
bound fraction of the aptamer was altered, and the decreased height or area of the free 
aptamer peak (or increase in the complex peak) was monitored as a function of target 
concentration to obtain a binding curve. Since we utilize automated microchip gel 
electrophoresis rather than zone electrophoresis as the separation mechanism, it is 
necessary to validate that this approach can be used to obtain aptamer-protein 
dissociation constants that are comparable to previous reports. The microchip 
electropherograms of THRaptA complexed with thrombin at different ratios are shown in 
Figure 2-2A. The concentration of THRaptA was fixed at 300 nM while thrombin was 
varied from 0 to 2.88 mM in separate 5 mL incubation mixtures. 1 mL aliquots of each 
mixture were loaded into the 11 sample wells of the electrophoretic microchip along with 
5 mL of the Experion DNA 1K loading buffer (Bio-Rad), lowering the concentrations of 
all components by 6-fold (THRaptA fixed at 50 nM; thrombin varied from 0 to 480 nM). 
By monitoring the THRaptA peak at ~50 s (filled blue peak in Figure 2-2A), a consistent 
trend was observed in which the peak intensity decreased as the thrombin concentration 
in the incubation mixture was increased. This decrease was a result of the mobility shift 
of the ssDNA aptamer as it formed a complex with thrombin.11,20 The peaks at ~39 s 
corresponded to a 15-base pair internal standard for DNA sizing and concentration 
measurements. The peak intensities at ~50 s were used to generate a binding curve as a  
34 
 
 
 
 
function of concentration of thrombin (Figure 2-2B; solid line, filled triangles), with error 
bars representing standard deviations of triplicate measurements (3 microchips). Using 
nonlinear least squares fitting of this data to the Hill equation, we determined the Kd,eff of 
THRaptA binding to thrombin to be 25.6 ?  2.9 nM at 30 ?C  (running temperature of 
the microchip electrophoresis system). This value is higher than previous measurements, 
likely due to the increased temperature or other effects such as aptamer sequence 
 
  
 
Figure 2-2:  ssDNA mobility shift assay for protein?aptamer Kd,eff 
measurements. (A) Example microchip electropherograms of THRaptA 
during titration with thrombin. The THRaptA peak around 50 s was 
observed to decrease proportionally as thrombin formed complexes with a 
fraction of the aptamer pool. (B) Binding curves generated from 
electropherograms for aptamer?target binding of THRaptA (solid line, filled 
triangles), THRaptB (dashed line, filled gray squares), and IgEapt (dash-
dotted line, open circles) against their respective targets. Nonlinear least 
squares analyses revealed Kd,eff values of 25.6 ? 2.9, 128 ? 27, and 112 ? 7 
nM, respectively. 
 
35 
 
 
differences (Table 2-2; discussion below). Essentially the same Kd,eff value, albeit with 
larger error, was obtained when using the presumed aptamer-protein complex peak at 107 
s (data not shown). It is noteworthy that only 28.8 and 16.5 pmol of thrombin and 
thrombin aptamer, respectively, were required for a single-point Kd,eff  measurement (one 
microchip, 11 sample wells), which represents a major advantage of this microchip 
electrophoresis based approach (Table 2-2). If duplicate or triplicate runs are desired, 
these amounts must be doubled or tripled. 
For further validation, this method was used to measure the Kd,eff values of THRaptB 
and IgEapt against their respective targets. The fixed concentrations of THRaptB and 
IgEapt in the final dilution on-chip were 200 and 50 nM in these sets of experiments, 
respectively. As shown by the binding curves in Figure 2-7B, the Kd,eff values of 
THRaptB binding thrombin (dashed line, filled gray squares) and IgEapt binding to IgE 
(dash-dot line, open circles) were 128 ?  27 and 112 ?  7 nM, respectively (triplicate 
incubations, 3 microchips). The THRaptB Kd,eff measurement was in good agreement 
with previous reports, which vary by about an order of magnitude depending on the 
methodology (Table 2-2). The IgEapt Kd,eff measurement was slightly higher than 
previous reports, again likely due to the increased temperature of measurement (30 ?C ; 
discussed in detail below). The total amounts of thrombin, THRaptB, IgE, and IgEapt 
required for single-point versions of these measurements were 239, 66.0, 28.8, and 16.5 
pmol, respectively. 
 
 
 
36 
 
 
 
 
 
2.4 Measurement of Aptamer-Small Molecule Dissociation Constants 
With small molecule targets such as ATP, the charge-to-size ratios?and hence the 
electrophoretic migration times?of target-bound aptamers are not significantly altered 
compared to that of the unbound aptamers, thus the peaks cannot be distinguished. This 
problem limits the flexibility of our approach or any other electrophoretic approach, as 
well as for fluorescence anisotropy-based methods. To remedy the issue, an 
aptamer/competitor strategy was utilized, in which a DNA competitor complementary to 
 
Table 2-2.   Method comparisons for Kd,eff  measurements of the four aptamers analyzed 
in this work. The microchip electrophoresis approach is the only one listed that can 
interrogate aptamer?small molecule binding without requiring a washing step; the 
amount of consumedDNAreagent is also small for this approach 
 
  
37 
 
 
the aptamer could be displaced by ATP (Figure 2-1B and C). The equilibrium between 
ATPapt (thick bold line) and competitor (thinner line) is depicted in Figure 2-1B, where a 
competitor sequence with a designed 7-base-pair hairpin loop possesses a 15-base region 
that is complementary to ATPapt. This equilibrium between ATPapt and competitor is 
defined by the dissociation constant, ,1ATPdeffK . In Figure 2-1C, this aptamer?competitor 
equilibrium is combined with the ATPapt binding to its target, ATP. As small molecules 
of ATP displace the competitor from ATPapt, the competitor hairpin reanneals, 
promoting an equilibrium shift to the left where ATP is then bound to its aptamer, 
ATPapt. This equilibrium is defined by dissociation constant, ,2ATPdeffK . Note also that an 
additional region was added in Figure 2-1C, namely a 26-nucleotide region 
complementary to the competitor, C26 (gray bold line) for the purpose of enhancing 
separation. For measurement of ,1ATPdeffK  (Figure 2-6B), solutions of competitor only were 
heated to 95 ?C , then cooled slowly to room temperature (22 ?C ) at ?3 ?C  min-1 using 
a thermal cycler. Fixed amounts of competitor were incubated with variable amounts of 
ATPapt for 2 h at room temperature prior to loading into the chip. To measure ,2ATPdeffK  
(Figure 2-1C), ATPapt?competitor?C26 complex was heated to 95 ?C , then cooled 
slowly to room temperature (22 ?C ) at ?3 ?C  min-1. The ratio of ATPapt : competitor : 
C26 was maintained at 1 : 1 : 1. Fixed amounts of ATPapt?competitor?C26 complex (0.5 
mM) were incubated with variable amounts of ATP for 2 h at room temperature prior to 
loading into the chip. 
From Figure 2-1B, we define the effective dissociation constant of aptamer?
competitor complex as: 
38 
 
 
,1
[ c o m p e t i t o r ] [ A T P a p t ][ A T P a p t c o m p e t i t o r ]AT P
d e ffK ? ?
                                        (2) 
Now, we define the effective dissociation constant of the competitive equilibrium 
shown in Figure 2-1C, which is a unitless constant: 
26
,2 26
[ c o m p e t i t o r C ] [ A T P a p t A T P ]
[ A T P a p t c o m p e t i t o r C ] [ A T P ]AT Pd e ffK
???
??
            (3) 
,2ATPdeffK  can also be defined as below, since the C26 strand is stably complexed with 
competitor and does not participate significantly in this equilibrium: 
 
,2
[com p et i t o r][ A T P apt A T P ]
[A T P apt compet i t o r][ A T P ]AT Pd ef fK
??
?
                         (4) 
It is from these two measurable equilibria that we can extract the effective 
dissociation constant of interest, Kd,ATPapt: 
,1,
,2
[A T P ap t ][ A T P ]
[A T P ap t A T P ]
A T Pd eff
d A T P a p t A T P
d eff
KK
K???
                               (5) 
Thus, we can separately measure ,1ATPdeffK  and ,2ATPdeffK  using microchip DNA mobility 
shift assays (Figure 2-3A and B) with the systems depicted in Figure 2-1B and C, then we 
can extract Kd,ATPapt from this data using eqn (5). In this way, from binding curves (Figure 
2-3C), we determined ,1ATPdeffK  and ,2ATPdeffK  to be 16.1 ?  1.9 nM and 2.27 ?  0.05 ?  10-4, 
respectively. Using eqn (5), Kd,ATPapt was then determined to be 70.9 ?  8.5 mM at 
39 
 
 
30 ?C . This extracted value also agrees well with the published value of ~50 mM,13,23 
especially considering the increased temperature (Table 2-2). 
 
 
 
 
 
 
 
Figure 2-3: ssDNA mobility shift 
assay for protein?small molecule 
Kd,eff measurements. (A) Microchip 
electropherograms of the system 
shown in Fig. 2-5B, as competitor 
is titrated with ATPapt. The 
competitor peak (43 s, filled gray 
peaks) was observed to decrease 
in height as it hybridized with 
ATPapt, shifting the equilibrium 
shown in Fig. 2-5B to the right. (B) 
Electropherograms of the system 
in Fig. 2-5C during titration with 
ATP. The ATPapt?competitor?C26 peak (~45 s, filled gray peaks) 
decreased in height as ATP complexed with ATPapt, shifting the 
equilibrium in Fig. 2-5C to the left. (C) Binding curves generated from 
decreases in the competitor peak (filled circles) and the ATPapt?
competitor?C26 peak (filled triangles)  gave           =16.1 ?  1.88  nM   and         
,2ATPdeffK  = 2.27 ?  0.05 ? 10
4, respectively. By eqn (5), K
d,ATPapt = 70.9 ?  
8.5 ?M at 30 ?C . 
,1ATPdeffK
40 
 
 
It is feasible that Kd,eff measurements can be made in this way for aptamers against 
any small molecule target, greatly enhancing the flexibility of our automated microchip 
electrophoresis approach for aptamer?target Kd,eff measurements. In doing so, one must 
first develop a similar aptamer?competitor strategy for each aptamer to be assayed, 
inferring an increased experimental commitment. As long as the aptamer sequences are 
known, it should be relatively straightforward to design similar hairpin-forming 
competitors for other small-molecule binding aptamers, but the energetics of competitor?
aptamer binding will need to be optimized empirically. 
 
2.5 Method and Dissociation Constant Comparisons 
The measured Kd,eff values of THRaptA, THRaptB, IgEapt, and ATPapt were 25.6 
?  2.9 nM, 128 ?  27.0 nM, 112 ?  7 nM, and 70.9 ?  8.5 mM, respectively (Table 2-2). 
These values agreed reasonably well with previously published numbers, especially 
considering the differences in temperature and solution conditions. The measured Kd,eff  
values for THRaptB (128 ?  27.0 nM) and ATPapt (70.9 ?  8.5 mM) were in agreement 
with the published values of 50?450 nM21,22,25-27,30 and ~50 mM,24,34 respectively. The 
measured dissociation constants for both THRaptA (25.6 ?  2.9 nM) and IgEapt (112 ?  
7 nM) were higher than the published values of 0.5 ? 1 nM21 and 8 ? 48 nM.23,28,31,32 
There are several possibilities that could cause this discrepancy. Most importantly, the 
running temperature in our experimental setup is fixed at 30 ?C , which is higher than that 
used in the other methods. A temperature increase can cause a significant increase in 
measured Kd value, with the magnitude of the shift dependent on the thermodynamics of 
binding for each system. Second, the salt concentrations and buffer constituents are 
41 
 
 
different among measurements, which can also significantly affect the binding constants. 
Third, the nucleotide sequences of THRaptA and IgEapt (as well as THRaptB) used in 
this paper were slightly different than those used in other measurements. Our sequences 
(Table 2-1) include the consensus sequences from selection (bold text), but also include 
additional sections at the 5? and 3? ends for use in proximity ligation assays.55,57 With the 
nitrocellulose filter binding method,21,23,25,26 the authors used only the consensus 
sequences of THRaptA and IgEapt, shown in bold in Table 2-1. German et al.22 used 
different versions of the IgE aptamer, namely the unlabeled aptamer (bold sequences in 
Table 2-1) and a labeled aptamer, with a fluorescent label at the 5?-end using an ethylene 
glycol linker. Their dissociation constants were determined to be 64 nM for labeled 
aptamer and 8 nM for unlabeled aptamer. It is obvious that the added fluorescent label 
caused a significant decrease in binding affinity. Likewise, our extended sequences at the 
3? and 5? ends of aptamers may have affected the binding with target molecules. Finally, 
since DNA intercalating dyes were used to label the free aptamer in this work, these dyes 
may have competed to some extent with target binding, increasing the measured Kd,eff. It 
is noteworthy that order of magnitude differences are often observed when comparing Kd 
measurements reported in the literature (e.g. ?THRaptB? column in Table 2-2), due to the 
lack of consistency in buffer conditions, temperature, labeling, etc. Nonetheless, our 
aptamer affinity measurements were in reasonably good agreement with previous reports, 
and our approach is greatly simplified compared to others. 
Perhaps equally important is the fact that the microchip electrophoresis-based 
approach requires much less time and reagent consumption than other methods (Table 2-
2). As shown in Table 2-2, our approach provides a 2- to 6-fold reduction in analysis time 
42 
 
 
compared to all other methods except the ?FFE method,31 which provides ~2-fold faster 
analysis than our approach. In this comparison, ?analysis time? refers to the time required 
for completion of washing and measurement steps, but excluding preparation and 
incubation times for titrations, which should be essentially equal among the methods. The 
automated microfluidic system also requires only 1 mL sample volumes in each of 11 
sample wells, allowing 5 mL incubation volumes for our experiments. In order to provide 
equal comparisons, the time and amount estimates shown in Table 2-2 were generated by 
assuming an equal number of points of titration (11 used here) and scaling the values 
reported for individual incubations. By comparison to all but the filter binding and ?FFE 
methods, our approach provides a 10- to 6000-fold reduction in DNA consumption, with 
only 17?170 pmol of DNA used for 11 of the 5 mL incubations. Although less DNA was 
required, the filter binding approach requires washing steps and specialized equipment 
for dealing with radioactive reagents. Specialized equipment and training will also be 
required for the ?FFE method,31 since the method is still in the development stage. On 
the other hand, this method is promising and would likely find much success if made 
widely available. A significant advantage of our approach is the commercial availability 
and ease of use of the automated instrumentation. Furthermore, the automated and 
homogeneous nature of our method reduces instances of human error and avoids 
repetitive washing steps. 
As a final point of comparison, our approach provides an added level of flexibility in 
comparison to other homogeneous methods based on electrophoresis22,28,31 or 
fluorescence anisotropy.33 The scheme developed for measurement of aptamer?small 
molecule dissociation constants (Figure 2-1B and C and 2-3) is a novel extension of 
43 
 
 
electrophoretic affinity measurements which allows the flexibility of our method to rival 
filter binding and affinity chromatography. In fact, our approach is the only one listed in 
Table 2-2 that can interrogate aptamer?small molecule binding without requiring a 
washing step. By exploiting the inherent nature of DNA aptamers to coexist in their 
target binding form (with tertiary structure) and in their duplex form with complementary 
sequences (e.g. antidotes for aptamer-based drugs), a competitive equilibrium was 
developed (Figure 2-1B and C). This system imparts large, measurable electrophoretic 
mobility shifts that are triggered by aptamers binding to small molecules (Figure 2-3). 
The approach combines the power of DNA-based computing or structure-switching with 
aptamer?target interactions to significantly improve the flexibility of our mobility shift 
assays. Moreover, the use of DNA-based competitors could also be applied to the other 
homogeneous methods shown in Table 2-2, such as ?FFE, ACE, NECEEM, or to optical 
biosensors to improve their flexibility of application. 
 
2.6 Feasibility for Kd Monitoring during Aptamer Selection 
To provide proof-of-principle that the microchip DNA mobility shift assays could be 
useful for monitoring dissociation constants during aptamer selection processes (e.g. 
SELEX5,6), a mock-SELEX process was carried out. A known DNA aptamer was spiked 
into a random sequence DNA library at increasing ratios to mimic enrichment within the 
library, and our approach was used to measure the bulk dissociation constant of these 
mixtures. Solutions of increasing percentages of THRaptA in a random DNA library 
were prepared (0, 1, 5, 10, 25, 75, 90, and 100%). Fixed total DNA concentrations of 
each THRaptA/library mixture were then incubated in 5 mL volumes (300 nM total 
44 
 
 
DNA, or 1.5 pmol) with variable amounts of thrombin (11 thrombin concentrations and 1 
microchip for each mixture) for 2 h at room temperature prior to loading into the 
microchip, injection, and separation. Since the DNA library and THRaptA were of 
approximately equal length (100 and 97 bp, Table 2-1), only one peak in the 
electropherogram was monitored. The THRaptA/library peak was observed to decrease 
with the addition of thrombin for all samples except the 0% THRaptA control (Figure 2-
4A). Data shown in this figure are peak heights of free THRaptA/library pool, normalized 
to the peak height with no thrombin added. The shape of the binding curve became more 
pronounced with increasing percentage of THRaptA in the pool. As shown in Figure 2-
4B, nonlinear least squares analysis revealed that the known Kd,eff of THRaptA could be 
extracted from this data at concentrations as low as 5% of the pool (75 fmol THRaptA + 
1425 fmol library in each incubation tube). Error bars in this data represent fitting errors 
in the least squares analysis. The Kd,eff  of the 1% mixture was determined to be 24.6 ?  
70.2 nM. Although this value was coincidentally near the true value of 25.6 nM, with 
such large error, a one-parameter t-test showed that it was significantly different from the 
true value of 25.6 nM (p < 0.05). In the same way, mixtures with 5% THRaptA or more 
were shown to be statistically equal to the true value (p > 0.2). In essence, the measured 
Kd,eff of the pools converged on the true value of the THRaptA Kd,eff  as the fraction of 
THRaptA in the library pool increased (Figure 2-4B). In a true SELEX process, the bulk 
Kd,eff  should decrease with each round as higher affinity ligands succeed in the 
evolutionary competition, rather than continuously converging toward a single value. 
Nonetheless, these results suggest that our approach could be used as an automated, 
homogeneous route for monitoring Kd,eff  values during each round of the SELEX 
45 
 
 
process, requiring only ~825 fmol (11 incubation tubes with 75 fmol each) of binding 
sequences (THRaptA) in a background of random DNA. 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2-4?   Mock-SELEX process in which THRaptA was spiked into a 
random sequence DNA pool at different concentrations. (A) Microchip 
electrophoresis binding curves for pools from 5 to 90% THRaptA. (B) 
Measured Kd,eff values of the pool converged to the true value with 
increasing THRaptA fraction, and our system was able to detect 
aptamers spiked into a random sequence pool at as low as 5% 
population (75 fmol THRaptA). A one sample t-test showed no difference 
from the previously measured value of 25.6 nM (p > 0.2) for all samples 
other than the 1% sample (p < 0.05). 
 
46 
 
 
2.7 Conclusions 
In this work, we have demonstrated that automated microchip electrophoresis is a 
useful method for measurement of aptamer?target Kd,eff  measurements. The approach is 
homogeneous, requires minimal time and reagent consumption, and can be applied to 
aptamer?protein or aptamer?small molecule binding. To achieve this flexibility, an 
aptamer?competitor strategy, akin to DNA-based logic operations, was used to extract 
Kd,eff  values for aptamer?small molecule interactions. This aspect should be applicable to 
any electrophoretic affinity assays22,28,31 or fluorescence anisotropy assays33 developed to 
date, which should greatly improve the flexibility of the assays. The approach was also 
proven feasible for Kd,eff  monitoring of DNA pools during SELEX processes. For a 
typical SELEX process, only one microchip electrophoresis run can be carried out 
between selection rounds to preserve precious material and time. This would still provide 
11 points on the titration curve to extract binding constants, while only consuming ~825 
fmol of DNA between rounds. In this respect, the advantage provided by the microfluidic 
platform in reducing reagent consumption should be particularly useful. 
 
 
 
47 
 
 
 
2.8 References 
 
(1)      Jones, S.; et al. Nucl. Acids Res. 2001, 29, 943-954. 
(2)      Jones, S.; et al. J.  Mol. Biol. 1999, 287, 877-896. 
(3)      Morozova, N.; et al. BIoinformatics 2006, 22, 2746-2752. 
(4)      Nagai, K. Curr. Opin. Struct. Biol. 1996, 6, 53-61. 
(5)      Ellington, A. D.; Szostak, J. W. Nature 1990, 346, 818-822. 
(6)      Tuerk, C.; Gold, L. Science 1990, 249, 505-510. 
(7)      Osborne, S. E.; et al. Curr. Opin. Chem. Biol. 1997, 1, 5-9. 
(8)      Famulok, M.; et al. Chem. Rev. 2007, 107, 3715-3743. 
(9)      Navani, N. K.; Li, Y. F. Curr. Opin. Chem. Biol. 2006, 10, 272-281. 
(10) Jiang, Y. X.; et al. Anal. Chem. 2003, 75, 2112-2116. 
(11) Griffin, L. C.; et al. Gene 1993, 137, 25-31. 
(12) Ferreira, C. S.; et al. Anal. Bioanal. Chem. 2008, 390, 1039-1050. 
(13) Balogh, Z.; et al. FASEB 2010, 24, 4187-4195. 
(14) Shangguan, D.; Li, Y.; Tang, Z.; Cao, Z. C.; Chen, H. W.; Mallikaratchy, P.;      
Sefah, K.;       Yang, C. J.; Tan, W. PNAS 2006, 103, 11838-11843. 
(15) Li, J. W.; et al. Biochem. Biophys. Res. Commun. 2002, 292, 31-40. 
(16) Gutsaeva, D. R.; et al. Blood  2011, 117, 727-735. 
(17) Lu, C.; et al. Cancer Biol. Ther. 2010, 9, 176-182. 
(18) Wu, Y.; Sefah, K.; Liu, H.; Tan, W. PNAS 2010, 107, 5-10. 
(19) Huang, Y. F.; et al. Langmuir 2008, 24, 11860-11865. 
(20) Chu, T. C.; et al. Nucl. Acids Res. 2006, 34, e73. 
(21) Tasset, D. M.; Kubik, M. F.; Steiner, W. J.  Mol. Biol. 1997, 272, 688-698. 
(22) German, I.; Buchanan, D. D.; Kennedy, R. T. Anal. Chem. 1998, 79, 4540-4545. 
(23) Wiegand, T. W.; et al. J. Immunol. 1996, 157, 221-230. 
(24) Huizenga, D. E.; Szostak, J. W. Biochemistry 1995, 34, 656-665. 
(25) Bock, L. C.; Griffin, L. C.; Latham, J. A.; Vermaas, E. H.; Toole, J. J. Nature    
1992, 355, 564-566. 
(26) Macaya, R. F.; Waldron, J. A.; Beutel, B. A.; Gao, H.; Joesten, M. E.; Yang, M.; 
Patel, R.; Bertelsen, A. H.; Cook, A. F. Biochemistry 1995, 34, 4478-4492. 
(27) Lee, M.; Walt, D. R. Anal. Biochem. 2000, 282, 142-146. 
(28) Mendonsa, S. D.; Bowser, M. T. Anal. Chem. 2004, 76, 5387-5392. 
(29) Mosing, R. K.; Bowser, M. T. J. Sep. Sci 2007, 30, 1420-1426. 
(30) Berezovski, M.; Nutiu, R.; Li, Y.; Krylov, S. N. Anal. Chem. 2003, 75, 1382-1386. 
(31) Turgeon, R. T.; Fonslow, B. R.; Jing, M.; Bowser, M. T. Anal. Chem. 2010, 82, 
3636-3641. 
(32) Gong, M.; Wehmeyer, K. R.; Limbach, P. A.; Heineman, W. R. Electrophoresis 
2007, 28, 837-842. 
(33) Potyrailo, R. A.; Conrad, R. C.; ellington, A. D.; Hieftje, G. M. Anal. Chem. 1998, 
70, 3407-3412. 
(34) M. Sassanfar; Szostak, J. W. Nature 1993, 364, 550-553. 
48 
 
 
(35) Walt, D. R. Chemical Society Reviews 2010, 39, 38-50. 
(36) Drabovich, A.; Berezovski, M.; Krylov, S. N. J. Am. Chem. Soc 2005, 127, 
11224-11225. 
(37) Berezovski, M.; Drabovich, A.; Krylova, S. M.; Musheev, M.; Okhonin, V.; 
Petrov, A.; Krylov, S. N. J. Am. Chem. Soc 2005, 127, 3165-3171. 
(38) Dolnik, V. Electrophoresis 2008, 29, 143-156. 
(39) Heller, C. Electrophoresis 2001, 22, 629-643. 
(40) Kostal, V.; Arriaga, E. Electrophoresis 2008, 29, 2578-2586. 
(41) Barrolier, V.; Watzke, E.; Gibian, H. Naturforschung 1958, 13B:754. 
(42) Hannig, K. Z. Anal. Chem. 1961, 181, 244-254. 
(43) Roman, M.; Brown, P. Anal. Chem. 1994, 66, 86-94. 
(44) Pruski, Z.; et al. Electrophoresis 1990, 11, 932-986. 
(45) Zeiller, K.; et al. Hoppe-Syler's Z Physiol Chem. 1975, 356, 1225-1244. 
(46) Rodkey, L. Appl. Theor. Electrophor. 1990, 1, 243-247. 
(47) Kessler, R.; Manz, H. Electrophoresis 1990, 11, 979-980. 
(48) Hoffstetter-Kuhn, S.; Wagner, H. Electrophoresis 1990, 11, 451-456. 
(49) Nath, S.; Schutte, H.; Weber, G.; Hustedt, H.; Deckwer, W. Electrophoresis 1990, 
11, 937-941. 
(50) Knisley, K.; Rodkey, L. Electrophoresis 1990, 11, 927-931. 
(51) Clifton, M.; Jouve, N.; de Balmann, H.; Sanchez, V. Electrophoresis 1990, 11, 
913-919. 
(52) Poggel, M.; Melin, T. Electrophoresis 2001, 22, 1008-1015. 
(53) Hill, A. V. Physiology 1910, 40, 190-224. 
(54) Weiss, J. N. FASEB J. 1997, 11, 835-841. 
(55) Kim, J.; Hu, J.; Sollie, R. S.; Easley, C. J. Anal. Chem. 2010, 82, 6976-6982. 
(56) Li, N.; Ho, C. J. Am. Chem. Soc. 2008, 130, 2380-2381. 
(57) Fredriksson, S.; Gullberg, M.; Jarvius, J.; Olsson, C.; Pietras, K.; G?stafsd?ttir, S. 
M.; ?stman, A.; Landegren, U. Nature Biotechnol. 2002, 20, 473-477. 
 
 
49 
 
 
 
CHAPTER 3  
DNA Models for Development of Proximity Assays 
 
3.1     DNA Models for Asymmetric Proximity Ligation Assays 
The proximity ligation assay (PLA) is one of the most sensitive and simple protein 
assays developed to date, yet a major limitation is the relatively narrow dynamic range 
compared to other assays such as enzyme-linked immunosorbent assays. In this work, the 
dynamic range of PLA was improved by 2 orders of magnitude and the sensitivity was 
improved by a factor of 1.57. To accomplish this, asymmetric DNA hybridization was 
used to reduce the probability of target-independent, background ligation. An 
experimental model of the aptamer-target-connector complex (aptA-T-aptB-C20, PLA) in 
PLA was developed to study the effects of asymmetry in aptamer-connector 
hybridization. Connector base pairing was varied from the PLA standard of 20 total bases 
(C20) to an asymmetric combination with 15 total bases (C15). The results of this model 
suggested that weakening the affinity of one side of the connector to one aptamer would 
significantly reduce target-independent ligation (background) without greatly affecting 
target-dependent ligation (signal). 
 
 
50 
 
 
 
3.1.1     Background of Proximity Ligation Assay (PLA) 
The proximity ligation assay1 (PLA) is among the most sensitive protein assays 
developed to date. PLA can be considered a descendent of immuno-PCR methods,2 in 
which antibodies are attached to oligonucleotides to allow the coupling of target binding 
with exponential amplification by polymerase chain reaction (PCR). The term ?proximity 
ligation? was first named in Landegren?s paper,1 the method depends on the simultaneous 
and proximate recognition of target molecules by pairs of affinity probes, giving rise to 
an amplifiable detection signal. In that paper, the molecule they targeted was the 
homodimer of the platelet-derived growth factor B-chain (PDGF-BB). A DNA aptamer 
with affinity for PDGF-BB was extended by additional sequence elements at either the 5? 
or the 3? end, forming a proximity probe pair. When pairs of probes bind PDGF-BB, the 
free ends of their sequence extensions are brought close to hybridize together to a 
subsequently added connector oligonucleotide, allowing the ends to be connected by 
enzymatic DNA ligation. So, the detected protein molecules promote the ligation reaction 
by bringing together the free ends of the proximity probe pairs. The ligation products can 
then be replicated by nucleic acid amplification through PCR (Figure 1-4). 
In the beginning stages of PLA method development, Landegren and co-workers 
determined the optimal concentrations of each proximity probe and the connector 
oligonucleotide that maximize the cooperative effect induced by target binding,1 (Figure 
3-1) and they further proved that this cooperative effect was inversely related to the Kd of 
the proximity probes.3 Major advantages of PLA are sensitivity, selectivity, and ease of 
use. Furthermore, the assay requires no washing steps due to its homogeneous nature. 
51 
 
 
However, PLA has the limitation of a relatively narrow dynamic range. The reason for 
this flaw is the necessity to use very low amounts of proximity probes to minimize a 
target-independent hybridization and ligation (background).  
 
3.1.2     Improvement of Sensitivity and Dynamic Range in PLAs by Asymmetric 
Connector Hybridization 
As shown in Figure 3-2A, the mode of action of PLA results in a four-part complex 
composed of the targeted protein analyte (T), two proximity probes (DNA-conjugated 
antibodies or oligonucleotide aptamers; aptA and aptB), and a connector oligonucleotide 
(C20,PLA). The three separate DNA molecules (or moieties) are designed to hybridize into 
 
 
 
 
Figure 3-1:   Optimization of reaction conditions for the proximity ligation 
assay. (A) Effect of connector oligonucleotide concentration and time of 
ligation on signal and background. (B) Influence of preincubation volume and 
probe concentration. The upper, yellow layers connect the signals recorded in 
reactions with 5.6 attomoles of PDGF-BB, while the lower, red layers 
represent the target-independent background, expressed in arbitrary units. 
Adapted by permission from Macmillan Publishers Ltd: Nature biotechnology (Fredriksson, S.; 
Gullberg, M.; Jarvius, J.; Olsson, C.; Pietras, K.; G?stafsd?ttir, S. M.; ?stman, A.; Landegren, U. 
Nature Biotechnol. 2002, 20, 473-477.), copyright? (2013) 
 
52 
 
 
a structure that promotes a DNA ligase enzyme to covalently couple the two proximity 
probes. In the presence of target, the four-part complex is preferred over target-
independent hybridization (three-part complex) due to a cooperative ?proximity effect? 
that essentially increases the local concentration of the oligo-nucleotide tails.4-6 Thus, the 
amount of ligation products is proportional to the amount of protein analyte, and these 
products can be made recognizable by PCR for exponential amplification and highly 
sensitive quantitation by qPCR. Due to its simplicity and sensitivity, PLA has found 
much interest of late and has been successfully utilized in a variety of applications, such 
as sensitive protein detection,1,3 in situ analysis of protein localization,7 DNA-protein 
interactions,8 clinical diagnostics,9 and protein-protein interaction assays.10 
Our goal in this study was to improve the dynamic range of PLA without sacrificing 
LOD. To address this issue, we made two hypotheses: (1) target-independent 
hybridization (background; Figure 3-2A, bottom) is inversely proportional to the Kd of 
the connector oligonucleotide at a given concentration of proximity probes, and (2) the 
width of the dynamic range is proportional to the amount of a proximity probe to be used. 
If confirmed, these hypotheses suggest that dynamic range and signal to background 
could be improved by simply reducing the number of bases in the connector. To test these 
hypotheses, first we developed and tested an experimental model of the proximity effect. 
This model suggested that asymmetric connectors with higher Kd values could 
significantly decrease target-independent hybridization (background; Figure 3-2A, 
bottom). In turn, this would lead to higher signal-to-background ratios and allow the use 
of more proximity probes, which should increase the dynamic range of PLA. 
 
53 
 
 
 
 
  
 
 
Figure 3-2. Schematics of PLA and model systems described in the text. 
(A) Standard PLA, in which a connector length of 20 bases (C20,PLA, or b 
= 20) is used. Ligated signal or background complexes are not 
differentiated by qPCR. (B) Asymmetric PLA, in which connector lengths 
of 15-19 bases (Cb,PLA, where 15 ? b < 20) are tested. 
 
 
54 
 
 
3.1.3     Reagents and Experimental Methods 
3.1.3.1   Reagents and Materials 
All solutions were prepared with deionized, ultrafiltered water (Fisher Scientific). 
T4 DNA ligase was purchased from New England BioLabs. Human thrombin was 
obtained from Sigma-Aldrich. All oligonucleotides except the Taqman probe were 
obtained from Integrated DNA Technologies (IDT; Coralville, Iowa), with purity and 
yield confirmed by mass spectrometry and HPLC, respectively. Sequences of ssDNA 
strands used in the experimental model are given in Table 3-1. Sequence FreeA was 
labeled at its 5?-end with Iowa black FQ quencher (IABkFQ; absorbance maximum at 
531 nm). Sequence FreeB was labeled at its 5?-end with a phosphate group (Phos) and at 
its 3?-end with 5(6)-carboxyfluorescein (6-FAM, mixed isomer; emission maximum at 
520 nm). Loop strands (Table 3-1) were synthesized by ligation of FreeA and FreeB using 
T4 DNA ligase; Sequences (listed 5? to 3?) for PLA or asymmetric PLA were as follows. 
Thrombin aptamer A, aptA, CAG TCC GTG GTA GGG CAG GTT GGG GTG ACT 
TCG TGG AAC TAT CTA GCG GTG TAC GTG AGT GGG CAT GTA GCA AGA 
GG; thrombin aptamerB, aptB, /5Phos/GT CAT CAT TCG AAT CGT ACT GCA ATC 
GGG TAT TAG GCT AGT GAC TAC TGG TTG GTG AGG TTG GGT AGT CAC 
AAA; symmetric connector, C20,PLA, AAG AAT GAT GAC CCT CTT GCT AAA A; 
asymmetric connector, C18,PLA, TTA TGA TGA CCC TCT TGC TAA AA; asymmetric 
connector, C16,PLA, TAG ATG ACC CTC TTG CTA AAA; forward PCR primer, GTG 
ACT TCG TGG AAC TAT CTA GCG; reverse PCR primer, AAT ACC CGA TTG CAG 
TAC GAT TC; Taqman probe (Applied Biosystems, ABI), 4-chlorofluorescein-TGT 
ACG TGA GTG GGC ATG TAG CAA GAG G-carboxytetramethylrhodamine. 
55 
 
 
 
 
3.1.3.2   Fluorescence Measurements for Model Titrations 
Fluorescence measurements were conducted using a Nanodrop 3300 
fluorospectrometer (Thermo Scientific). The intensity of FAM emission was measured at 
520 nm with LED excitation at 485 nm. The buffer used for titrations was 50 mM Tris-
HCl (pH 7.5), 100 mM NaCl, 1 mM MgCl2. Titrations sets (14-16 points) were carried 
out in duplicate, and each fluorescence measurement was repeated six times. Fixed 
 
Table 3-1.   Oligonucleotide Sequences Used in the Experimental Model of Proximity 
Hybridizationa 
 
 
56 
 
 
amounts of Loop or FreeA and FreeB strands were incubated with variable amounts of Cb 
overnight at room temperature prior to fluorescence measurements. 
 
3.1.3.3   Synthesis of Loop strands 
Loop strands (Table 3-1) were synthesized by ligation of FreeA and FreeB using T4 
DNA ligase. The Linker strand was used to hybridize with and adjacently position the 
strands for ligation. Ligation was carried out by T4 DNA ligase (New England Biolabs) 
in a total volume of 100 ?L for 3 h at 16 ?C with a Linker:FreeA,/B molar ratio of 20:1. 
Ligated products were further concentrated to 20 ?L total volume and mixed with 100?l 
of 1.2x DNA loading dye (0.5mg/ml bromophenol blue, 8M urea, 1% (v/v) NP-40, 1mM 
Tris-HCl pH 8;)11. Then, samples were denatured at 90 ?C for 5 min, and quickly loaded 
onto 1 M urea containing TAE 5% agarose gel. Gel electrophoresis was run on a VWR 
Mini Gel apparatus at 70 volts for 5 h with a 100 nt ssDNA as well as the 50 nt FreeA and 
FreeB strands for markers (Figure 3-3). Under UV light, the ligated product bands were 
excised and eluted using Wizard? SV Gel and PCR Clean-Up Systems (Promega) under 
the manufacturer?s protocol. Ligated products were quantified using both UV absorbance 
at 260 nm and FAM absorbance at 490 nm with a Nanodrop 1000 small-volume 
spectrophotometer (Thermo Scientific). 
 
 
 
 
 
57 
 
 
 
 
 
3.1.3.4   Proximity Ligation Assays 
Temperature control was achieved using a Mastercycler-EP gradient thermal cycler 
(Eppendorf). Thrombin and oligonucleotide solutions were made as described.1 An 
amount of 4 ?L of a reaction mixture containing human thrombin, aptA, and aptB was 
incubated at 37 ?C  for 15 min before adding 1 ?L of connector (C16,PLA, C18,PLA, or 
C20,PLA). An additional 30 min of incubation at 22 ?C  was performed to stabilize 
complexes. These 5 ?L samples were then brought to a total volume of 55 ?L for the 
 
  
 
Figure 3-3: Schematic of Loop strand synthesis by ligation. ssDNA 
separation by agarose gel electrophoresis was used to purify ligation 
products.  
 
58 
 
 
ligation and amplification mixture, which contained 50 mM KCl, 10 mM Tris-HCl, pH 
8.3, 1.9 mM MgCl2, 0.4 Weiss unit T4 DNA ligase, 73 ?M ATP, 0.18 mM dNTPs, 0.45 
?M forward and reverse PCR primers, 45 nM Taqman probe for the 5? nuclease assay, 
and 1.5 units of AmpliTaq Gold polymerase (ABI). The reactions were transferred to a 
real-time PCR instrument (CFX96, Bio-Rad) for temperature cycling: 5 min at 22 ?C  for 
ligation, 10 min at 95 ?C , and then cycling for 15 s at 95 ?C  and 60 s at 60 ?C , repeated 
45 times. Signal-to-background (S/BG) values were calculated from the relative number 
of ligation products detected in a sample with thrombin to that in a sample without 
thrombin. The relative quantity of the ligated products was calculated by using the 
comparative threshold cycle (Ct) method.12 For measurements of PLA background, 15 
pM aptA and 20 pM aptB were used, with different connector concentrations as follows: 
C16,PLA ) 178 nM; C18,PLA ) 112 nM; C20,PLA ) 45 nM. For asymmetric PLA, 900 pM 
aptA and 1200 pM aptB were used with 480 nM C16,PLA; and for symmetric PLA, 15 pM 
aptA and 20 pM aptB were used with 400 nM C16,PLA (same conditions as previously 
reported1). 
 
3.1.3.5   Connector Kd Calculations 
Base-pairing dissociation constants for the invariable segment (Kd,A) and for each 
variable segment (Kd,B) of the connectors were calculated using the well-established, 
thermodynamic nearest-neighbor model.13 Dissociation constants were derived from 
thermodynamic parameters calculated via the BioPHP melting temperature calculator 
(http://insilico.ehu.es/tm.php), a free online service that uses SantaLucia?s method.13 
Source code is freely downloadable at http://www.biophp.org/. Kd,B values were also 
59 
 
 
experimentally confirmed; in Table 3-2, the column labeled ?Kd,B? shows the calculated 
values, and the column labeled ? ,BGdeffK ? shows the experimentally confirmed values. 
 
 
3.1.3.6   DNA Melting Curve Measurement 
The melting temperatures (Tm) of DNA oligomers C16 through C20 (sequences in 
Table 3-1) were determined using a Bio-Rad CFX96 qPCR instrument (Figure 3-4). 
Working solutions consisted of varying concentrations of oligomers and 200 nM of the 
DNA intercalating dye PicoGreen (em ~520 nm; Invitrogen) in 50 mM Tris-HCl (pH 
7.5), 100 mM NaCl, and 1 mM MgCl2. Oligomer concentrations were intentionally 
adjusted to approximately 3 ? Kd,B from Table 3-2 (calculated values). In this way, a 
constant Tm for C16 through C20 inferred accuracy of SantaLucia?s13 thermodynamic 
 
Table 3-2.   Analysis of Connector Sequences Used in the Experimental Model of 
Proximity Hybridizationa 
 
 
60 
 
 
nearest-neighbor modeling approach for estimating Kd,B values. Fluorescence was 
measured as a function of temperature, and the derivative of the fluorescence was fit to 
Gaussian functions using nonlinear least squares fitting to determine Tm values (peak 
center). 
 
 
 
 
 
 
 
 
 
 
Figure 3-4: (A) Fluorescence melting curves of DNA oligomers from 
Table 3-1. (B) Melting temperatures (Tm) were kept essentially constant 
using variable oligomer concentrations of 3 ? Kd,B.  
 
61 
 
 
3.1.4     Signal and Background in Proximity Ligation 
A schematic of the original, aptamer-based PLA1,3 is shown in Figure 3-2A, in the 
context of protein detection with aptamer probes. The quantitative capability of PLA is 
most heavily dependent upon two steps: the cooperative hybridization of the four-part 
complex, aptamer A?target?aptamer B?connector (aptA?T?aptB?C20,PLA), followed by 
the proximity-dependent ligation of the two aptamers using a DNA ligase. This process 
results in an amount of ligated product that is proportional to the original amount of 
protein analyte (?signal? ligations), albeit with some analyte-independent ligation 
products present (?background? ligations). Although qPCR does not differentiate signal 
from background ligations, under optimized conditions1,3 the signal will be dominant 
over background to allow highly sensitive, indirect detection of the protein analyte. 
Signal enhancement over background in PLA is based on the proximity effect. The 
effective concentrations of two aptamers are significantly increased due to their 
proximity, afforded by simultaneous binding to the same protein.4-6 
The chief interference in PLA is target-independent, background ligation (Figure 3-
2A, bottom). This background ligation has been reported to result from two sources.1,3,14 
First, based simply on binding equilibria, a fraction of aptamers will always hybridize 
with the 20-base connector sequences (C20,PLA), even in the absence of protein analyte, 
resulting in target-independent ligations that increase the background levels in the 
assay.1,3 Second, the T4 DNA ligase is capable of ligating a small fraction of ssDNA 
sequences, even in the absence of a connector sequence.14,15 This will also increase assay 
background. Importantly, the presence of this background limits the amount of usable 
aptamer (or antibody) probes, which affects the sensitivity and partially defines the upper 
62 
 
 
limit of the dynamic range.3 Methods to widen dynamic range would address a major 
weakness of PLA and significantly improve the applicability of the assay. 
 
3.1.5     Design of Experimental Model of Proximity Hybridization 
We hypothesized that by decreasing connector binding affinities for aptamers using 
an asymmetric connector (Figure 3-2B), the amount of background ligation products 
would be greatly reduced compared to signal ligation products. In turn, this should allow 
the use of higher aptamer concentrations, which can increase the upper limit of dynamic 
range and improve sensitivity. This hypothesis was based on the inherent signal 
enhancement given by cooperative binding in the proximity effect,4-6 which does not 
apply to background. To understand the impact of connector binding affinity on S/BG 
ratio in PLA, we developed a simple experimental model using DNA hybridization and 
fluorescence quenching (Figure 3-5). First, we define an ?asymmetric connector? in the 
model system as a connector with one side shortened, compared to the original PLA 
connector, C20,PLA. Herein, model connector sequences are represented by Cb, where b is 
equal to the total number of bases in the connector that are complementary to the Loop 
sequence. For example, the symmetric connector used in the model is represented by C20, 
similar to the 20-base connector used in the original PLA work,1,3 whereas an asymmetric 
connector with only six bases in the variable region is represented by C16. Sequences are 
noted in Table 3-1, with variable regions in bold. (Note that connectors used later for 
PLA or asymmetric PLA are differentiated from model connectors using the subscript 
?PLA,? such as C20,PLA or C16,PLA.) In our model system, signal and background were 
treated separately. The equilibria shown in Figure 3-5 represent signal complex formation 
63 
 
 
with a sequential binding mode and background complex formation with an independent 
binding mode. These are not blind assumptions but are based upon Hill coefficients 
which were determined by nonlinear least-squares fitting of our experimental data 
(discussed in detail below). 
 
 
 
 
  
 
Figure 3-5:  Schematic of the experimental model system used to predict 
signal and background in PLA methods. The ?signal? complex is defined as 
Loop-Cb, whereas the ?background? complex is defined as FreeA-Cb-FreeB. 
Model connectors were varied from C15 to C20 in this study. 
 
 
 
64 
 
 
To model target-dependent signal in PLA, a single DNA loop was hybridized with 
connector sequences (Figure 3-5, top). The aptA?T?aptB complex in PLA was modeled 
using a 100-nucleotide DNA loop (Loop) that is capable of hybridizing with a connector 
sequence at both ends, representing the ideal case in which aptamers possess infinite 
affinities for their protein analyte (Kd?s ?  0). To measure the extent of hybridization, the 
Loop was covalently labeled with IABkFQ quencher at the 5?-end (black circle) and with 
FAM at the 3?-end (green circle). Quenching of FAM fluorescence ensued when both the 
5?- and 3?-ends of the Loop were hybridized to a connector sequence, mimicking the 
proximity hybridization complex in PLA (i.e., Loop?Cb represents aptA?T?aptB?C20,PLA). 
Thus, a decrease in Loop fluorescence was referred to as ?signal? in this case. The model 
of background ligations in PLA (Figure 3-5, bottom) follows from the model of signal. 
The FAM- and IABkFQ-labeled Loop sequence was essentially split into two free strands 
(FreeA and FreeB), representing aptamers unbound by protein analytes. Upon combination 
of both free strands and a connector sequence, the extent of fluorescence quenching 
represented target-independent hybridization in PLA and was proportional to the amount 
of background ligation products that would be expected (i.e., FreeA?Cb?FreeB represents 
aptA?C20,PLA?aptB). Thus, a decrease in fluorescence was referred to as ?background? in 
this case. 
 
3.1.6     Complex Formation with Asymmetric Connectors 
Moving toward our goal of reduced background in PLA, the experimental model 
(Figure 3-5) was used to determine effective dissociation constants (Kd,eff) of signal and 
background complexes at different connector lengths (Cb, where 15 ? b < 20). The 
65 
 
 
assumption was that the Kd,eff values could represent the relative amounts of signal and 
background. Separate titrations of the Loop and Free strands with the Cb strand gave the 
corresponding Kd,eff values of the complexes. The S/BG was then calculated as 
,,
,,
S / B G S B Ga e f f d e f fB G S
a e f f d e f f
KK
KK??
                                         (1) 
This S/BG represents the ideal case in which aptamers have infinite affinities for the 
target (Kd?s ?  0). 
Connectors? affinities for the FreeB strand (or for one arm of the Loop) were reduced 
by decreasing the number of complementary bases to FreeB or the Loop arm (blue 
segments in Figure 3-1B; blue text in Table 3-1). Base-pairing dissociation constants for 
the invariable segment (Kd,A) and for each variable segment (Kd,B) of the connectors were 
calculated (Table 3-2) using the well-established, thermodynamic nearest-neighbor 
model.13 Starting with the 20-base connector (C20) similar to that used in standard PLA, 
the variable segment was decreased in length, resulting in five different asymmetric 
connectors (C15 through C19). Titrations with these asymmetric connectors and the 
symmetric control connector (C20) were carried out on both the signal and background 
models (Figure 3-5), and the decrease in fluorescence was used to assay stability. 
3.1.7     Background Reduction in the Experimental Model 
Modeling of cooperative complex formation has proven difficult, particularly in the 
realm of predicting the affinity of bivalent ligands based on known affinities of each 
individual ligand.4-6 Rather than attempting to develop an extensive model of the system 
shown in Figure 3-5, titration data was analyzed using the Hill equation,16 a simplified 
66 
 
 
model that has found wide use in biochemistry, physiology, and pharmacology to analyze 
binding equilibria and ligand-receptor interactions.17 According to Weiss,17 the Hill 
coefficient is best described as an ?interaction coefficient? that reflects the extent of 
cooperativity among multiple ligand binding sites. Since it is expected that cooperativity 
is inherent to the formation of the aptA?T?aptB?C20 complex in PLA, or the Loop?Cb 
complexes in our model, we employed nonlinear least-squares (NLLS) fitting of our data 
to the Hill equation (eq 2) to assay this cooperativity and to determine Kd,eff values.  
,
[]()
( ) [ ]
nb
i f i nn
d e ff b
CS S S S
KC? ? ? ?                            (2) 
where S is the fluorescence signal measured as a function of [Cb], the connector 
concentration, Si and Sf are the initial and final signals, respectively, and Kd,eff represents 
the effective dissociation constant of the Loop?Cb complex ( ,SdeffK ) or the 
FreeA?Cb?FreeB complex ( ,BGdeffK ). In this way, fluorescence measurements during 
titration with increasingly higher [Cb] could be fit to the Hill equation (eq 2) and the Kd,eff 
values could be extracted. Furthermore, each extracted Hill coefficient, n, represented the 
extent of cooperativity in that particular system. Our expectation was that the Loop?Cb 
complexes (signal) would have significantly increased cooperativity (higher n) compared 
to the FreeA?Cb?FreeB complexes (background). 
 
 
 
 
67 
 
 
 
 
 
 
 
Figure 3-6: Results of experimental model of proximity hybridization. 
Fluorescence response curves for the signal (Loop, red squares) and 
background (Free, blue triangles) complexes when titrated with C15 
through C20. NLLS fits to the Hill equation (Eq. 2) are shown as solid 
curves. It is clear from this depiction that as the connector affinity 
decreases, the background stability is drastically decreased, while the 
signal stability is only moderately decreased.  
 
 
68 
 
 
Example data sets from these titrations are shown in Figure 3-6, with error bars 
representing standard deviations of the triplicate measurements. Figure 3-6 shows that the 
Loop?Cb complex (red squares) is more thermodynamically favored compared to the 
FreeA?Cb?FreeB complex (blue triangles). All titrations were fit to eq 2 via NLLS fitting 
to determine Kd,eff values and S/BG (via eq 1) for each asymmetric connector (Cb); 
measured and calculated values are shown in Table 3-2. This figure clearly shows that the 
relative stability of the signal complex (Loop?Cb, red squares) over the background 
complex (FreeA?Cb?FreeB, blue triangles) is increasing with decreasing connector length 
through C20 to C15. With the use of the symmetric connector, C20, the S/BG value was 
determined to be 5.72, the S/BG value using C16 was determined to be 838. Thus, a 146-
fold increase in S/BG was achieved by simply decreasing the connector length in an 
asymmetric manner. As shown in Figure 3-7A, the background levels with C20 were 382-
fold larger than the background with C16, whereas the signal differed by only 3.16-fold. 
These large decreases in background and only modest decreases in signal gave the 
expected increase in S/BG with decreasing connector affinity (Figure 3-7B). The 
agreement between measured ,BGdeffK  and calculated Kd,B values was also encouraging 
(Figure 3-7C). It must be noted that the background complex titration experiment for C15 
was impractical, due to the expense involved in working with millimolar concentrations 
of DNA. Although, using nearest-neighbor calculations,13 it is predicted that the S/BG 
could be as high as 1.7 ?  104 for C15. These results provided a first step toward 
confirmation of our hypothesis. Also shown in Figure 3-7B is a curve (dashed line) 
calculated from data presented by Tian and Heyduk,6 who developed a similar 
experimental model using bivalent ligands of various affinities. It is clear that the general 
69 
 
 
trend observed in our data matches with that seen by Tian and Heyduk with an obvious 
offset in magnitude. This offset is likely due to the model differences, in which Tian and 
Heyduk used a 6 nm spacer between their binding sites, whereas our model has 
essentially no spacer between the binding sites. Inclusion of an equivalent spacer in our 
system would be unreasonable, since PLA requires subsequent ligation of the aptamers 
after hybridization of the signal complex. Nonetheless, the similarities were encouraging. 
Figure 3-7B presents the relationship between S/BG and the affinities of the variable 
segments of the connectors, which are represented by the calculated Kd,B values13 (Table 
3-2). This analysis is useful, since Kd,B will ultimately become the independent variable in 
the asymmetric PLA approach. As shown in the figure, S/BG increases proportionally 
with Kd,B. In theory, one could increase Kd,B even further. In the extreme example, one 
could use C11 (only one base pair in the variable segment). However, the S/BG parameter 
is not useful if there is insufficient signal present with respect to the noise of 
measurement, and it is likely that the amount of either signal or background (or both) 
complexes using C11 would be negligible. The connector length, b, could be considered 
an alternative for the independent variable (shown as upper axis in Figure 3-7, parts A 
and B), but care must be taken, since b remains dependent upon base composition, 
making it less precise than Kd,B. 
 
 
 
 
 
70 
 
 
 
 
 
Hill coefficients (n) were also determined by NLLS fitting of the data to eq 2. The 
mean n values for formation of all of the background (FreeA?Cn?FreeB) and signal 
(Loop?Cn) complexes were 1.42 ?  0.19 and 1.76 ?  0.11, respectively. These values 
were shown to be statistically different (p < 0.01). Cooperativity enhancement in the 
 
 
 
 
 
points follow a slope of unity (black line), the measured and calculated 
values are shown to be well correlated.  
 
 
Figure 3-7:   (A) relative amounts 
of signal and background using 
each connector and (B) signal-to-
background ratios as functions of 
the dissociation constants (Kd,B) 
of the variable segments of each 
connector, Cb. (C) Data from 
Table 3-2 is shown in graphical 
form to highlight the experimental 
confirmation of SantaLucia?s 
nearest-neighbor calculations of 
Kd,B. Ideally, should equal 
Kd,B. Since the majority of the  
71 
 
 
signal complex was expected, since this effect gives PLA the ability to discriminate 
signal from background. Furthermore, as reported by Weiss,17 these Hill coefficients 
suggest that the signal complex forms in a sequential manner, whereas the background 
complex forms in an independent manner, as reflected in Figure 3-5. 
In summary, the improvements in S/BG in the experimental model is derived from 
the more than 2 orders-of-magnitude increase in ,BGdeffK  (reduced affinity) that is achieved 
using asymmetric connectors, whereas ,SdeffK  is increased by less than 1 order of 
magnitude (Table 3-2, Figure 3-7A); thus, by eq 1, S/BG is increased significantly 
(Figure 3-7B). However, application of asymmetric connectors to PLA requires 
experimental confirmation, as given below, since we began our model with the 
assumption of infinite aptamer affinities. 
 
3.1.8     Confirming the Experimental Model in Asymmetric PLA 
3.1.8.1   Design of Asymmetric PLA 
A schematic of asymmetric PLA is shown in Figure 3-2B. Since decreased 
background was achieved in the experimental model (Figure 3-6, 3-7), the next step was 
to confirm that this approach was useful for PLA. The real assay is inherently more 
complicated than the experimental model, requiring not only a four-part complex 
formation but also ligation of aptA to aptB, followed by qPCR.1,3 Furthermore, aptamer 
affinities are finite in the real system, compared to the assumption of infinite affinities in 
the model. On the basis of these factors, we expected that the observations from the 
model would be less pronounced in the real system. Asymmetric PLA (Figure 3-2B) is 
based on the formation of the complex aptA?T?aptB?Cb,PLA, where b < 10 is achieved by 
72 
 
 
decreasing connector lengths on the side complementary to aptB. Shorter connectors, 
therefore, have higher Kd,B (Table 3-2), or lower affinity. 
For the standard PLA system (C20,PLA, b = 20), the optimal connector concentration 
was determined to be > 40 nM.1 This correlates well with the model results shown in 
Figure 3-6. At C20,PLA = 40 nM, the stability of the signal complex is near maximal, 
whereas the stability of the background complex is minimized as much as possible. 
Working on this principle, the results from the titrations in the experimental model 
(Figure 3-6, 3-7) were used to estimate the optimal connector concentrations, [Cn]opt, for 
each model connector of different length. These values are reported in the rightmost 
column of Table 3-2 and were used for all asymmetric PLA experiments to follow. This 
approach allowed the interrogation of each asymmetric PLA system at its optimal values 
of signal and background. 
 
3.1.8.2   Background Reduction Using Asymmetric PLA 
Starting with the identical reagents used by Landegren and co-workers1,3 for 
sensitive detection of human thrombin, connector lengths were systematically decreased 
from the initial C20,PLA sequence. Background ligation levels were measured by carrying 
out the assay without the target protein (human thrombin) present. As shown in Figure 3-
8A, the background levels in asymmetric PLA were significantly reduced by increasing 
Kd,B (or decreasing length, b), matching well with the experimental model (Figure 3-7A). 
Background levels in asymmetric PLA with C16,PLA were 46.7-fold lower than with 
standard, symmetric PLA (C20,PLA), thus confirming our hypothesis. It should, therefore, 
73 
 
 
be possible to use higher concentrations of aptamer probes with asymmetric PLA to 
improve the dynamic range and sensitivity. 
 
3.1.8.3   Dynamic Range and Sensitivity Enhancement with Asymmetric PLA 
A major limitation of standard PLA is the relatively narrow dynamic range. This 
disadvantage is directly related to the generation of background ligations. The original 
work in PLA for human thrombin detection determined the optimal concentrations of 
aptamer probes to be 15 and 20 pM for aptA and aptB in the 5 ?L incubation mixture, 
respectively.1 At higher probe concentrations, increased background ligations were 
shown to be detrimental to the S/BG of the assay. Since the assay relies on the 
simultaneous binding of two aptamers to the same target, simple probability theory shows 
that the population of signal complexes (aptA?T?aptB?Cb,PLA) will first increase then 
decrease as the concentration of target is increased. The upper limit of the dynamic range 
is thus positively related to both the concentrations and the Kd?s of the aptamer probes. 
Gullberg et al.3 showed that target at 10 times higher concentration than probes could not 
be assayed with probe Kd < 0.4 nM but could be assayed with a probe Kd = 2.5 nM, albeit 
with relatively low S/BG. This implies that in standard PLA, a lower LOD is achievable 
with higher affinity probes at the expense of narrow dynamic range. 
Previous publications using standard, aptamer-based PLA for thrombin detection did 
not report the LOD or the dynamic range.1,3 However, by analyzing the presented data, 
the reported LOD was determined to be 16 amol (significantly different from 
background, p < 0.05), with a dynamic range from 16 to 400 amol. In our laboratory, 
with the standard PLA system using C20,PLA, we obtained a detection limit about 3-fold 
74 
 
 
higher than 16 amol, but we were able to achieve a significantly wider dynamic range 
compared to the previous reports. As shown in Figure 3-8B (open blue squares), our 
standard PLA detection limit was 50 amol, and our dynamic range was from 50 to 3000 
amol (7.68-fold wider than previous reports). 
With significant reductions in background using asymmetric PLA (Figure 3-8A), we 
hypothesized that the dynamic range could be improved using higher aptamer 
concentrations with connector C16,PLA. This hypothesis was confirmed by titration of the 
system with human thrombin (Figure 3-8B). Asymmetric (closed black circles) and 
standard (open blue squares) PLA were carried out under the following optimal 
conditions in the 5 ?L incubation mixture: asymmetric, [C16,PLA] = 480 nM, [aptA] = 0.9 
nM, and [aptB] = 1.2 nM; standard, [C20,PLA] = 45 nM, [aptA] = 15 pM, and [aptB] = 20 
pM. Using asymmetric PLA, the dynamic range was increased by 2 orders of magnitude, 
and the sensitivity was improved by a factor 1.57 compared to the standard system with 
C20,PLA (Figure 3-8, parts B and C). In addition, the maximum achievable S/BG was 
larger by a factor of 3.15. These improvements were accomplished with an additional 3-
fold improvement in LOD to 5.0 amol of thrombin. When compared to previously 
reported data,1,3 our dynamic range was improved by an even larger factor of 323. In 
Figure 3-8B, horizontal bars mark the dynamic ranges (blue line is for asymmetric PLA, 
orange line is for standard PLA), with the upper bounds defined as the highest 
concentration with a statistically higher (p < 0.05) value of signal-to-background ratio 
compared to the adjacent lower concentration. Error bars represent standard errors of 
quadruplicate runs of qPCR. 
 
75 
 
 
 
 
  
Figure 3-8:  Asymmetric PLA. (A) Amount of background ligations vs 
dissociation constant (Kd,B) of the variable segments of each connector, 
Cb,PLA, in the absence of target protein. (B) Titration with human 
thrombin. Asymmetric connector C16,PLA (filled black circles) and 
symmetric connector C20,PLA (open blue squares) were used for 
comparison. Error bars represent the standard error of quadruplicate 
qPCR measurements. Horizontal bars define the assay dynamic 
ranges. (C) Linear scale comparisons of dynamic range, sensitivity, and 
limit of detection (LOD) improvements using asymmetric connector 
C16,PLA (white bars) compared to the standard connector C20,PLA (blue 
bars). Error bars depict standard deviations of linear sensitivities. The 
cross-hatched bar represents the LOD obtained in the original reports of 
PLA. 
76 
 
 
Since logarithmic scales were used in Figure 3-8B, linear scale comparisons of 
dynamic ranges, sensitivities, and limits of detection are shown in Figure 3-8C. 
Sensitivities were determined as the slopes of the linear regressions of both data sets in 
the range from the LOD to 3000 amol. This figure helps to clarify the significant 
advantages provided by asymmetric PLA, in comparison to standard PLA. Most notably, 
our approach of using asymmetric connectors has allowed a significant improvement in 
one of the major limitations of the assay, dynamic range. This improvement should be 
generally applicable to any of the previously developed PLA methods.1,3-11 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
77 
 
 
3.2     DNA-Based Experimental Model of Electrochemical Proximity Assay (ECPA) 
3.2.1     Background 
Sandwich immunoassays are now widely adopted for bioanalytical quantitation of a 
multitude of proteins in various fields, and there are numerous assay kits available for 
purchase. The assays possess both high sensitivity and selectivity, due in large part to the 
requirement of simultaneous binding of two high-affinity antibodies, and they are highly 
scalable. However, from a biosensing perspective, sandwich immunoassays still have 
room for improvement. These heterogeneous assays cost ~$10 per sample and require 
washing steps, making protocols tedious and time-consuming, with significant user 
intervention and reagent volumes required. Point-of-care analysis is difficult to envision 
using sandwich ELISAs. 
Proximity ligation assays (PLA) (Figure 3-9A) are some of the most simple-to-use 
and sensitive protein assays developed to date. PLA is homogeneous and can achieve 
protein detection limits in the zeptomole range (fM in 1 ?L). A key concept in PLA is the 
?proximity effect,? i.e. when two oligonucleotide-conjugated antibodies are bound to 
target, the local concentration of the oligos is largely increased, making weak interactions 
with connectors more kinetically feasible. Probes are then enzymatically ligated, and the 
products are PCR amplifiable. PLA can be used with aptamers or with antibodies. A 
modified version of this assay is the molecular pincer assay (Figure 3-9B).19 Although 
this assay will have lower sensitivity, it can achieve highly specific, homogeneous protein 
detection (pM detection limits), with direct readout using fluorescence resonance energy 
transfer (FRET) between dyes attached to the oligo arms, making it amenable to point-of-
care analysis. 
78 
 
 
 
 
A disadvantage of the aforementioned assays is that purification of the Ab-oligo 
conjugate is a determining factor for sensitivity. Use of aptamers in these assays has the 
potential to significantly improve performance. Aptamers are RNA or single stranded 
DNA (ssDNA) molecules that have been isolated from random-sequence nucleic acid 
libraries and selected for high affinity against a target protein, small molecule, or even 
 
 
 
 
Figure 3-9:   (A) Proximity ligation assays (PLA) and (B) molecular 
pincer assays.  
 
 
79 
 
 
whole cells. Aptamers exhibit a number of advantages over antibodies as molecular 
recognition elements, especially for diagnosis. Their in vitro isolation eliminates the need 
for animal hosts, chemical modification of nucleic acids is highly flexible and well-
developed, and aptamer binding affinity can be tailored. Furthermore, aptamers can bind 
virtually any molecule, even toxic or non-immunogenic targets. Lastly, aptamers will 
refold into functional structures after exposure to extreme conditions (pH, temperature, 
etc.), making them amenable to reusable biosensors. Thus, SELEX has opened a new and 
promising field of research, giving 
more than mere alternatives to 
antibodies. 
Our proposed approach will 
combine the proximity effect 
(Figure 3-9) with an 
electrochemical approach recently 
proven feasible by the Revzin 
group. They used a structure-
switching aptamer to move an 
electrochemically active species 
away from a gold electrode as 
target was bound, providing 
homogeneous, electrochemical (EC) detection of interferon gamma with 0.06 nM 
detection limit. However, the flexibility of this approach is severely limited due to the 
requirement of a structure-switching aptamer, which will not exist against many targets. 
 
 
 
Figure 3-10:   The electrochemical 
proximity assay (ECPA). In the 
presence of the target protein, this 
five-part complex moves the redox-
active methylene blue (MB) near the 
gold surface, thus increasing current 
in proportion to the protein analyte.  
 
 
 
80 
 
 
As shown in Figure 3-10, we propose to use the proximity effect to move the EC-active 
methylene blue (MB) closer to a gold electrode upon binding of two antibody-oligo 
conjugates to protein targets. This novel combination of proximity-dependent assays with 
EC current enhancement allows wash-free, highly-sensitive, and highly selective 
detection of target molecules, in a format amenable to point-of-care use. The assay 
format will be extremely flexible, since any target with two antibodies could be assayed. 
 
3.2.2     Model Systems of ECPA 
As in our previous work with PLA,18 here we utilized a DNA loop to model the 
probe-target complex in ECPA (Figure 3-11), making the assumption that probe affinity 
for the target protein is infinite. The 80-nucleotide DNA loop mimics formation of the 
ECPA complex, bringing MB near the gold surface and increasing redox current. The 
background was modeled using only the thiolated DNA and MB-DNA (Figure 3-11). 
This experimental model greatly simplified the optimization of experimental parameters. 
Since the surface-dependent ECPA involves a different type of cooperative complex 
formation compared to homogeneous PLA, we devised two new strategies for 
minimizing background in ECPA. 
 
 
 
 
81 
 
 
 
 
The first strategy was to decrease the binding affinity between thiolated DNA and 
MB-DNA by reducing the number of complementary bases in the thiolated DNA (Figure 
3-12A). The hypothesis was that the amount of background hybridization between 
thiolated DNA and MB-DNA would be greatly reduced, thereby reducing background 
current greatly without a large decrease in signal current. However, the background 
reduction was accompanied by a large decrease in signal (more detail is shown in chapter 
4), since the weakened connection also weakened hybridization of the DNA Loop (model 
of signal). In an attempt to reduce background without such a large signal reduction, our 
second strategy was to utilize a short DNA competitor with the thiolated DNA. We 
hypothesized that when using a competitor sequence, background hybridization would 
occur more slowly than signal hybridization, since both signal and background 
 
 
 
 
Figure 3-11:  DNA-based model for ECPA. A continuous DNA loop is 
used to model the Signal complex shown in Figure 3-10. Background is 
modeled by simply adding MB-DNA without the loop.  
 
 
 
 
 
 
 
82 
 
 
complexes must displace the short competitor prior to current enhancement by the MB-
DNA strand. Figure 3-12B shows a representation of the delayed background formation 
over time, mediated by competition with competitor strands. This way, signal should 
form rapidly, while background would be delayed kinetically by the competitor.  
 
 
 
 
  
 
Figure 3-12:  Model system strategies. (A) Depiction of background 
reduction in strategy 1. Fewer base pairs (weaker hybridization) 
between thiolated DNA and MB-DNA results in lower background 
current. (B) Depiction of background reduction in strategy 2, where a 
competitor strand prevents or slows background formation over a given 
time window.  
 
83 
 
 
3.3    Conclusions 
This chapter describes experimental models of proximity hybridization that were 
successfully developed and tested. The first part of the chapter describes modeling and 
use of asymmetric connectors to significantly improve the dynamic range of PLA, 
thereby addressing a major limitation of the assay. Learning from the DNA-based 
models, sensitivity, LOD, and maximum S/BG were also improved in the real assay. By 
enhancing its already successful predecessor, this asymmetric PLA method approaches 
an ideal assay system, providing a homogeneous and highly specific protein assay with 
the capability to detect very small quantities of protein in free solution, at high 
throughput, and with good dynamic range. 
This DNA-based experimental model system ultimately improved our understanding 
of the proximity effect and its inherent cooperative assembly in PLA or any other 
proximity assay, highlighting the importance of the interplay between connector-to-
aptamer affinities and aptamer-to-target affinities in maximizing S/BG and dynamic 
range. Generally speaking, low connector affinities can be paired with high probe 
affinities (aptamers or antibodies) to achieve this goal. 
The second part of this chapter reports the use of our DNA model systems to study 
ECPA.  We compared two strategies, in order to decrease background current and widen 
the gap between background and signal current. The DNA model system design and 
experimental results provided very important information for the aptamer-based and 
antibody-based ECPA tests discussed in later chapters. 
 
 
84 
 
 
 
3.4    References 
 
(1) Fredriksson, S.; Gullberg, M.; Jarvius, J.; Olsson, C.; Pietras, K.; Gu? stafsdo? 
ttir, S. M.; Ostman, A.; Landegren, U. Nat. Biotechnol. 2002, 20, 473?477. 
(2) Adler, M.; Wacker, R.; Niemeyer, C. M. Analyst 2008, 133, 702?718. 
(3) Gullberg, M.; Gu? stafsdo? ttir, S. M.; Schallmeiner, E.; Jarvius, J.; Bjarneg?rd, 
M.; Betsholtz, C.; Landegren, U.; Fredriksson, S. Proc. Natl. Acad. Sci. 
U.S.A. 2004, 101, 8420?8424. 
(4) Zhou, H. X. Biochemistry 2001, 40, 15069?15073. 
(5) Zhou, H. X. J. Mol. Biol. 2003, 329, 1?8. 
(6) Tian, L.; Heyduk, T. Biochemistry 2009, 48, 264?275. 
(7) So?derberg, O.; Leuchowius, K. J.; Gullberg, M.; Jarvius, M.; Weibrecht, I.; 
Larsson, L. G.; Landegren, U. Methods 2008, 45, 227?232. 
(8) Gustafsdottir, S. M.; Schlingemann, J.; Rada-Iglesias, A.; Schallmeiner, E.; 
Kamali-Moghaddam, M.; Wadelius, C.; Landegren, U. Proc. Natl. Acad. Sci. 
U.S.A. 2007, 104, 3067?3072. 
(9) Nordengrahn, A.; Gustafsdottir, S. M.; Ebert, K.; Reid, S. M.; King, D. P.; 
Ferris, N. P.; Brocchi, E.; Grazioli, S.; Landegren, U.; Merza, M. Vet. 
Microbiol. 2008, 127, 227?236. 
(10) Gustafsdottir, S. M.; Wennstro?m, S.; Fredriksson, S.; Schallmeiner, E.; 
Hamilton, A. D.; Sebti, S. M.; Landegren, U. Clin. Chem. 2008, 54, 1218?1225. 
(11) Heged?s, E.; K?kai, E.; Kotlyar, A.; Dombr?di, V.; Szab?, G. Nucleic Acids 
Research 2009, 37, e112. 
(12) Livak, K. J. User Bulletin No. 2: ABI PRISM 7700 Sequence Detection System; 
PE Applied Biosystems: Foster City, CA, 1997; pp 11-15. 
(13) SantaLucia, J. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 1460?1465. 
(14) Leslie, D. C.; Sohrabi, A.; Ikonomi, P.; McKee, M. L.; Landers, J. P. 
Electrophoresis 2010, 31, 1?8. 
(15) Kuhn, H.; Frank-Kamenetskii, M. D. FEBS J. 2005, 272, 5991?6000. 
(16) Hill, A. V. J. Physiol. 1910, 40, iv?vii. 
(17) Weiss, J. N. FASEB J. 1997, 11, 835?841. 
(18) Kim, J.; Hu, J.; Sollie, R. S.; Easley, C. J. Anal. Chem. 2010, 82, 6976-6982. 
(19) Heyduk, E.; Dummit, B.; Chang, Y.-H.; Heyduk, T. Anal.Chem 2008, 80, 5152-
5159. 
 
85 
 
 
 
CHAPTER 4  
The Electrochemical Proximity Assay (ECPA) 
 
4.1     Introduction 
Diagnostics is one of the most critical steps in health care and medical treatment.1  
Specific protein detection is of great importance in this realm, since it is currently one of 
the predominant methods to diagnose the onset or progression of disease states.2,3  Unless 
specialized point-of-care assays are available for the protein of interest, quantitation is 
typically performed in a centralized laboratory by technicians.4  This process is expensive 
and could waste time that is critical to patient care.  Over the years, clinical approaches 
for point-of-care testing have addressed this challenge for select analytes,5-9 yet these 
assay formats are highly specialized to the particular target molecule, thus inflexible to 
apply to other targets.  To keep pace with expectations in future point-of-care testing, 
there is a need for more flexible, yet highly sensitive, quantitative, and easy-to-use 
methods.4 
Although point-of-care devices are welcome in clinical and research laboratories, 
the existence of surrounding infrastructure places fewer constraints on methodology.  
Based on their inherent flexibility, sandwich enzyme-linked immunosorbent assays 
(ELISA) have emerged as the method of choice for protein quantitation in clinical and 
86 
 
 
research laboratories.3  Unfortunately, these heterogeneous assays require expert users 
with dedicated instrumentation, and they are time-consuming, laborious, and expensive.  
Quantitative, point-of-care protein analysis is thus not possible with standard sandwich 
ELISA formats.  Nonetheless, the flexibility of the dual-antibody recognition concept is 
highly valuable and has served as a guide to various alternative strategies in recent 
years.10-16 
Proximity immunoassays such as the proximity ligation assay (PLA)11,12 or the 
molecular pincer assay13 can overcome some of the limitations of ELISA.  PLA, for 
example, is one of the most simple-to-use and sensitive protein assays developed to 
date.17 The assay is homogeneous (no washing steps), and detection limits rival or 
outperform ELISAs, even with much smaller sample volumes.  A key concept in PLA is 
the ?proximity effect,? which relies on simultaneous recognition of a target molecule by a 
pair of affinity probes.  The bound probes can then be covalently linked by enzymatic 
ligation of their oligonucleotide tails, and qPCR is used as the readout, with products 
proportional to target protein concentration.  PLA has been shown functional with 
aptamer pairs11 and with a variety of antibody pairs.12 Although nucleic acid aptamers 
have garnered significant attention in the analytical and biosensing communities based on 
their many potential advantages,18-24 the use of aptamers as affinity probes in PLA is 
severely limited.  PLA requires two aptamers binding at separate sites on the same 
protein target, but aptamer pairs unfortunately do not yet exist against most targets.  In 
PLA12 or in the pincer assays,13 this limitation was overcome by employing antibody-
oligonucleotide conjugates as probes, since the popularity and success of sandwich 
methodology (ELISA, Western blots) has afforded a large, commercially available library 
87 
 
 
of antibody pairs against many proteins.  These assays thus provide simpler and less 
expensive alternatives to ELISA. 
Nonetheless, limitations in current proximity assays impede their use in a point-of-
care setting.  Although the use of qPCR gives PLA its high sensitivity, this readout 
requires that each sample be added to a tube with ligation and PCR reagents, and then be 
inserted into a qPCR instrument followed by 1-2 hours of amplification and analysis.  
The molecular pincer assays are simpler and more rapid (<20 min),13 making them more 
amenable to point-of-care measurements by fluorescence readout; however, the limit of 
detection of these assays is several orders of magnitude higher than PLA.  Thus, there is a 
need for a more sensitive yet simpler readout for proximity assays that is amenable to 
point-of-care testing. 
Electrochemical detection is of particular interest in the development of biosensors 
because it offers great signal stability, simple instrumentation, high sensitivity, and ease 
of calibration compared to fluorescence, as well as excellent compatibility with 
miniaturization technologies.25,26  Here, we present the marriage of the proximity assay 
concept with electrochemical detection to give a simple, highly sensitive, flexible 
strategy for specific protein quantitation, termed the electrochemical proximity assay 
(ECPA).  ECPA uses the proximity effect to move an electrochemically active label, 
methylene blue (MB),27 closer to a gold electrode upon binding of two probes to a protein 
target, an approach akin to electrochemical DNA sensing28,29 or specialized aptamer-
based protein sensing30-32 reported by others.  In the presence of protein targets, the redox 
current in ECPA is quantified using square wave voltammetry (SWV) and is found to 
depend directly on the concentration of target. This detection strategy is based largely 
88 
 
 
upon pioneering work by the Plaxco group,33-37 in using MB-labeled DNA for biosensing.  
Building upon this work and on aptamer-based protein sensing by Zhang et al.,32 we have 
added the antibody-based proximity assay concept.  We used a DNA-based experimental 
model to optimize signal-to-background ratios, ultimately providing a direct insulin 
detection limit that is lower than most commercially available ELISAs, with a dynamic 
range >40-fold wider than these ELISAs.  These results were achieved with direct 
electrochemical readout, i.e., without requiring washing steps, which bodes well for the 
future of ECPA in point-of-care settings.  In contrast to other approaches for 
electrochemical protein sensing,24,30 ECPA should be useful for any protein with 
available antibody pairs. 
 
4.2     Reagents and Experimental Methods 
4.2.1     Reagents and Materials 
All solutions were prepared with deionized, ultrafiltered water (Fisher Scientific). 
The following reagents were used as received: insulin antibodies (clones 3A6 and 8E2; 
Fitzgerald Industries), 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) 
(99.5%), tris-(2-carboxyethyl) phosphine hydrochloride (TCEP), (Sigma-Aldrich, St. 
Louis, MO), bovine serum albumin (BSA, 98%; EMD Chemicals Inc.), human thrombin, 
immunoglobulin E (IgE), and insulin (Sigma Aldrich). Methylene blue-conjugated DNA 
(MB-DNA) was purchased from Biosearch Technologies (Novato, CA), purified by RP-
HPLC. Oligonucleotides were obtained from Integrated DNA Technologies (IDT; 
Coralville, Iowa), with purity and yield confirmed by mass spectrometry and HPLC, 
respectively. Sequences (listed 5?  to 3? ) for aptamer based ECPA were as follows. 
89 
 
 
Thrombin aptamer A (THRaptA): AGTCCGTGGTAGGGCAGGTTGGGGTGA-
CTTTTTTTTTTTTTTTTTATATTTTTTTTTTCTCGCGGATTTGAACCCTAACG;  
 
 
Thrombin aptamer B (THRaptB): TAGGAAAAGGAGGAGGGTGGG-
ATTGGTGTGTGTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTGGTTGGTGTGGT
TGG. Sequences (listed 5?  to 3? ) for antibody-based ECPA were as follows. Insulin 
antibody arm 1 (AbArm1): /5AmMC6//iSp18/CCCACTTAAACCTC-
AATCCACGCGGATTTGAACCCTAACG; Insulin antibody arm 2 (AbArm2): 
TAGGAAAAGGAGGAGGGTGGCC-CACTTAAACCTCAATCCA/iSp18//3AmMC6/. 
Sequences of ssDNA strands used in the experimental model are given in Table 4-1. 
 
4.2.2     Preparation of the Electrode and DNA Monolayer Assembly 
ECPA sensors for the model system, for thrombin detection, and for insulin 
detection were fabricated using a gold working electrode (Bioanalytical Systems Inc., r = 
 
Table 4-1.   Single-Stranded DNA Sequences Used in the ECPA Model Systems 
(Strategies 1 and 2)a 
 
 
90 
 
 
0.75 mm). The gold electrode was polished carefully to a mirror surface with an aqueous 
slurry of 0.05 ?m diameter alumina particles and then successively washed in an 
ultrasonic cleaner with water. The electrode was then immersed into fresh piranha 
solution (H2SO4/H2O2, 3:1) for 5 min, rinsed with D. I. water, and dried under a stream of 
nitrogen gas. (Caution: piranha solution is dangerous to human health and should be used 
with extreme caution and handled only in small quantities). Finally, the gold electrode 
was electrochemically polished by scanning the potential from ?0.5 to +1.5 V in 0.1 M 
H2SO4 at a scan rate of 0.1 V s?1 for 50 cycles. The cleaned gold electrode was 
thoroughly washed with D. I. water and ethanol and dried under flowing nitrogen. 
Prior to modification of the electrode, 1 ?L of 200 ?M thiolated-DNA and 1 ?L of 
200 ?M MB-DNA were each separately mixed with 2 ?L of 10 mM TCEP in two 200-
?L PCR tubes. These tubes were incubated for 90 min at room temperature (21 ?C) for 
reduction of disulfide bonds in the thiolated-DNA and to reduce the MB-moiety of the 
MB-DNA. Both of these solutions were then diluted to a total volume of 200 ?L in 
HEPES/NaClO4 buffer (10 mM HEPES and 0.5 M NaClO4, pH 7.0)29 to a final 
concentration of 1 ?M. Unless otherwise noted, all solutions used in the experiments to 
follow were carried out at pH 7.0. For immobilization, the previously cleaned gold 
electrode was transferred directly to the diluted and reduced thiolated-DNA solution and 
incubated for 16 h at room temperature in the dark. Following the formation of a self-
assembled monolayer (SAM), excess thiolated-DNA physically adsorbed on the electrode 
surface was removed via a room temperature-deionized water rinse (~ 20 s). For all assay 
strategies employing the competitor DNA strands, this same process was followed, 
except after reduction by 10 mM TCEP, the reduced thiolated-DNA solution was diluted 
91 
 
 
to a total volume of 200 ?L in HEPES/NaClO4 buffer and incubated with 2 ?M 
competitor DNA sequence (C9) for 60 min at room temperature in the dark. For 
immobilization in competitor systems, the cleaned gold electrode was transferred directly 
to this equilibrated thiolated-DNA/competitor solution then incubated for 16 h at room 
temperature in the dark. 
 
4.2.3     ECPA Probe Assembly and Electrochemical Measurements 
Electrochemical measurements were performed using an Epsilon electrochemistry 
workstation (Bioanalytical Systems, Inc.) with a standard three-electrode configuration 
consisting of a Ag|AgCl(s)|KCl(sat) reference electrode (Bioanalytical Systems, Inc.), a 
homemade platinum gauze flag (0.77 cm2) counter electrode, and a gold working 
electrode. All potentials are reported relative to the saturated Ag|AgCl reference 
electrode. Electrochemical measurements were performed in HEPES/NaClO4 buffer 
using square wave voltammetry (SWV) with a 50 mV amplitude signal at a frequency of 
60 Hz, over the range from ?0.45 to 0.00 V versus Ag|AgCl reference. The characteristic 
voltammetric peak of MB was detected by SWV at ?210 mV (vs Ag/AgCl). MB was 
chosen as the redox tag due to its excellent shelf life and robust electrochemical response 
in serum compared to other redox tags, such as ferrocene.29,31 The electrochemical 
response of each sensor was measured as follows: (1) reference and measurement SWV 
data sets were collected; (2) both raw data sets were smoothed using a 21-point boxcar 
function and baseline corrected (all data corrected with B-spline generated baseline in 
Origin 8 using two regions: ?0.40 to ?0.35 V and ?0.08 to 0.00 V); and (3) difference 
traces were generated. Signal (with target) and background (no target) voltammograms 
92 
 
 
were treated in this manner and are presented as difference traces. To prepare calibration 
graphs and calculate standard deviations, traces were integrated from ?0.330 to ?0.100 
V. In the case of the aptamer-based system, we report the average of three measurements, 
while in the case of the antibody-based system the average of two measurements is 
reported. 
Model System Strategy 1: Decreasing Binding Affinity by Reducing the Number of 
Complementary Bases. The electrode was modified as described above and was placed 
into a glass electrochemical cell with HEPES/NaClO4 buffer. Three different thiolated 
DNA sequences, G5, G7, and G10 (Table 4-1), were used in strategy 1 of the model 
system. In this way, the affinity of thiolated DNA and MB-DNA were adjusted through 
changes in the number of complementary bases between them. For modeling signal, the 
sensor was immersed in 10 nM ECPA-loop and 15 nM MB conjugated DNA sequences 
in 3 mL of HEPES/NaClO4 buffer. For modeling background, the sensor was immersed 
in 15 nM MB conjugated DNA in 3 mL of HEPES/NaClO4 buffer. Both signal and 
background currents were measured at the 15-min time point. 
Model System Strategy 2: Use of a Short DNA Competitor. The electrode was 
modified as described above and was placed into a glass electrochemical cell with 
HEPES/NaClO4 buffer. Three different competitor DNA sequences, C7, C8, and C9, 
were used in strategy 2 of the model system (Table 4-1). The sensor was allowed to 
equilibrate in 3 mL HEPES/NaClO4 buffer with various concentrations of competitors for 
6 h. For modeling background in the competitor systems, redox current was measured at 
each 10 min of the first hour, then at 90 and 120 min. Once C9 was chosen, 1:3, 1:7, 
93 
 
 
1:10, and 1:25 molar ratios of MB-DNA:C9 were tested at a fixed concentration of 15 
nM MB-DNA. 
Aptamer-Based ECPA System. The sensor was allowed to equilibrate in 3 mL of 
HEPES/NaClO4 buffer with 100 nM C9 for 6 h. Thrombin aptamers (THRaptA and 
THRaptB) were first folded by heating to 95 ?C and cooled rapidly by immersion in ice 
water to promote intramolecular interactions. Thrombin of various concentrations (from 
50 pM to 50 nM) was incubated with folded 10 nM THRaptA and 15 nM THRaptB in 
HEPES/NaClO4 buffer for 90 min prior to measurements. The thrombin/aptamer 
incubations were then added into the glass electrochemical cell. Before conducting 
voltammetric measurements, the sensor surface was allowed to react with analytes for 90 
min. Selectivity tests with other proteins (IgE, insulin, or BSA) were made under the 
same conditions. 
Antibody-Based ECPA System. The sensor was equilibrated in 500 ?L of 
HEPES/NaClO4 buffer with 300 nM C9 for 6 h. Prior to measurements, HEPES/NaClO4 
buffer was supplemented with 0.5% BSA (w/v) (to minimize antibody adsorption), 10 
nM Ab1, 10 nM Ab2, and 10 nM MB (for background measurements), and various 
concentrations of insulin (from 128 fM to 2 nM). Before conducting voltammetric 
measurements, the sensor surface was allowed to react with analytes for 40 min. 
Selectivity tests were performed in the same manner by substituting 2 nM C-peptide or 
insulin-like growth factor 1 (IGF-1) for insulin. 
 
 
 
94 
 
 
4.2.4     Preparation of Antibody-Oligonucleotide Conjugates 
The antibody-oligonucleotide conjugates used in the insulin ECPA, AbArm1?3A6 and 
AbArm2?8E2, were prepared by conjugating AbArm1 to insulin antibody 3A6 (Kd ?  1 
nM) and AbArm2 to insulin antibody 8E2 (Kd ?  0.1 nM), respectively (antibodies 
obtained from Fitzgerald Industries). Conjugation reactions and purification steps were 
accomplished using an Antibody-Oligonucleotide All-In-One Conjugation Kit (Solulink), 
according to the manufacturer? s instructions. Briefly, the oligonucleotides were first 
activated with sulfo-S-4FB, and their quantities and qualities were confirmed using 
absorbance, specifically A260 nm of unmodified activated oligonucleotides and the A260 
nm to A360 nm ratio after the modification of activated oligonucleotides. Antibodies 
were also activated with S-HyNic. Activated oligonucleotides and antibodies were then 
mixed and incubated at room temperature for 2 h. Once the conjugation reaction was 
stopped, conjugates were further purified using the supplied magnetic affinity matrix. 
(Figure 4-1) The final concentrations of the conjugates were determined by the Bradford 
protein assay. AbArm1?3A6 and AbArm2?8E2 were synthesized with 45% and 86% 
recovery from the initial amount of antibodies (100 ?g). 
 
 
 
 
95 
 
 
 
 
 
4.3     Signal and Background in ECPA 
The principle of the electrochemical proximity assay (ECPA) is shown in Figure 4-
2. The sensor is prepared by self-assembly of thiolated DNA strands onto a gold 
electrode via the alkanethiol moiety at the 5?  terminus. The quantitative capacity of 
 
 
 
 
 
Figure 4-1:   Reaction of a HyNic-modified IgG with a 4FB-modified oligo 
leads to the rapid formation of a stable antibody-oligonucleotide 
conjugate. 
Image from Solulink technical manual: Antibody-oligonucleotide all-in-one 
conjugation kit. 
 
96 
 
 
ECPA stems from cooperative hybridization of the five-part complex shown in Figure 4-
2A: thiolated DNA?DNA conjugated antibody 1?target protein?DNA conjugated 
antibody 2?MB conjugated DNA. The five-part complex forms a circular structure on the 
sensor surface through proximity-dependent hybridization of the thiolated DNA and MB-
DNA, which is the step that brings MB close enough to the gold electrode surface for 
electrochemical current enhancement. This process results in a quantity of electrons 
transferred from MB to the electrode that is proportional to the original amount of protein 
analyte ( ? signal ? ), albeit with some analyte-independent current generated by 
hybridization of thiolated DNA and MB-DNA only (? background? ). Although SWV 
does not differentiate signal and background currents, under optimized conditions, the 
signal will greatly exceed the background to allow highly sensitive, direct 
electrochemical quantitation of the protein analyte. Similar to what has been observed in 
PLA11,17 or the molecular pincer assays,13 signal enhancement over background in ECPA 
is based on the proximity effect, that is, the marked increase in the effective 
concentrations of the MB-DNA and thiolated DNA due to the simultaneous binding of 
the two probes to the same protein. This allows the MB-DNA/thiolated DNA interaction 
to be weak in the absence of protein (? background? ) yet strong in the presence of the 
protein (? signal? ). Finally, it should be noted that the detection limits of proximity 
assays are often well below the Kd values of the individual probes, which can be 
attributed to the chelate-like effect of utilizing two probes in a cooperative fashion, often 
termed the ? proximity effect.?  
 
97 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 4-2:   Principle of the electrochemical proximity assay (ECPA). 
(A) the final, five-part cooperative complex and (B) the stepwise 
operation of the assay, in which the electrode with a preassembled 
DNA/competitor monolayer is immersed into a premixed solution of 
EPCA probes (two Ab-oligos and MB-DNA) and target protein to 
generate current. 
 
98 
 
 
Through binding equilibria, a fraction of thiolated DNA will always hybridize with 
the MB-DNA sequences, even in the absence of target analyte, resulting in target-
independent hybridization, recruitment of MB to the gold surface, and an increase in 
current. A portion of this background current could also result from nonspecific 
adsorption of MB-DNA to the surface, although our results suggest that specific binding 
is the major cause. The presence of this background current is obviously detrimental to 
the assay. We applied two strategies in attempt to lower the background using our model 
system, as discussed below. 
 
4.4     DNA-Based Experimental Model of ECPA 
In chapter 3, I introduced the model 
systems of ECPA. As shown in Figure 4-
3, a 80-nucleotide DNA loop mimics 
formation of the ECPA complex (Figure 
4-2A), moving MB-conjugated 
oligonucleotide close to the gold surface, 
where it participates in oxidation and 
reduction reactions facilitated by a regenerator in solution. Thus, electrical current is 
increased in proportion to the concentration of loop strands. 
In the first strategy, binding affinity between thiolated DNA and MB-DNA was 
decreased by reducing the number of complementary bases in the thiolated DNA (Figure 
4-4A). The hypothesis was that the amount of background hybridization between 
thiolated DNA and MB-DNA would be greatly reduced, thereby reducing background 
 
  
Figure 4-3:  DNA-based model for 
ECPA.  
 
 
 
99 
 
 
current greatly without a large decrease in signal current. Figure 4-4B compares the 
signal and background responses of the system with 5, 7, and 10 complementary bases 
(G5, G7, and G10 strands). Comparing G10 to G7, as hypothesized, the background 
current was reduced by 2-fold while signal current was reduced by only 1.6-fold. 
Furthermore, compared to a background peak current of 54 nA with G10, it was indeed 
possible to reduce the background current to baseline using G5. However, the 
background reduction was accompanied by a large decrease in signal peak current from 
104 nA down to 38 nA, since the weakened connection also weakened hybridization of 
the DNA Loop (model of signal). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
100 
 
 
 
 
So in order to reduce background without such a large signal reduction, a short DNA 
competitor with the G10 system was utilized in the second strategy. The hypothesis is 
background hybridization would occur more slowly than signal hybridization, since both 
signal and background complexes must displace the short competitor prior to current 
enhancement by the MB-DNA strand. Figure 4-5A shows a representation of the delayed 
 
 
 
 
Figure 4-4:   Model system strategy 1: Decreasing binding affinity by reducing 
the number of complementary bases. (A) Fewer base pairs (weaker 
hybridization) between thiolated DNA and MB-DNA results in lower 
background current. (B) Experimental confirmation of strategy 1, with both 
signal and background currents reduced in the voltammograms as the 
number of base pairs (N) is reduced. At N = 5, background is minimized, but 
signal is reduced significantly. 
101 
 
 
background formation over time, mediated by competition with competitor strands. This 
way, signal of similar magnitude to that in the N = 10 case above should form rapidly, 
while background would be delayed kinetically by the competitor. Figure 4-5B shows 
signal and background responses of the system with 7-, 8-, and 9-base competitors (C7, 
C8, and C9). As hypothesized, the hybridized competitor sequences blocked access of 
MB-DNA to the thiolated DNA, thereby slowing background formation. Figure 4-5B 
shows that with C7 and C8, background currents of 47 and 24 nA were detected even 10 
min after addition of MB-DNA, while no background was detected for as long as 40 min 
using C9. Since C9 allowed a 40-min time window for detection, we chose C9 as the 
competitor for further experiments. Upon addition of the Loop (model of signal), 
significant signal current of 81 nA was possible after 30 min, while C9 prevented 
background formation (Figure 4-5C). Optimal conditions were determined to be 15 nM 
MB-DNA and 100 nM C9, and these were applied in the aptamer-based ECPA system, 
below. 
 
 
 
 
 
102 
 
 
 
 
4.5     Aptamer-Based ECPA 
A schematic of aptamer-based ECPA is shown in Figure 4-6A (upper right). Two 
thrombin aptamers (THRaptA, THRaptB) that bind thrombin at different sites were 
applied as affinity probes, and competitor C9 was used to minimize background. Using 
 
  
 
Figure 4-5:   Model System Strategy 2: Use of a Short DNA Competitor. 
(A) Depiction of background reduction in strategy 2, where a competitor 
strand prevents or slows background formation over a given time 
window. (B) Experimental confirmation of strategy 2. The 9-base 
competitor (C9) was the only one to show baseline current for up to 40 
min. (C) Signal and background voltammograms are shown with C9 
under optimal conditions, showing more than double the signal current 
and equal background current compared to N = 5 from strategy 1. 
 
103 
 
 
conditions optimized by the model system, background levels were measured in the 
absence of target protein (human thrombin). Similar to the model system, background 
remained at baseline current for up to 90 min, after which an increasing peak current at 
?210 mV was detected, indicating that MB-DNA was beginning to displace the 
competitors. This 90-min detection window was actually wider than the 40-min window 
observed in the model system. This difference is attributed to the decreased diffusion 
coefficient38 of the MB-DNA (40-base; ~70 ?m2 s?1) when hybridized with THRaptB 
(120-bases; ~30 ?m2 s?1), which would slow the kinetics of the competitor displacement 
process by ~2.3-fold in comparison to the model system. This estimation agrees very well 
with the 2.25-fold increase in time required for background formation. The lower right 
plot in Figure 4-6A shows the background with no thrombin (black trace) and a typical 
MB oxidation peak appearing at ?210 mV (red trace) in the presence of 2.5 nM thrombin 
after the 90 min incubation. As expected, the saturated peak current at 10 nM thrombin 
(52 nA) was of lower magnitude than the model system (81 nA), which had assumed 
probes with infinite affinity. This aptamer-based ECPA system calibrated versus 
thrombin concentration (Figure 4-6A, left plot), with sensor responses recorded in 
triplicate as integrated MB peak areas from ?330 to ?100 mV. ECPA was capable of 
detecting thrombin levels as low as 50 pM using a direct electrochemical readout, with a 
dynamic range up to 10 nM at these probe concentrations. 
To demonstrate specificity, the aptamer-based ECPA was challenged with 
nonspecific proteins including human IgE, insulin, and BSA. Figure 4-6B shows that 
essentially no response was observed in the presence of 10 nM insulin or IgE; even with 
4-fold lower thrombin (2.5 nM), the signal was ~ 40-fold larger than that of IgE or 
104 
 
 
insulin. In addition, baseline current was observed in the presence of 2% BSA, while the 
signal from 2.5 nM thrombin was recovered by 93% in 2% BSA. This result is 
encouraging for future application of ECPA to biological samples and point-of-care 
settings. 
 
 
105 
 
 
 
 
  
 
Figure 4-6:   Aptamer-based ECPA. (A) With a direct readout, a human 
thrombin detection limit of 50 pM was achieved, with a dynamic range 
up to 10 nM. Upper right image shows the principle of the assay, with 
the lower right plot showing example voltammagrams for the blank 
(black) and in the presence of 2.5 nM thrombin (red). (B) The dual 
probe assay shows high selectivity, as expected, with 93% recovery of 
signal in the presence of 2% BSA. 
 
106 
 
 
4.6     Antibody-Based ECPA 
The flexibility of the aptamer-based approach is limited because of the requirement 
of two aptamers for the target protein, since aptamer pairs exist only for a few select 
proteins. As noted above, the use of antibody-oligonucleotide conjugates as probes can 
overcome this challenge. 12,13 With the success of sandwich immunoassays, there exists a 
large, commercially available library of antibody pairs against many proteins. As proof of 
concept that ECPA can be applied to a wide variety of protein targets, we show herein 
that insulin can be directly detected using two antibody-oligonucleotide conjugates as 
ECPA probes. 
A schematic of antibody-based ECPA is shown in Figure 4-7A (upper right), again 
employing the short DNA competitor strategy. With this new assay format, a different set 
of conditions were determined as optimal, including the addition of 0.5% BSA to reduce 
nonspecific antibody adsorption and a C9 concentration of 300 nM. Using 10 nM of each 
antibody-oligo and 10 nM MB-DNA, the assay showed a 40-min detection window 
before competitor began to be displaced by MB-DNA. Since the antibody-oligo 
conjugates will significantly alter the diffusion rates of most components, we did not 
expect the kinetics of signal and background formation to follow trends observed in the 
model system or aptamer-based ECPA; nonetheless, the detection time window was 
similar to the other systems. The lower right plot in Figure 4-7A shows the background 
with no insulin (black trace) and a typical MB oxidation peak appearing at ?210 mV (red 
trace) in the presence of 2 nM insulin after 40 min. This antibody-based ECPA system 
was then calibrated versus insulin concentration (Figure 4-7A, left plot), with sensor 
response recorded in duplicate as integrated MB peak areas from ?330 to ?100 mV. 
107 
 
 
Remarkably, using a direct electrochemical readout, ECPA was capable of detecting 
insulin levels as low as 128 fM (7.43 ? 10?4 ng mL?1) with a dynamic range extending to 
2 nM (11.6 ng mL?1). The selectivity of antibody-based ECPA was tested against insulin-
like growth factor 1 (IGF-1), which has similar structure to insulin, and against C-
peptide, which is consecrated with insulin into the bloodstream. As expected, the sensor 
did not respond to higher concentrations of either IGF-1 or C-peptide (Figure 4-7B). The 
drastically improved performance of the antibody-based ECPA compared to the thrombin 
aptamer ECPA was expected, since the aptamer Kd values44 are several orders of 
magnitude higher than the typical antibody Kd. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
108 
 
 
 
 
 
 
 
Figure 4-7:  Success of antibody-based ECPA greatly improves the flexibility 
of the assay, since a large variety of protein targets could be quantified this 
way. (A) Insulin as low as 128 fM was detected with direct readout, with a 
dynamic range up to 2 nM. Upper right image shows the principle of the 
assay, with the lower right plot showing example voltammagrams for the 
blank (black) and in the presence of 2 nM insulin (red). (B) The dual-antibody 
assay also shows high selectivity, as expected. 
 
109 
 
 
Finally, Table 4-2 shows a comparison of our antibody-based ECPA to 
commercially available sandwich ELISAs for insulin detection.45-50 In order to facilitate 
equal comparison of the direct-readout ECPA with various heterogeneous ELISAs, the 
concentrations of insulin in the incubation solution of each ELISA is reported in Table 4-
2. ECPA outperforms all of the kits in terms of assay dynamic range (from 43- to 312-
fold wider range). The impressive ECPA dynamic range of 15 600 (from 128 fM to 2 
nM) should provide enhanced flexibility in sample preparation. Only one of the ?
ultrasensitive?  versions of ELISA (25-?L sample volume) has an essentially equal 
detection limit (1.1-fold higher) compared to ECPA. Compared to ? standard?  ELISA 
kits, ECPA shows between 15.6- and 60.9-fold lower limit of detection for insulin. In 
fact, using the noise level of the blank, the linearly extrapolated LOD for insulin was 
calculated to be 20 fM, lower than all ELISAs shown in Table 4-2. These performance 
improvements come with the additional benefit of a direct-readout format, making ECPA 
amenable to point-of-care analysis. To our knowledge, ECPA represents the highest 
performing direct-readout insulin assay reported to date. Looking toward future 
application in point-of-care insulin measurements in human serum, if we leverage the 
pioneering efforts of the Plaxco group using similar DNA-based electrochemical 
sensors,28,29,33-37 it should be possible to detect a variety of proteins in undiluted serum. 
Of course, since the ECPA detection limit for insulin (128 fM) is over 400-fold lower 
than the normal human serum insulin levels (~60 ? 80 pM), serum samples could be 
simply diluted to minimize interferences in this case. 
 
 
110 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table 4-2.   Performance Comparisons between ECPA and Various Commercially 
Available ELISA Kitsa 
 a
ECPA has a lower detection limit than five of the six kits in the table (as much as 60-fold lower), with a 
comparable detection limit to one ?ultrasensitive? human insulin ELISA. The dynamic range of ECPA 
is >40-fold wider than all ELISAs shown in the table (as much as 300-fold wider). These improvements 
shown by ECPA come with the added benefit of a direct electrochemical readout, i.e., without requiring 
washing steps. Abbreviations: ECPA = electrochemical proximity assay, ELISA = enzyme-linked 
immunosorbent assay, LOD = limit of detection. bTo provide valid method comparisons, LODs and 
dynamic ranges for ELISAs are defined by the concentrations in the incubation mixture, prior to the 
washing step and secondary antibody incubation. 
 
111 
 
 
4.7     Conclusions 
In this chapter, I described the development of the electrochemical proximity assay 
(ECPA), which leverages two aptamer or antibody-oligonucleotide probes and proximity-
dependent DNA hybridization to move a redox active molecule near a gold electrode. A 
DNA-based experimental model was used to optimize the assay format, and aptamer- and 
antibody-based ECPA were shown functional with high sensitivities and low detection 
limits, employing a short DNA competitor to limit background current. This background-
reduced ECPA was shown to match or outperform currently used ELISA kits for insulin 
detection. 
Of particular importance is the proof-of-concept provided by antibody-based ECPA. 
Judging from to the successes of other proximity immunoassays,12,13 it is reasonable to 
assume that ECPA should perform well in quantifying any other protein with an available 
antibody pair. Combining the assay?s flexibility and high sensitivity with the simplicity 
of direct electrochemical readout, ECPA should be useful in a variety of settings in the 
future, including medical diagnostics, biological research, and point-of-care testing. 
 
 
 
 
 
 
 
 
112 
 
 
 
4.8     References 
 
(1) Martinez, A.; Phillips, S. T.; Whitesides, G. M.; Carrilho, E. Anal. Chem. 2010, 
82, 3?10. 
(2) Burtis, C. A.; Ashwood, E. R. Tietz Textbook of Clinical Chemistry; Saunders: 
Philadelphia, PA, 1999. 
(3) Rusling, J. F.; Kumar, C. V.; Gutkind, J. S.; Patel, V. Analyst 2010, 135, 2496?
2511. 
(4) von Lode, P. Clin. Biochem. 2005, 38, 591?606. 
(5) Sorell Go?mez, L.; Rojas, G. Clin. Chim. Acta 1997, 260, 65?71. 
(6) Jung, K.; Zachow, J.; Lein, M.; Brux, B.; Sinha, P.; Lenk, S.; Schnorr, D.; 
Loening, S. A. Urology 1999, 53, 155?160. 
(7) Stivers, C. R.; Baddam, S. R.; Clark, A. L.; Ammirati, E. B.; Irvin, B. R.; Blatt, 
J. M. Diabetes Technol. Ther. 2000, 2, 517?526. 
(8) Davies, S; Byrn, F; Cole, L. A. Clin. Lab. Med. 2003, 23, 257?264. 
(9) Seamark, D. A.; Backhouse, S. N.; Powell, R. Ann. Clin. Biochem. 2003, 40, 
178?180. 
(10) Haab, B. B.; Dunham, M. J.; Brown, P. O. Genome Biol. 2001, 2, 1?5. 
(11) Fredriksson, S.; Gullberg, M.; Jarvius, J.; Olsson, C.; Pietras, K.; Gustafsdottir, 
S. M.; Ostman, A.; Landegren, U. Nat. Biotechnol. 2002, 20, 473?477. 
(12) Gullberg, M.; Gustafsdottir, S. M.; Schallmeiner, E.; Jarvius, J.; Bjarneg rd, 
M.; Betsholtz, C.; Landegren, U.; Fredriksson, S. Proc. Natl. Acad. Sci. U.S.A. 
2004, 101, 8420?8424. 
(13) Heyduk, E.; Dummit, B.; Chang, Y. H.; Heyduk, T. Anal. Chem. 2008, 80, 
5152?5159. 
(14) Konry, T.; Hayman, R. B.; Walt, D. R. Anal. Chem. 2009, 81, 5777?5782. 
(15) Kim, D.; Daniel, W. L.; Mirkin, C. A. Anal. Chem. 2009, 81, 9183?9187. 
(16) Luminex Corporation, Technologies and Science, 
http://www.luminexcorp.com/TechnologiesScience/index.htm; accessed 
December 2011. 
(17) Kim, J.; Hu, J.; Sollie, R. S.; Easley, C. J. Anal. Chem. 2010, 82, 6976?6982. 
(18) Osborne, S. E.; Ellington, A. D. Chem. Rev. 1997, 97, 349?370. 
(19) Famulok, M. Curr. Opin. Struct. Biol. 1999, 9, 324?329. 
(20) Drummond, T. G.; Hill, M. G.; Barton, J. K. Nat. Biotechnol. 2003, 21, 1192?
1199. 
(21) The Aptamer Handbook. Functional Oligonucleotides and Their Applications; 
Klussmann, S., Ed.; WILEY-VCH Verlag GmbH & Co. KGaA: Weinheim, 
Germany, 2006. 
(22) Herr, J. K.; Smith, J. E.; Medley, C. D.; Shangguan, D.; Tan, W. Anal. Chem. 
113 
 
 
2006, 78, 2918?2924. 
(23) Shangguan, D.; Li, Y.; Tang, Z.; Cao, Z. C.; Chen, H. W.; Mallikaratchy, P.; 
Sefah, K.; Yang, C. J.; Tan, W. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 11838
?11843. 
(24) Liu, Y.; Tuleouva, N.; Ramanculov, E.; Revzin, A. Anal. Chem. 2010, 82, 
8131?8136. 
(25) Hianik, T.; Wang, J. Electroanalysis 2009, 21, 1223-1235. 
(26) Sassolas, A.; Blum, L. J.; Leca-Bouvier, B. D. Electroanalysis 2009, 21, 1237-
1250. 
(27) Kelley, S. O.; Barton, J. K.; Jackson, N. M.; Hill, M. G. Bioconjug. Chem. 
1997, 8, 31?37. 
(28) Fan, C.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 
9134?9137. 
(29) Kang, D.; Zuo, X.; Yang, R.; Xia, F.; Plaxco, K. W.; White, R. J. Anal. Chem. 
2009, 81, 9109?9113. 
(30) Zhang, Y. L.; Huang, Y.; Jiang, J. H.; Shen, G. L.; Yu, R. Q. J. Am. Chem. Soc. 
2007, 129, 15448?15449. 
(31) Ferapontova, E. E.; Olsen, E. M.; Gothelf, K. V. J. Am. Chem. Soc. 2008, 130, 
4256?4258. 
(32) Zhang, Y. L.; Pang, P. F.; Jiang, J. H.; Shen, G. L.; Yu, R. Q. Electroanalysis 
2009, 21, 1327?1333. 
(33) Bonham, A. J.; Hsieh, K.; Ferguson, B. S.; Valle?e-Be?lisle, A.; Ricci, F.; Soh, 
H. T.; Plaxco, K. W. J. Am. Chem. Soc. 2012, 134, 3346?3348. 
(34) Valle?e-Be?lisle, A.; Ricci, F.; Plaxco, K. W. J. Am. Chem. Soc. 2012, 134, 
2876?2879. 
(35) Xia, F.; Zuo, X.; Yang, R.; White, R. J.; Xiao, Y.; Kang, D.; Gong, X.; Lubin, 
A. A.; Valle?e-Be?lisle, A.; Yuen, J. D.; Hsu, B. Y.; Plaxco, K. W. J. Am. Chem. 
Soc. 2010, 132, 8557?8559. 
(36) Rowe, A. A.; Chuh, K. N.; Lubin, A. A.; Miller, E. A.; Cook, B.; Hollis, D.; 
Plaxco, K. W. Anal. Chem. 2011, 83, 9462?9466. 
(37) Lubin, A. A.; Plaxco, K. W. Acc. Chem. Res. 2010, 43, 496?505. 
(38) Nkodo, A. E.; Garnier, J. M.; Tinland, B.; Ren, H.; Desruisseaux, C.; 
McCormick, L. C.; Drouin, G.; Slater, G. W. Electrophoresis 2001, 22, 2424?
2432. 
(39) Hu, J.; Easley, C. J. Analyst 2011, 136, 3461?3468. 
(40) Millipore Corportation, Human Insulin ELISA, 
http://www.millipore.com/catalogue/item/ezhi-14k; accessed December 2011. 
(41) Mercodia AB, Insulin ELISA, 
http://www.mercodia.se/index.php?page=productview2&prodId=9; accessed 
December 2011. 
(42) AbCam plc., Insulin Human ELISA Kit ? 1 ? 96 Well Plate (ab100578), 
http://www.abcam.com/Insulin-Human-ELISA-Kit-1-x-96-Well-Plate-
ab100578.html; accessed December 2011. 
114 
 
 
(43) Mercodia AB, Insulin, Ultrasensitive ELISA, 
http://www.mercodia.se/index.php?page=productview2&prodId=11; accessed 
December 2011. 
(44) Millipore Corporation, Rat/Mouse Insulin ELISA, 
http://www.millipore.com/catalogue/item/ezrmi-13k; accessed December 2011. 
(45) ALPCO Diagnostics, Mouse Ultrasensitive Insulin ELISA, 80-INSMSU-E01, 
http://www.alpco.com/products/Insulin_Ultrasensitive_Mouse_ELISA.aspx; 
accessed December 2011. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
115 
 
 
 
CHAPTER 5  
The Reusable Electrochemical Proximity Assay (ECPA) 
 
5.1     Abstract of ECPA 
We previously developed an electrochemical proximity assay (ECPA) format for 
direct detection of a wide variety of target proteins (Chapter 4). The first generation of 
ECPA was composed of two aptamers or antibody-oligo conjugates that form a 
cooperative complex only in the presence of target protein molecules. Electrochemical 
current was thus increased in proportion to protein concentration, and detection limits 
were in the fM range with direct readout.  
We have further improved ECPA towards reusable protein quantitation. 
Regeneration of the base DNA monolayer was accomplished enzymatically. First, we 
incorporated uracils (U) into selected probe strands, then these strands were 
enzymatically cleaved using a uracil-DNA excision mix. After reducing complementary 
bases of thiolated DNA, assembled on gold electrode surface, to methylene blue (MB)-
conjugated oligonucleotide, this allowed measurement time to be reduced 30-fold, from 
90 min to 3 min, without significant loss of sensitivity. A single electrode could be 
enzymatically regenerated and reused at least 17 times without loss of signal, even when 
detecting picomolar concentrations of thrombin or femtomolar concentrations of insulin. 
116 
 
 
Selectivity was again confirmed against IgE and insulin in aptamer-based, and against 
IGF-1 and C-peptide in antibody-based reusable ECPA systems. This technique was also 
proven useful for hormone secretion measurements in vitro from murine pancreatic islets, 
and it was validated for insulin quantitation in murine blood serum. The reusable format 
of ECPA is faster and easier to implement than standard ECPA, while maintaining high 
sensitivity and selectivity. It is feasible to envision that ECPA could be easily integrated 
with portable electronic devices and applied in point-of-care settings and that reusable 
ECPA could be expanded upon and used in clinical laboratories and basic research 
settings. 
 
5.2     Introduction 
Early medical diagnosis of critical health changes could enable the prevention of 
fatal events because it can ensure appropriate and timely care and treatment to patients.1-3 
Highly sensitive detection and accurate analysis of diseases-associated proteins is 
demanded for early diagnosis.4-6 Current standard detection methods such as Western 
blots and enzyme-linked immunosorbent assays (ELISA) are usually slow, cumbersome, 
laboratory-bound processes that require hours to days to obtain clinical information5. 
Point of care approaches could overcome those drawbacks and speed implementation of 
treatment processes and improve patient outcomes.3,7-11 
There are some challenges that must be surmounted from laboratory sensor 
development to point of care application.12,13 First, the technology must be sufficiently 
selective to reject false signals; second, it must operate continuously and cannot rely on 
117 
 
 
batch process steps such as separations; and third, it should be reversible. Reversibility is 
a very important factor, since it could greatly reduce the cost of the product.  
With the development of biosensing techniques, DNA based electrochemical sensors 
become more and more promising towards point of care applications because they are 
rapid, low-cost, suitable for miniaturized devices, and ideal for application in chemically 
complex environments such as blood serum.13,14 Types of designs that are published 
include electrochemical DNA (E-DNA), traditional sandwich/supersandwich assays, 
competition-type assays, electrochemical aptamer-based sensors (E-AB), double-
antibody sandwich assay, and aptamer-based sandwich assays.15-20 Generally, DNA based 
electrochemical sensors are engineered by attaching an electro-active reporter to one 
position of DNA sequence which transfers electrons to or from the electrode surface, thus 
inducing electrochemical signal change at a detector. The principle can be based on 
target-induced strand displacement mechanisms, analyte-induced conformational changes, 
or combinations with sandwich and competitive assays, by forcing redox-labeled DNA 
into proximity with the electrode surface.16   
Motivated by the point of care application,s the first generation of standard 
electrochemical proximity assay (ECPA) was successful developed.21 The sensor used 
the proximity effect22,23 to move an electrochemically active label, methylene blue (MB), 
closer to a gold electrode upon binding of two probes to a protein target with high 
sensitivity and selectivity. Using antibody-oligo probes in standard ECPA, the detection 
limit for insulin was 20 fM, which is lower than most commercially available ELISAs, 
with a much wider dynamic range. These improvements come with the added benefit of 
direct electrochemical readout. 
118 
 
 
Although standard ECPA has great achievements, the limitations are also obvious. 
First, it is not reusable, meaning that repeated preparation of the electrode and DNA 
monolayer assembly (many hours) is required after every measurement. Secondly, probes 
and target are needed to be incubated for a long time before measurement either in 
aptamer-based or antibody-based assays. The popular way to make a DNA-based surface 
assay reusable is using a water rinse.17,24,25 However, this reusable method has some 
deficiencies. First, it is only useful for small numbers of base pairs of dsDNA. If two 
DNA strands have a significant number of complementary bases?such as with hairpin 
structures or complex structures formed by proximity effect making three or more DNA 
strands binding together?a water rinse may not be a good way to impart reusability. 
Second, a water rinse results in a signal decrease, but a signal decrease is not sufficient 
basis to prove the sensor is regenerated, it can only explain the reporter molecules are not 
near the sensor surface; the dsDNA may not be denatured, or target molecules may still 
be attached on the surface.  Here, regeneration was accomplished enzymatically by 
incorporating uracils into selected probe DNA strands, where these strands could be 
cleaved into small pieces using a uracil-DNA excision mix. Following enzymatic 
cleavage, the leftover short DNA strands were less than six oligonucleotides, which will 
easily denature at 37 ?C , or even at room temperature. This reusable version of ECPA 
performed as well as standard ECPA, in categories such as optimized signal-to-
background ratios and an equal direct insulin detection limit.  Reusable ECPA even 
showed a 4-fold wider dynamic range than standard ECPA, meaning that it has >160-fold 
wider range than ELISAs, and we also proved its utility in endocrine tissue sampling tests 
119 
 
 
and blood serum tests. These achievements further prove the reusable ECPA should be 
useful in medical research or clinical measurements.  
 
5.3     Reagents and Experimental Methods 
5.3.1     Reagents and Materials 
All solutions were prepared with deionized, ultrafiltered water (Fisher Scientific). 
The following reagents were used as received: insulin antibodies (clones 3A6 and 8E2; 
Fitzgerald Industries), 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) 
(99.5%), tris-(2-carboxyethyl) phosphine hydrochloride (TCEP), (Sigma-Aldrich, St. 
Louis, MO), bovine serum albumin (BSA, 98%; EMD Chemicals Inc.), human thrombin, 
immunoglobulin E (IgE), and insulin (Sigma Aldrich). Methylene blue-conjugated DNA 
(MB-DNA) was purchased from Biosearch Technologies (Novato, CA), purified by RP-
HPLC. Oligonucleotides were obtained from Integrated DNA Technologies (IDT; 
Coralville, Iowa), with purity and yield confirmed by mass spectrometry and HPLC, 
respectively. Sequences (listed 5?  to 3? ) for aptamer based ECPA were as follows. 
Thrombin aptamer A (THRaptA): AGTCCGTGGTAGGGCAGGTTGGGGTGA-
CTTTTTTTTTTTTTTTTTATATTTTTTTTTTCTCGCGGAUUUGAACCCUAACG; 
Thrombin aptamer B (THRaptB): TAGGAAAAGGAGGAGGGTGGG-
ATTGGTGTGTGTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTGGTTGGTGTGGT
TGG. Sequences (listed 5?  to 3? ) for antibody-based ECPA were as follows. Insulin 
antibody arm 1 (AbArm1): /5AmMC6//iSp18/CCCACTTAAACCTC-
AATCCACGCGGAUUUGAACCCUAACG; Insulin antibody arm 2 (AbArm2): 
120 
 
 
TAGGAAAAGGAGGAGGGTGGCC-CACTTAAACCTCAATCCA/iSp18//3AmMC6/. 
Sequences of ssDNA strands used in the experimental model are given in Table 5-1. 
 
 
 
5.3.2     Preparation of Sensor Surface  
Reusable ECPA sensors for the model system, for thrombin detection, for insulin 
detection, for measurement of hormone secretion, and for insulin measurement of blood 
serum were fabricated using a gold working electrode (2 mm diameter, CH instruments, 
Inc., Austin, TX). Gold electrodes were polished using a previously reported protocol 
(JACS) with minor alterations. In brief, the gold electrodes were first polished with a 
slurry of 0.05 ?m aluminum oxide in water (Buehler, Lake Bluff, IL) followed by 
sonication in ethanol for 10 min. This was followed by pipetting fresh piranha solution 
(H2SO4/H2O2, 3:1) on gold surface for 10 min, rinsed with D. I. water. (Caution: piranha 
solution is dangerous to human health and should be used with extreme caution and 
handled only in small quantities). The gold electrodes were further cleaned by a series of 
 
Table 5-1. Single-Stranded DNA Sequences Used in the Reusable ECPA Model 
Systemsa 
 
 
 
121 
 
 
oxidative and reductive scans in 0.5 M H2SO4 as previously described. The cleaned gold 
electrodes were thoroughly washed with D. I. water and ethanol and dried under flowing 
nitrogen. 
The volume for preparation of electrode-immobilized probe is dependent upon the 
number of electrodes to be prepared and is intended to conserve reagents.  We will 
provide info for our conditions when preparing 7 electrodes.  14 ?L (2 ?L per electrode) 
of 200 ?M thiolated-DNA was mixed with 28 ?L (4 ?L per electrode) of 10 mM TCEP 
in a 1.5-mL microcentrifuge tubes.  This tube was incubated for 120 min at room 
temperature (21 ?C) in the dark for reduction of disulfide bonds in the thiolated-DNA.  
This solution was then diluted to a total volume of 1400 ?L in HEPES/NaClO4 buffer (10 
mM HEPES and 0.5 M NaClO4, pH 7.0)16,21  to a final concentration of 2 ?M thiolated 
DNA. 
The MB-DNA must also be reduced in a similar way, but less volume is needed.  To 
generate sufficient reduced MB-DNA for 20-30 measurements, 3 ?L of 200 ?M MB-
DNA was mixed with 6 ?L of 10 mM TCEP in a 1.5-mL microcentrifuge tubes.  This 
tube was incubated for 120 min at room temperature (21 ?C) in the dark for reduction of 
the MB-moiety of the MB-DNA.  This solution was then diluted to a total volume of 600 
?L in HEPES/NaClO4 buffer to a final concentration of 1 ?M MB-DNA. (Critical step: 
Thiolated-DNA should be reduced with TCEP immediately before immobilization on a 
clean electrode. MB-DNA is mildly light-sensitive, so it is preferable to perform 
reduction step in the dark.) Unless otherwise noted, all solutions used in the experiments 
to follow were carried out at pH 7.0. 
122 
 
 
For immobilization, the 
previously cleaned electrode 
was directly transferred to 
the reduced thiolated-DNA 
solution and Incubated 
overnight (~12-16 h) at room 
temperature in the dark. 
Following incubation and 
SAM formation, remove any 
excess probe DNA 
physically adsorbed on 
electrode surface via a room 
temperature-deionized water rinse (~20 s). The stream of deionized water should flow 
over the modified electrode. And then the modified electrode was immediately 
transferred to 3 mM 6-mercaptohexanol solution for 40 min at room temperature in the 
dark. (Figure 5-1) (Critical step: 6-Mercaptohexanol solution should be prepared 
immediately before use. It?s better to do it in the hood.)  
 
5.3.3     ECPA Probe Assembly and Electrochemical Measurements 
Electrochemical measurements were performed as described previously21 using an 
GAMRY Reference 600 electrochemistry workstation with a standard three-electrode 
configuration consisting of a Ag|AgCl(s)|KCl(sat) reference electrode (Bioanalytical 
Systems, Inc.), a homemade platinum gauze flag (0.77 cm2) counter electrode, and a gold 
 
 
 
Figure 5-1:  Cyclic voltammetric response of 
modified gold electrode before transferred to 
3 mM 6-mercaptohexanol solution (blue 
trace) and after (red trace). 
123 
 
 
working electrode. All potentials are reported relative to the saturated Ag|AgCl reference 
electrode. Electrochemical measurements were performed in HEPES/NaClO4 buffer 
using square wave voltammetry (SWV) with a 50 mV amplitude signal at a frequency of 
60 Hz, over the range from ?0.45 to 0.10 V versus Ag|AgCl reference.  
Model system. The electrode was modified as described above and was placed into a 
glass electrochemical cell with HEPES/NaClO4 buffer. For modeling signal, the sensor 
was immersed in 10 nM ECPA-loop and 15 nM MB- DNA in 500 ?L of HEPES/NaClO4 
buffer. For modeling background, the sensor was immersed in 15 nM MB-DNA in 500 
?L of HEPES/NaClO4 buffer. Both signal and background currents were measured at the 
3-min time point. 
Aptamer-Based reusable ECPA System. Seven gold working electrodes were used 
for signal measurement of each thrombin concentration. Thrombin of seven different 
concentrations (from 0 to 10 nM) was incubated with 10 nM THRaptA and 15 nM 
THRaptB in HEPES/NaClO4 buffer prior to measurements. The thrombin/aptamer 
incubations were then added into the glass electrochemical cell. Before conducting 
voltammetric measurements, the sensor surface was allowed to react with analytes for 3 
min. After measurement, each sensor was regenerated by immersing in 2 unit uracil DNA 
excision mix in 100 ?l uracil excision enzyme buffer (50 mM Tris-HCl (pH 9.0), 20 mM 
[NH4]2SO4 and 10 mM EDTA) for 7 min at 37 ?C . Removing the enzyme solution and 
cleaved DNA pieces physically adsorbed on electrode surface via a room temperature-
deionized water rinse (~20 s), the stream of deionized water should flow over the 
modified electrode, and then electrodes were dried by using nitrogen gas. After that, the 
124 
 
 
sensor was ready for next measurement. Regeneration and selectivity tests with other 
proteins (IgE, insulin, or BSA) were made under the same conditions. 
Antibody-Based Reusable ECPA System. Antibody-oligonucleotide conjugates were 
prepared using the same method as described.21 The measurements of insulin secretions 
from pancreatic islets were performed with our novel cyclic methodology (see Figure 5-
7). The sensor was incubated with 50 nM Ab1 (DNA conjugated antibody 1: 
AbArm1?3A6) for 3 min by dropping 10 ?L solution on the sensor surface, and then 
washed within HEPES/NaClO4 buffer for ~15-second. Next, the sensor was transferred to 
a tube with live pancreatic islets in solution (300?L), which was kept at 37 ?C  by heat 
block. After 3 min incubation in the sample solution, the electrode was washed by 
HEPES/NaClO4 buffer for ~15-second, and then was incubated with 50 nM Ab2 (DNA 
conjugated antibody 2: AbArm2?8E2) for 3 min by dropping 10 ?L solution on the gold 
surface. Finally, the sensor was transferred to 500 ?L of HEPES/NaClO4 buffer with 50 
nM MB-DNA. Both signal and background currents were measured at the 3-min time 
point. For insulin calibration, we also used seven gold working electrodes for signal 
measurement of each insulin concentration by adopting the same protocol as 
measurement of secretions from pancreatic islets. 
 
5.4     Principle of reusable ECPA 
The principle of the reusable electrochemical proximity assay (ECPA) is shown in 
Figure 5-2. The sensor is prepared by self-assembly of thiolated DNA strands onto a gold 
electrode via the alkanethiol moiety at the 5? terminus, and also treated by 3 mM 6-
mercaptohexanol solution, ensuring monolayer formed on the whole gold surface. The 
125 
 
 
five-part complex is shown in Figure 5-2A: thiolated DNA?Ab1?target 
protein?Ab2?MB conjugated DNA. Uracils are incorporated into Ab1 oligo and MB-
conjugated DNA, so that the Ab1 and MB conjugated DNA will be enzymatically 
cleaved using a uracil-DNA excision mix into three and two small DNA pieces, 
respectively. Thus, after rinsing, only thiolated-DNA and 6-mercapto-1-hexanol are left 
on the sensor surface. Uracil-DNA excision mix contains two enzymes, HK-UNG (Heat-
Killable Uracil N-Glycosylase [UNG]) and Endonuclease IV. HK-UNG cleaves the 
uracil base from a uracil-deoxynucleotide in any DNA, creating an abasic site at the 
location of dUTP incorporation. Endonuclease IV then cleaves the phosphodiester bond 
at this abasic site, generating a series of digestion fragments. (Figure 5-2B) 
 
 
 
 
 
 
 
126 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 5-2:   Principle of reusable ECPA. (A) Stepwise operation of the assay. 
Electrode was immersed into uracil enzyme solution, the selected probe 
strands were enzymatically cleaved by uracil-DNA excision mix, the base 
DNA monolayer was regenerated. (B) Illustration of uracil-DNA excision mix 
mechanism. 
 
 
 
 
 
127 
 
 
5.5     DNA-Based Experimental model of reusable ECPA 
The same ssDNA strands are utilized as in our previous work ,21 except DNA loop 
and MB-DNA strands incorporated with uracils (Table 5-1). In standard ECPA, the signal 
measurement time of DNA-based experimental model is 40 min. Here, the purpose of 
this experimental model is to shorten the measurement time. As we know, in standard 
ECPA, we compared the signal and background responses of the system with 5, 7, and 10 
complementary bases (G5, G7, and G10 strands), and G10-competitor system. The 
advantages and disadvantages of G5 system and G10-competitor system are as follows. 
In the G5 system, measurements could be made quickly, and the background current was 
reduced to baseline, but with a large signal reduction. In the G10-competitor system, 
there was a large signal current and a baseline level background; however, both signal 
and background currents were necessarily measured at the 40-min time point.  
In an attempt to increase signal and reduce the measurement time, a higher 
concentration of MB-DNA was adopted in G5, G7, and G10-competitor systems. 
Previously, in standard ECPA, the concentration of MB-DNA used was 15 nM, but here 
we utilized 50 nM MB-DNA. The hypothesis was that the higher amount of MB-DNA 
would not greatly affect the background current in the G5 system, due to the restriction of 
fewer base pairs where the binding between G5 and MB-DNA was very weak at room 
temperature. However, forsignal current, the proximity effect (discussed in our previous 
work with both PLA and ECPA21,23) should provide even more enhancement. Figure 5-3 
compares the signal and background responses of the G5, G7, and G10-competitor 
systems. As hypothesized, the background current of G5 system still kept at the baseline 
level while signal current was increased to the same current level as in G7 and G10 
128 
 
 
systems. For G7 and G10 systems, because of more complementary bases in the thiolated 
DNA than G5, the background current of G7 was above baseline level. Even with the 
short DNA competitor utilized in G10 system, at higher concentration of MB-DNA, both 
signal and background formed rapidly. After comparing G5, G7, and G10 systems, we 
choose G5 for further experiments, since G5 allowed a short measurement time without 
sacrificing signal current.  Furthermore, the G5 system is simpler to implement than the 
G10 system, since a short DNA competitor was not required. 
 
 
 
Figure 5-3:   Comparisons of the signal and background responses of the G5, 
G7, and G10 with C9 competitor systems in DNA-based model for reusable 
ECPA. Experiments confirmed that the background current of G5 system 
could keep at the baseline level, while the signal current could be increased at 
high concentration of MB-DNA. 
 
 
 
 
129 
 
 
5.6     Regeneration comparisons between uracil enzyme and several general 
methods 
Before making the decision of 
using the uracil excision enzymes as a 
regeneration reagent, we compared it 
with other methods such as water rinse 
and water or HEPES/NaClO4 buffer 
bath at 37 ?C . Figure 5-4 shows the 
comparison results. The red line in 
Figure 5-4 shows the signal current of 
10 nM thrombin measured with 10 nM 
THRaptA, 15 nM THRaptB and 50 nM 
MB-DNA in HEPES/NaClO4 buffer at 
the 3-min time point. The pink curve shows the result after a 7-min soak in 
HEPES/NaClO4 buffer at 37 ?C . The signal current went down to 80% of original value. 
In work by the Plaxco group, E-AB sensors could be regenerated by washing for ~30s in 
room temperature-deionized water24. However, this water flow method was not sufficient 
for our ECPA system. The regeneration result of water rinse in our experiment is shown 
by the blue curve, where the current value was about 35% of the original one, which 
means only about 65% of the five-part cooperative complex was denatured. The green 
curve shows the result of immersing the sensor in 37 ?C  warm water for 7 min, which can 
denature 80% of cooperative complex. The black line shows the electrode regeneration 
 
  
Figure 5-4: Regeneration comparisons 
between uracil enzyme and several 
general methods. 
 
 
 
 
 
 
130 
 
 
result using uracil-DNA excision mix, indicating that this approach is far superior to the 
other approaches for regenerating the biosensor surface in ECPA. 
 
5.7     Aptamer-Based reusable ECPA 
A schematic of aptamer-based reusable ECPA is shown in Figure 5-5A (inside 
figure). Two thrombin aptamers (THRaptA, THRaptB) that bind thrombin at different 
sites were applied as affinity probes. Using conditions optimized by the model system, 
background levels were measured in the absence of target protein (human thrombin). 
Similar to the model system, background remained at baseline current at the 3 min time 
point. Figure 5-5A shows the background with no thrombin (black trace) and typical MB 
oxidation peaks appearing at ?210 mV in the presence of various amount of thrombin 
measured at 3 min time point. Seven current signals of various concentration of thrombin 
(from 0 to 10 nM) were measured by seven sensors, which were prepared under the same 
condition. Each sensor was regenerated seven times, so signal currents of 7 different 
concentrations of thrombin were measured by every sensor. Figure 5-5B shows this 
reusable aptamer-based ECPA system calibrated versus thrombin concentration with each 
sensor response recorded as integrated MB peak areas from ?330 to ?100 mV. The curve 
with error bars is the average value of 7 sensors (dash line). It is noteworthy that 150 pM 
is clearly not at the detection limit of the reusable ECPA.  This result was also very 
promising in that electrode surface preparation proved to be highly consistent.  
Calibrations from seven differently-prepared electrodes were statistically equal. 
The next step was to test how many times the sensor can be regenerated with 
reusable ECPA. In order to ensure reliability, two sensors with two different 
131 
 
 
concentrations of thrombin were chosen to test this experiment. The signal of both 
sensors starts to decrease after 19 times regeneration by uracil-DNA excision enzyme 
(Figure 5-5C). 
To demonstrate specificity and sustainable signal reproducibility, the reusable ECPA 
was challenged with nonspecific proteins including human IgE, insulin, and BSA by 
using only one sensor. In Figure 5-5D, the signal of four measurements (first, second, 
ninth, and tenth) was 5 nM thrombin, the signal of four measurements (fifth, sixth, 
thirteenth, and fourteenth) was 5 nM thrombin in 3% BSA. Figure 5-5D shows that 
essentially no response was observed in the presence of 15 nM IgE (eleventh, twelfth) or 
insulin (fifteenth, sixteenth); even with 3-fold lower thrombin (5 nM), the signal was ~ 65 
? 95-fold larger than that of IgE or insulin. In addition, the signal of ninth and tenth 
measurements of thrombin was 94% of the first and second measurements, the same 
situation for 5 nM thrombin in 3% BSA, showing good signal recovery. Furthermore, 
baseline current was observed in the presence of 3% BSA, while the signal from 5 nM 
thrombin was recovered by 92% in 3% BSA. This result is encouraging for future 
application of reusable ECPA to continuous measurement of biological samples and 
possibly for point-of-care settings. 
 
 
 
 
 
 
132 
 
 
 
 
 
 
 
 
  
Figure 5-5:   Aptamer based reusable ECPA. (A) Current changes under 
different concentrations of thrombin in pM to nM range. (B) Seven 
different electrodes generated similar calibration curves. The thrombin 
limit of detection was 150 pM with reusable ECPA. (C) 1.25 and 5 nM 
thrombin were chosen to test regeneration. A single electrode can be 
reused 18 times without loss of signal. (D) A single electrode test shows 
high selectivity throughout regeneration steps.  
 
 
 
 
133 
 
 
5.8     Antibody-Based reusable ECPA 
The flexibility of the aptamer-based approach is limited because of the requirement 
of two aptamers for the target protein, since aptamer pairs exist only for a few select 
proteins. The use of antibody-oligonucleotide conjugates as probes can overcome this 
challenge.26,27 
A schematic of antibody-based reusable ECPA is shown in Figure 5-6A (inside 
figure). A different set of conditions were determined for this antibody-based assay.  We 
adopted a cyclic protocol, and insulin standard solutions were incubated at 37 ?C  to 
ensure relevance to secretion sample measurements from pancreatic islets. Since we used 
50 nM of each antibody-oligo and 50 nM MB-DNA, the assay showed a higher signal 
than the model system (10 nM loop and 50 nM MB-DNA). Figure 5-6A shows the 
background with no insulin (black trace) and typical MB oxidation peaks appearing at 
?210 mV in the presence of various amount of insulin measured at 3 min time point (at 
37 ?C). This antibody-based reusable ECPA system was then calibrated versus insulin 
concentration (Figure 5-6B), with response recorded in 7 sensors as integrated MB peak 
areas from ?330 to ?100 mV. Remarkably, using a direct electrochemical readout, ECPA 
was capable of detecting insulin levels as low as 100 fM with a dynamic range extending 
to 6.25 nM. The number of possible regenerations of antibody-based reusable ECPA is 
17 times (Figure 5-6C). The selectivity of antibody-based reusable ECPA was tested 
against insulin-like growth factor 1 (IGF-1), which has similar structure to insulin, and 
against C-peptide, which is co-secreted with insulin into the bloodstream. As expected, 
the sensor did not respond significantly to higher concentrations of either IGF-1 or C-
peptide (Figure 5-6D). 
134 
 
 
 
 
 
 
 
 
 
 
Figure 5-6:   Antibody based reusable ECPA. (A) Current changes under 
different concentrations of insulin in fM to nM range. (B) Seven different 
electrodes generated similar calibration curves. The insulin limit of 
detection was 100 fM with reusable ECPA. (C) 1.25 and 0.25 nM insulin 
were chosen to test regeneration. A single electrode can be reused 17 
times without loss of signal. (D) A single electrode test shows high 
selectivity throughout regeneration steps.  
 
 
 
 
 
135 
 
 
5.9     Temporal Measurement of Hormone Secretion from Endocrine Tissue  
In order to monitor secretions from pancreatic islets, we cannot add antibody-
oligonucleotide conjugates into the same solution with pancreatic islets as we did in the 
aptamer-based reusable ECPA system. First, the insulin antibodies have been shown to 
bind pancreatic islets, not just insulin in solution, which can affect the function of the 
cells.  Second, we want to continuously monitor the secretion changes from pancreatic 
islets in real-time, not just make a one-time measurement. A schematic of the cyclic 
methodology of antibody-based reusable ECPA used for measurement of secretions from 
pancreatic islets is shown in Figure 5-7A. In this cycle, 10 ?L of oligo-modified Ab1 
solution was first dropped on the gold surface of the pre-fabricated sensor and incubated 
for 3 min. After washing, the sensor was loaded into a tube containing pancreatic islets at 
37 ?C  (using a heat block). At this time, some of the secreted insulin molecules would 
bind with Ab1, staying on the sensor surface.  Next, the electrode was washed briefly in 
HEPES/NaClO4 buffer. The third step included pipetting 10 ?L of oligo-modified Ab2 
solution onto the sensor surface and incubating for 3 min.  The sensor was then placed 
into 50 nM MB-DNA solution, at which time the signal current was measured after 3 min. 
Finally, after measurement, the sensor was enzymatically regenerated in uracil-DNA 
excision mix solution to prepare for the next time measurement.  
Figure 5-7B shows quantitation of insulin secretion from living pancreatic islets 
(two different mice) using antibody-based reusable ECPA, during basal (3 mM) and 
stimulatory (11 mM) glucose levels. The reported signal here is integrated current change 
versus time.  Sensor responses were recorded as integrated MB peak areas from ?330 to 
?100 mV, then each point was subtracted from its subsequent measurement in time. The 
136 
 
 
first 3 measurements were made at 37 ?C  with islets experiencing basal glucose levels (3 
mM). Immediately following the 3rd measurement, glucose was increased to 11 mM. As 
expected, an increase was observed on the 4th measurement after changing from 3 to 11 
mM glucose. This ?spike? in insulin was expected and is shown as the third current 
difference data point in all data sets shown in Figure 5-7B.  With the group of 5 islets, the 
integrated signal increased from 132 to 212 nA(80 nA spike). With the groups of 15 islets, 
the integrated signal spike was from 266 to 505 nA (239 nA spike) and 272 to 491 nA 
(219 nA spike), respectively. The integrated signal current of control measurements of 50 
pM insulin in the buffer with 3 mM glucose was 239 nA, which falls between the signals 
measured from 5 islets and 15 islets. 
 
 
 
 
 
 
 
 
 
 
 
 
 
137 
 
 
 
 
 
 
 
 
 
 
 
Figure 5-7:   Temporal measurement of hormone secretion from endocrine 
tissue via reusable ECPA. (A) This cyclic protocol allowed insulin secretion 
measurements from pancreatic islets at 37 ?C. (B) Insulin secretion 
measurements by antibody based reusable ECPA from living pancreatic islets 
(2 different mice), during basal (3 mM) and stimulatory (11 mM) glucose 
levels. 
 
 
Cycle time 
~ 5 ? 10-min 
138 
 
 
5.10     Measurement of Insulin from Murine Blood Serum 
The same cyclic methodology as used in measurement of hormone secretions was 
also adopted in the insulin measurement from murine blood serum. Since various 
enzymes are present in serum, especially DNases, we need to inhibit the enzymes? 
activities to avoid degradation of DNA strands used in our assay.  For this reason, the 
second step of loading the sensor into the blood serum sample in the cyclic protocol was 
performed on ice, keeping the temperature around 0 ?C . Figure 5-8A shows that a single 
electrode can be reused ~9 times without loss of signal in 20-fold dilution of blood serum. 
Figure 5-8B shows insulin quantitation in blood serum from two different mice 
(C57BL/6J), in 20-fold diluted serum samples. The concentration of insulin in blood 
serum was measured by standard addition of fixed amounts of insulin into the serum 
sample. This way, the measured insulin concentrations were 120 and 140 pM, separately 
for the two blood samples. The published result of insulin concentration in the blood of 
normal C57BL/6J mice was 16.4 ? 4.5 mU/ml (113.8 ? 31.2 pM).28 This result proved 
that our reusable ECPA approach could be used successfully to quantify proteins in 
complex biological samples such as serum.  The results bode well for the possible use of 
ECPA in clinical research or integrated with medical devices applied in point-of-care 
settings. 
 
 
 
 
 
139 
 
 
 
 
 
5.11     Characterization by surface plasmon resonance 
We have begun characterization of our electrochemical proximity assay with surface 
plasmon resonance (SPR). SPR has several advantages that the electrochemical detection 
method does not have, such as real-time sensing and label free nature. As noted, to make 
our ECPA biosensor, we first need to prepare DNA monolayers on the gold electrode 
surface. We hypothesized that monolayer formation of thiolated-DNA on gold surface 
would be faster in a flowing system. Using an SPR system with our collaborator, Dr. 
 
  
Figure 5-8:   Insulin measurements from murine blood serum. (A) single 
electrode can be reused 9 times without loss of signal in 20-fold dilution 
of blood serum performed on ice (0 - 4 ? C) (B) Insulin measurements of 
blood serum from two different mice (C57BL/6J), 20-fold dilution in 
HEPES/NaClO4 buffer. The measured insulin concentrations were 120 
and 140 pM separately. 
 
 
 
 
 
 
140 
 
 
Simonian (Materials Engineering, Auburn University), the flow rate was first set at 2 ?L 
min-1 for 60 min of continuous flow of running buffer through flow cell for stabilization 
of the gold surface. When the signal of gold surface was stable, we then switched to the 
sample channel, and the thiolated-DNA solution was pumped into the flow cell. Figure 5-
9A shows real-time interrogation of self-assembled monolayer (SAM) formation at the 
gold surface at different concentrations of thiolated-DNA. As we expected, the self-
assembled monolayer was formed faster than that under static mode. Usually in ECPA, 
for DNA monolayer assembly on the gold electrode surface, the electrode was incubated 
for 16 h at room temperature in the dark, since optimization experiments had not been 
peformed. In Figure 5-9A, after 500 s continuous flow of 1 ?M thiolated-DNA, 95% of 
gold surface was covered. For 0.3, 0.1, and 0.03 ?M thiolated-DNA, coverage required 
2000 s, 2500 s, and 5000 s respectively. For saving reagents yet maintaining rapid SAM 
formation, we chose the 0.1 ?M thiolated-DNA for future monolayer formation. The total 
amount of thiolated-DNA used to cover 95% of the surface was ~ 8.3 pmol at 2 ?L min-1 
flow rate.  However, the total amount of thiolated-DNA used in ECPA was 200 pmol. In 
the experiments shown by Figure 5-9B, 0.1 ?M thiolated-DNA was flowed through first 
to form a monolayer on the gold surface, and after reaching equilibrium, we switched to 
the buffer channel for rinsing physically absorbed thiolated-DNA.  This rinsing resulted 
in aslightly reduced response. The DNA Loop model of ECPA signal, consisting of 15 
nM loop and 15 nM MB-DNA, was then added. The figure clearly shows the response 
values increased in the presence of the loop model solution. These preliminary results are 
encouraging for future experiments in optimizing the timing and kinetics of ECPA, as 
well as for fundamental characterization of the electrode surface preparation in ECPA. 
141 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
 
Figure 5-9:   Characterization of ECPA with surface plasmon resonance 
(SPR). (A) The figure shows real-time interrogation of self-assembled 
monolayer (SAM) formation at the gold surface at different 
concentrations of thiolated-DNA. (B) Real-time sensing of SAM forming, 
buffer rinsing, and loop model binding with thiolated-DNA. 
 
 
 
 
 
142 
 
 
5.12     Conclusions 
In this chapter, I described the development of reusable ECPA, which overcomes 
limitations of standard ECPA, namely the slow measurement time and the slow electrode 
regeneration steps. A DNA-based experimental model was used to optimize the assay 
format, and uracils were incorporated into selected probe strands, making the assay 
reusable after a simple enzymatic surface ?cleaining.? Aptamer-based reusable ECPA 
clearly showed faster measurement without significant loss of sensitivity, and selectivity 
was again confirmed for thrombin against IgE, insulin, and BSA. Furthermore, the faster 
measurement with antibody-based reusable ECPA made it possible to monitor secretions 
from pancreatic islets in near real-time. Finally, insulin quantitation in murine blood 
serum further proved this assay could be highly useful in clinical research. 
Due to its high sensitivity, direct-readout format, and flexibility to assay multiple 
targets, ECPA has good potential for quantitation of a variety of proteins in clinical 
laboratories, research laboratories, or at the point-of-care.  Reusable ECPA is faster and 
easier to implement than standard ECPA, while maintaining high sensitivity and 
selectivity. The reusable version of ECPA should be more useful in research settings, 
such as secretion sampling or high-throughput screening applications.   
 
 
 
 
 
 
143 
 
 
 
5.13     References 
 
(1) Hay Burgess, D.; Wasserman, J.; Dahl, C. Nature 2006, 444, 1-2. 
(2) Yager, P.; Domingo, G. J.; Gerdes, J. Annual Review of Biomedical 
Engineering 2008, 10, 107-144. 
(3) Martinez, A.; Phillips, S. T.; Whitesides, G. M.; Carrilho, E. Anal. Chem. 2010, 
82, 3?10. 
(4) Burtis, C. A.; Ashwood, E. R. Tietz Textbook of Clinical Chemistry; Saunders: 
Philadelphia, PA, 1999. 
(5) Rusling, J. F.; Kumar, C. V.; Gutkind, J. S.; Patel, V. Analyst 2010, 135, 2496?
2511. 
(6) Wang, J. Biosensors and Bioelectronics 2006, 21, 1887-1892. 
(7) Sorell Go?mez, L.; Rojas, G. Clin. Chim. Acta 1997, 260, 65?71. 
(8) Jung, K.; Zachow, J.; Lein, M.; Brux, B.; Sinha, P.; Lenk, S.; Schnorr, D.; 
Loening, S. A. Urology 1999, 53, 155?160.    
(9) Stivers, C. R.; Baddam, S. R.; Clark, A. L.; Ammirati, E. B.; Irvin, B. R.; Blatt, 
J. M. Diabetes Technol. Ther. 2000, 2, 517?526. 
(10) Davies, S; Byrn, F; Cole, L. A. Clin. Lab. Med. 2003, 23, 257?264. 
(11) Seamark, D. A.; Backhouse, S. N.; Powell, R. Ann. Clin. Biochem. 2003, 40, 
178?180. 
(12) Drummond, T.; Hill, M. G.; Barton, J. K. Nature Biotechnol. 2003, 21, 1192-
1199. 
(13) Plaxco, K. W.; Soh, H. T. Trends in Biotechnology 2011, 29, 1-5. 
(14) Cagnin, S.; Caraballo, M.; Guiducci, C.; Martini, P.; Ross, M.; SantaAna, M.; 
Danley, D.; West, T.; Lanfranchi, G. Sensors 2009, 9, 3122-3148. 
(15) Xia, F.; White, R. J.; Zuo, X.; Patterson, A.; Xiao, Y.; Kang, D.; Gong, X.; 
Plaxco, K. W.; Heeger, A. J. J. Am. Chem. Soc. 2010, 132, 14346-14348. 
(16) Kang, D.; Zuo, X.; Yang, R.; Xia, F.; Plaxco, K. W.; White, R. J. Anal. Chem. 
2009, 81, 9109-9113. 
(17) Zuo, X.; Xiao, Y.; Plaxco, K. W. J. Am. Chem. Soc. 2009, 131, 6944-6945. 
(18) Baker, B. R.; Lai, R. Y.; Wood, M. S.; Doctor, E. H.; Heeger, A. J.; Plaxco, K. 
W. J. Am. Chem. Soc. 2006, 128, 3138-3139. 
(19) Xiao, Y.; Lubin, A. A.; Baker, B. R.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. 
Acad. Sci. USA 2006, 103, 16677-16680. 
(20) Liu, Y.; Tuleouva, N.; Ramanculov, E.; Revzin, A. Anal. Chem. 2010, 82, 
8131?8136. 
(21) Hu, J.; Wang, T.; Kim, J.; Shannon, C.; Easley, C. J. J. Am. Chem. Soc. 2012, 
134, 7066-7072. 
(22) Fredriksson, S.; Gullberg, M.; Jarvius, J.; Olsson, C.; Pietras, K.; G?stafsd?ttir, 
S. M.; ?stman, A.; Landegren, U. Nature Biotechnol. 2002, 20, 473-477. 
(23) Kim, J.; Hu, J.; Sollie, R. S.; Easley, C. J. Anal. Chem. 2010, 82, 6976-6982. 
144 
 
 
(24) Xiao, Y.; Lai, R. Y.; Plaxco, K. W. Nat. Protocols 2007, 2, 2875-2880. 
(25) Rowe, A. A.; Chuh, K. N.; Lubin, A. A.; Miller, E. A.; Cook, B.; Hollis, D.; 
Plaxco, K. W. Anal. Chem. 2011, 83, 9462-9466. 
(26) Gullberg, M.  Gustafsdottir, S. M.  Schallmeiner, E.   arvius,  .  Bjarneg rd, 
M.; Betsholtz, C.; Landegren, U.; Fredriksson, S. Proc. Natl. Acad. Sci. U.S.A. 
2004, 101, 8420?8424. 
(27) Heyduk, E.; Dummit, B.; Chang, Y. H.; Heyduk, T. Anal. Chem. 2008, 80, 
5152?5159. 
(28) Surwit, R. S.; Kuhn, C. M.; Cochrane, C.; McCubbin, J. A.; Feinglos, M. N. 
Diabetes 1988, 37, 1163-1167. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
145 
 
 
 
CHAPTER 6  
Future Directions 
 
6.1     Future Portable Electronic Device 
Based on the electrochemical proximity assay (ECPA) presented in Chapters 4 and 5, 
it is feasible to envision that ECPA could be integrated with portable electronic devices 
and applied in point-of-care settings. In terms of protein quantitation, a standard 
antibody-based ECPA has been proven to detect insulin as low as 20 fM by introducing a 
short DNA competitor. The reusable version of ECPA has proven that a single electrode 
can be regenerated and reused 18 times without loss of signal. Selectivity tests based on 
thrombin aptamers in reusable ECPA again showed high selectivity and high signal 
recovery.  The signal of 5 nM thrombin was ~ 65 to 100-fold larger than that of 15 nM 
IgE or insulin. And the signal of 5 nM thrombin was recovered by over 90% in 3% BSA. 
These results strongly encouraged future applications to biological samples and blood 
serum tests. It was shown that it is possible to build a cyclic protocol for insulin secretion 
measurements from pancreatic islets on a chip. With this in mind, Figure 6-1 shows a 
proposed integrated electronic device including 10 sensor sets, each set containing one 
reference electrode, one counter electrode, and three gold working electrodes. Each set 
stays in a reservoir for 3 min, then is clockwise rotated to the next reservoir. Based on the 
146 
 
 
result of regeneration test from aptamer based reusable ECPA, a single electrode could be 
regenerated and reused for 18 times. Thus, the proposed chip could complete 180 
individual measurements, each results being an average of signal from the 3 electrodes 
shown in the inset at the top right of the figure. 
 
 
  
Figure 6-1:  Future portable electronic device. The bottom left figure shows 
10 sensor sets built on a chip, each set containing one reference, one 
counter, and three working electrodes. P1R is probe1 reservoir, TR is 
sample solution reservoir, P2R is probe2 reservoir, MR is MB-DNA 
solution reservoir, ER is uracil-DNA excision mix reservoir, WR is washing 
buffer reservoir. The chip mimics the cyclic protocol (bottom right).  
 
 
 
147 
 
 
6.2     Prospect of Detection Technology 
Besides integrated with portable electronic devices, the reusable version of ECPA 
could also be useful in high-throughput screening applications, such as conjugating with 
HPLC fraction collectors or for drug discovery applications using 96-well plates with 
cells treated with different drugs (Figure 6-2).  
 
 
 
 
 
 
 
Figure 6-2:  Reusable ECPA integration with high throughput well-plate 
formats: P1 (probe1 well), T (collection sample wells), P2 (probe2 wells), 
MB (MB-DNA & signal measurement wells), En (uracil-DNA excision mix 
wells), W1 ? W5 (washing buffer wells).