Point-of-need detection using surface-based biosensors with an examination of protein immobilization and development of magnetic labels

Over the past decade, our research group has worked on developing surface-based immunoassays to detect disease biomarkers. Our immunoassay platforms use a gold surface coated with an N-hydroxysuccinimide (NHS)-based monolayer and a layer of antibodies to capture a target antigen. Readout is achieved by surface-enhanced Raman scattering (SERS) or giant magnetoresistance (GMR) after labeling of the captured antigen with Raman dye-modified gold nanoparticles or magnetic particles, which are also coated with antibodies. Both of these platforms enable the low-level detection of numerous biomarkers and have the potential for translation into a point-of-need (PON) (i.e., rapid, easy to use, and field deployable) test. As part of an effort to develop a PON test, this dissertation includes investigations of: (1) SERS-based detection of botulinum neurotoxins (BoNTs), (2) protein immobilization procedures, and (3) magnetic microcapsules (MMCs) for use with GMR detection. First, a SERS-based immunoassay for bioterrorism agents, botulinum neurotoxins A (BoNT-A) and B (BoNT-B) with picomolar (or lower) detection limits for BoNT-A and; BoNT-B in buffer and serum is described. These results not only demonstrate sufficient detection of these markers at levels important to homeland security and human health monitoring, but also the potential to translate this methodology to a PON test. Next, the reactivity of NHS ester-terminated monolayers, a common approach in protein immobilization chemistry, is investigated to assess the competition of the purported amidization reaction to that of hydrolysis. Results of kinetic studies on hydrolysis and aminolysis under relevant assay conditions show the rate of hydrolysis is 300× faster than that of aminolysis. These results indicate that it is highly unlikely that proteins are covalently linked to the surface and suggest that the protein layer is adsorbed via hydrophobic, hydrogen bonding, and electrostatic interactions. The last section examines the development of an MMC-based label. With marked improvement in both stability and magnetization over commercially-available magnetic nanoparticles, these MMCs show potential for the eventual enhanced function as a label in a GMR-based immunoassay. With these results, this dissertation aims to set the stage for the rational development of assays that will facilitate a paradigm shift towards PON tests.

POINT-OF-NEED DETECTION USING SURFACE-BASED BIOSENSORS WITH AN EXAMINATION OF PROTEIN IMMOBILIZATION AND DEVELOPMENT OF MAGNETIC LABELS by China Ye-Ling Lim A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemical Engineering The University of Utah December 2016 Copyright © China Ye-Ling Lim 2016 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of China Ye-Ling Lim has been approved by the following supervisory committee members: Marc D. Porter , Chair 10/26/2016 Mikhail Skliar , Member 10/26/2016 James N. Herron , Member 10/26/2016 Terry A. Ring , Member 10/26/2016 Jennifer S. Shumaker-Parry , Member 10/26/2016 and by the Department/College/School of Date Approved Date Approved Date Approved Date Approved Date Approved , Chair/Dean of Milind D. Deo Chemical Engineering and by David B. Kieda, Dean of The Graduate School. ABSTRACT Over the past decade, our research group has worked on developing surface-based immunoassays to detect disease biomarkers. Our immunoassay platforms use a gold surface coated with an N-hydroxysuccinimide (NHS)-based monolayer and a layer of antibodies to capture a target antigen. Readout is achieved by surface-enhanced Raman scattering (SERS) or giant magnetoresistance (GMR) after labeling of the captured antigen with Raman dye-modified gold nanoparticles or magnetic particles, which are also coated with antibodies. Both of these platforms enable the low-level detection of numerous biomarkers and have the potential for translation into a point-of-need (PON) (i.e., rapid, easy to use, and field deployable) test. As part of an effort to develop a PON test, this dissertation includes investigations of: (1) SERS-based detection of botulinum neurotoxins (BoNTs), (2) protein immobilization procedures, and (3) magnetic microcapsules (MMCs) for use with GMR detection. First, a SERS-based immunoassay for bioterrorism agents, botulinum neurotoxins A (BoNT-A) and B (BoNT-B) with picomolar (or lower) detection limits for BoNT-A and BoNT-B in buffer and serum is described. These results not only demonstrate sufficient detection of these markers at levels important to homeland security and human health monitoring, but also the potential to translate this methodology to a PON test. Next, the reactivity of NHS ester-terminated monolayers, a common approach in protein immobilization chemistry, is investigated to assess the competition of the purported amidization reaction to that of hydrolysis. Results of kinetic studies on hydrolysis and aminolysis under relevant assay conditions show the rate of hydrolysis is 300× faster than that of aminolysis. These results indicate that it is highly unlikely that proteins are covalently linked to the surface and suggest that the protein layer is adsorbed via hydrophobic, hydrogen bonding, and electrostatic interactions. The last section examines the development of an MMC-based label. With marked improvement in both stability and magnetization over commercially-available magnetic nanoparticles, these MMCs show potential for the eventual enhanced function as a label in a GMR-based immunoassay. With these results, this dissertation aims to set the stage for the rational development of assays that will facilitate a paradigm shift towards PON tests. iv "We are the music makers, And we are the dreamers of dreams." -Arthur O'Shaughnessy TABLE OF CONTENTS ABSTRACT....................................................................................................................... iii LIST OF TABLES ............................................................................................................. ix ACKNOWLEDGMENTS ................................................................................................. xi Chapters 1. INTRODUCTION .........................................................................................................1 1.1 Overview ............................................................................................................1 1.2 Botulinum Neurotoxins (BoNTs) ......................................................................2 1.3 Point-of-Need (PON) Test Considerations ........................................................5 1.4 Molecular Diagnostics and Solid-Phase Immunoassays (SPIs).........................7 1.5 Surface-Enhanced Raman Scattering (SERS) Immunoassay ............................9 1.6 Giant Magnetoresistance (GMR) Immunoassay ..............................................14 1.7 Protein Immobilization in Immunoassays .......................................................17 1.8 Protein Immobilization via NHS Chemistry ....................................................18 1.9 Interfacial Reactions of Spontaneously Adsorbed Monolayers (SAMs) .........21 1.10 Dissertation Summary....................................................................................24 1.11 References ......................................................................................................26 2. SERS-BASED DETECTION OF CLOSTRIDIUM BOTULINUM NEUROTOXINS A AND B TOWARD THE DEVELOPMENT OF A BIODEFENSE POINT-OFNEED TEST PLATFORM ..........................................................................................35 2.1 Introduction ......................................................................................................35 2.2 Experimental ....................................................................................................39 2.2.1 Reagents and Materials .........................................................................39 2.2.2 Capture Substrate Preparation ...............................................................40 2.2.3 Extrinsic Raman Label (ERL) Preparation ...........................................41 2.2.4 Immunoassay Procedure........................................................................41 2.2.5 Raman Instrumentation .........................................................................42 2.3 Results and Discussion ....................................................................................42 2.3.1 Assay Tuning .........................................................................................43 2.3.2 Dose Response Plots for BoNT-A and BoNT-B in PBS.......................51 2.3.3 Dose Response Plots for BoNT-A and BoNT-B in Human Serum ......54 2.4 Conclusions ......................................................................................................54 2.5 References ........................................................................................................57 3. SUCCINIMIDYL ESTER SURFACE CHEMISTRY: IMPLICATIONS OF THE COMPETITION BETWEEN AMINOLYSIS AND HYDROLYSIS ON COVALENT PROTEIN IMMOBILIZATION ...........................................................61 3.1 Introduction ......................................................................................................61 3.2 Experimental ....................................................................................................65 3.2.1 Reagents and Materials .........................................................................65 3.2.2 Monolayer Preparation ..........................................................................66 3.2.3 Infrared Spectroscopy (IRS) ..................................................................66 3.2.4 X-ray Photoelectron Spectroscopy (XPS) .............................................66 3.2.5 Electrochemistry....................................................................................67 3.2.6 UV-Vis Spectroscopy ............................................................................67 3.2.7 Contact Angle Measurements ...............................................................68 3.2.8 Adlayer Surface Concentration by XPS ................................................68 3.2.9 Interfacial Kinetics ................................................................................69 3.3 Results and Discussion ....................................................................................69 3.3.1 Characterization of As-Formed Adlayer ...............................................70 3.3.2 Compositional Analysis of Reacted Adlayers .......................................77 3.3.3 Adlayer Base Hydrolysis .......................................................................78 3.3.4 Homogeneous Base Hydrolysis.............................................................84 3.3.5 Adlayer Aminolysis ...............................................................................87 3.3.6 Implications of the Kinetic Measurements on the Aminolytic Immobilization of Protein .....................................................................91 3.4 Conclusions ......................................................................................................93 3.5 References ........................................................................................................94 4. HYDROLYSIS AND AMINOLYSIS OF SUCCINIMIDYL ESTER MONOLAYERS: IMPACT OF METHYLENE CHAIN-LENGTH AND SOLUTION COMPOSITION ON EFFICIENT PROTEIN IMMOBILIZATION ....99 4.1 Introduction ......................................................................................................99 4.2 Experimental ..................................................................................................102 4.2.1 Reagents and Materials .......................................................................102 4.2.2 Adlayer Preparation .............................................................................102 4.2.3 Infrared Spectroscopy Instrumentation ...............................................103 4.2.4 Interfacial Kinetic Studies ...................................................................103 4.3 Results and Discussion ..................................................................................104 4.3.1 Characterization of NHS-terminated Monolayers by IR-ERS ............105 4.3.2 Reaction Products for the Hydrolysis and Aminolysis of NHSterminated Monolayers ........................................................................109 4.3.3 Hydrolysis Reaction Kinetics of DSP-based (short, n=2) and DSUbased (long, n=10) Monolayers by IR-ERS ........................................111 vii 4.3.4 Thermodynamic Activation Parameters of DSP- and DSU-based Monolayers ..........................................................................................116 4.3.5 Hydrolysis of DSP-based Monolayers in NaOH, borate buffer, and phosphate buffer ..................................................................................122 4.3.6 Aminolysis of NHS-terminated Monolayers .......................................124 4.3.7 Kinetic Implications of Hydrolysis Conditions on Protein Immobilization ....................................................................................128 4.4 Conclusions ....................................................................................................129 4.5 References ......................................................................................................130 5. LAYER-BY-LAYER MAGNETIC POLYELECTROLYTE MICROCAPSULES AS IMMUNOASSAY LABELS ...............................................................................134 5.1 Introduction ....................................................................................................134 5.2 Experimental ..................................................................................................139 5.2.1 Reagents and Materials .......................................................................139 5.2.2 Magnetic Nanoparticle (MNP) Synthesis............................................140 5.2.3 Magnetic Microcapsule (MMC) Synthesis .........................................140 5.2.4 MC Characterization by Zeta (ζ) Potential, Raman Spectroscopy, and Atomic Force Microscopy (AFM) ................................................141 5.2.5 MMC Characterization by Scanning Electron Microscopy (SEM), Optical Microscopy, and Vibrating Sample Magnetometry ...............142 5.2.6 Calculating 𝜌𝜌𝑝𝑝 for Dynabeads, MCs, and MMCs ...............................142 5.2.7 Giant Magnetoresistance (GMR) Coupon Design, Sensor Design, and Magnetic Detection Station ..........................................................143 5.3 Results and Discussion ..................................................................................145 5.3.1 Characterization of MC Formation by ζ Potential, Raman Spectroscopy, and AFM ...............................................................................146 5.3.2 Characterization of MMCs by SEM and Optical Microscopy ............153 5.3.3 Sedimentation Calculations for MMCs by the Stokes and MasonWeaver Equations................................................................................155 5.3.4 Characterization of MMCs by VSM ...................................................163 5.3.5 MMCs as Immunoassay Labels ..........................................................166 5.4 Conclusions ....................................................................................................168 5.5 References ......................................................................................................169 6. CONCLUSION ..........................................................................................................173 APPENDIX: SUPPORTING INFORMATION FOR CHAPTER 4 ..............................178 viii LIST OF TABLES 3.1. Infrared spectral peak positions and band assignments for DSP and NHS dispersed in KBr and for the DSP-based adlayer on gold .............................................................72 3.2. XPS band assignments and positions for as-prepared and reacted DSP-based monolayers ................................................................................................................75 3.3. Infrared spectral peak positions and band assignments for aminolysis reaction products of the DSP-based monolayer ......................................................................89 4.1. IR spectral peak positions and band assignments for DSP-, DSH-, DSO-, and DSUbased adlayers on gold ............................................................................................107 4.2. Activation parameters for the hydrolysis of DSP- and DSU-based monolayers on gold by immersion in 50 mM BB (pH 8.50) ...................................................................120 4.3. Heterogeneous hydrolysis rate constants, 𝑘𝑘ℎ , for the hydrolysis of DSP-based adlayers in sodium hydroxide (NaOH), borate buffer (BB), and phosphate-buffered saline (PBS) ............................................................................................................123 5.1. Calculated sedimentation velocities, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , per Equation 5.1. For labels with multiple components, the particle density is approximated by volume fractions, 𝑉𝑉𝑓𝑓 , to obtain the apparent particle density, 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 .......................................................................144 5.2. Raman band assignments for PAH and PSS deposited on PS beads ......................150 5.3. Calculated sedimentation velocities, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , per Equation 5.1. For labels with multiple components, the particle density, 𝜌𝜌𝑝𝑝 , is approximated by volume fractions, 𝑉𝑉𝑓𝑓 , to obtain the apparent particle density, 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 , Table 5.1 ............................................157 5.4. Magnetization parameters for MMCs, in-house PEI-coated MNPs, and Dynabeads ...............................................................................................................165 A.1. XPS band assignments and positions for as-prepared DSP-, DSH-, DSO-, and DSUbased adlayers on gold .........................................................................................180 A.2. IR spectral peak positions and band assignments for reaction products of DSP-, DSH-, DSO-, and DSU-based adlayers on gold after 1 h immersion in hydrolysis (50 mM BB, pH 8.50) and aminolysis (500 mM EA in NaOH, pH 8.50) solutions ...............................................................................................................185 A.3. IR-ERS band positions of C-H stretches in high energy region (3000 to 2800 cm1) for DSP-, DSH-, DSO-, and DSU-based adlayers before (as-is) and after hydrolysis (hydro-) or aminolysis (amino-) by immersion in borate buffer (50 mM, pH 8.50) or ethylamine (500 mM, pH 8.5), respectively.....................................186 A.4. Student t values for thermodynamic activation parameters for the hydrolysis of DSP- and DSU-based monolayers. The calculated t values are compared to a table of critical values (e.g., for n=16, t=0.865, 1.337, 1.746, 2.120, 2.583, and 2.921 for confidence levels of 60%, 80%, 90%, 95%, 98%, and 99%, respectively) to determine to what significance level (i.e., a confidence level of 99% would give a significance level of 1%) the two values are different .........................................187 x ACKNOWLEDGMENTS First and foremost, I would like to thank all of my friends and family for their support. Without them, I would not have had the strength that it took to complete this training. Second, I am grateful for the Porter group members who have shared in many lively discussions about research, life, and music preferences. The support and laughter they provided created a great environment where we are colleagues and friends. Last but not least, I would like to thank my advisor, Dr. Marc Porter, for leading a laboratory that values creativity and considers the impact that research has on the betterment of lives. Above all, I am so thankful to have participated in research that I believe in and imagine will help move society forward to a brighter future. CHAPTER 1 INTRODUCTION 1.1 Overview Point-of-need (PON) detection strategies are critical in a variety of technical areas including disease diagnosis, pharmaceutical quality control, food safety, and bioterrorism agent screening.1-3 One of the most common architectures for such a test, the sandwichstyle surface-based immunoassay, relies on the selective capture of target antigens that are indicative of a disease or bioterrorism agent by immobilized antibodies, with subsequent labeling by a tagged antibody. Today's tests, however, often require specialized testing facilities, costly instrumentation, and long workup times, which limits their utility as PON tests.4 Thus, the development of a sensitive, low-cost, and rapid detection method that is easy to use in a variety of settings (e.g., field-deployable) continues to be a major research thrust in the biomedical and bioterrorism arenas.5-6 This dissertation focuses on the development of an immunoassay for botulinum neurotoxin (BoNT) coupled with in-depth studies on protein immobilization and synthesis of magnetic assay labels that can be applied to a variety of immunoassay platforms. In general, this dissertation will discuss the development of sandwich-style surfacebased immunoassays and methods to improve performance. Chapter 2 describes the 2 development of an immunoassay for BoNT, a potential bioterrorism agent, using surfaceenhanced Raman scattering (SERS) as a readout for antigen concentration. Chapter 3 discusses the use of N-hydroxysuccinimide (NHS) esters in immobilizing proteins such as antibodies at a surface for use as a capture substrate in immunoassay platforms. Chapter 4 elaborates on Chapter 3 with studies investigating methods to improve the efficiency of NHS ester-based protein coupling by varying methylene chain length, solution composition, and monolayer structure. Chapter 5 introduces the synthesis of highly magnetic microcapsules (MMCs) for use as assay labels in a giant magnetoresistance (GMR)-based platform. Altogether, this dissertation aims to contribute to the development of surface-based immunoassays using SERS- and GMR-based readouts with reactivity studies of NHS esters to determine optimum procedures for robust protein immobilization. Ultimately, this work is directed toward the development of a sensitive, low-cost, fielddeployable platform for the rapid detection of diseases and bioterrorism agents. 1.2 Botulinum Neurotoxins (BoNTs) Botulinum neurotoxins (BoNTs), produced by Clostridium botulinum, are some of the most lethal toxins known to humankind, and upon infection can result in the potentially fatal disease, botulism.7-8 The median lethal dose (LD50), or the dose that causes death in 50% of the test population, is 1.3 to 2.1 ng/kg intravenously, 1 μg/kg orally, or 10 to 13 ng/kg by inhalation.9 Even at doses below the LD50, the side effects of botulism, primarily local to systemic paralysis, can take months of palliative care.10-11 Due to the pathogenicity of BoNTs, terrorists have attempted to weaponize BoNTs, with a failed attempt at aerosol dispersion in Japan in the 1930s and, more recently, a threat from Iraq between 1985 and 3 1998.12 Due to the extreme lethality of BoNTs, and their history of use in bioterrorism attacks, the Center for Disease Control and Prevention (CDC) has designated BoNTs a category A pathogen and serious bioterrorism threat.9-10,13 BoNTs consist of a ~100 kDa heavy chain that targets axon terminals and a ~50 kDa light chain which contains a zinc metalloprotease that blocks neurotransmitter release.14-15 A representative structure is shown in Figure 1.1.16-18 Initially, the two chains exist as a single polypeptide chain (Figure 1.1A) which is activated by cleavage into two separate chains.14 The receptor binding domains of the heavy chain (red region in Figure 1.1B) are responsible for binding to the presynaptic terminus, allowing endocytosis, while the translocation domain of the heavy chain (green region in Figure 1.1B) penetrates the endosome and forms a pore.18 This pore then allows the light chain (blue region in Figure 1.1B) to pass across the cell membrane and into the cytosol, where it can block neurotransmitter release and cause paralysis. As shown, the heavy and light chains are connected by a single disulfide bond that, when broken, produces the two distinct chains. Because of the decrease in toxicity upon separation of the heavy and light chains, these chain fragments are often used in the development of PON tests.19-20 Current diagnosis for botulism is based on clinical suspicion or the mouse bioassay, which is the gold standard for molecular diagnosis of BoNTs.21 The mouse bioassay involves monitoring the symptoms and viability of a mouse after injection of the suspect sample.21-22 Thus, the mouse bioassay is a qualitative test, with a limit of detection of ~20 pg/mL, that does not differentiate between different toxins (i.e., yield a quantitative assessment of toxin levels or differentiate between toxins that produce similar symptoms).22 In addition, the mouse bioassay is resource intensive due to the necessary 4 Figure 1.1. Schematic of BoNT structure before and after cleavage. Before cleavage (A) shows a single polypeptide chain and after cleavage (B) shows the catalytic domain of the light chain (blue) with zinc (Zn2+) metalloprotease, translocation domain of the heavy chain (green), and binding domain of the heavy chain (red). 5 labor involved in the use, care, and monitoring of live animals; it is also time intensive as the test requires up to 7 days. The time required and costs associated with the mouse bioassay limits its use as a screening tool. To effect rapid response to biodefense events, much research has been aimed at developing a PON test capable of rapid and sensitive detection of BoNTs at low cost.6,23 1.3 Point-of-Need (PON) Test Considerations There are several important considerations involved in test development (i.e., analytical sensitivity and specificity) and specifically for PON tests (i.e., speed, cost, and ease of use).24-25 It is worthwhile to note that the analytical definitions for sensitivity and specificity differ from the clinical definitions, with the former being used during early development phases of an immunoassay before incorporation of patient samples. For our purposes, references to sensitivity or specificity will refer only to analytical sensitivity or analytical specificity, respectively. During the development of an immunoassay, the analytical sensitivity and specificity are used as metrics to estimate the analytical measurement capabilities of the platform. The analytical sensitivity of an assay is related to the limit of detection (LoD) or the antigen concentration that can be reliably differentiated from the blank (i.e., signal at zero concentration).24 The term analytical sensitivity may also be used to designate the ability to differentiate changes in antigen concentration.24 In this case, an increase in analytical sensitivity would infer an increase in the difference between signals from two different antigen concentrations. On the other hand, analytical specificity refers to the ability of the assay to detect a specific antigen in the presence of other proteins that may or 6 may not be structurally similar. The analytical specificity of an immunoassay is important to achieve accurate results (i.e., reliable detection of the target antigen). In the case of detection of bioterrorism agents, the speed, cost, and ease of use are additional metrics in the evaluation of an immunoassay as a PON test.25 First, the speed is especially important in biosafety screening or patient diagnosis during a bioterrorism attack. In these time-sensitive situations, the test must be rapid enough to enable a quick response to bioterrorism threats and effectively limit damage. Second, the cost of an immunoassay dictates its widespread use. The practicality of moving an immunoassay beyond the research laboratory is highly contingent on maintaining low costs for infrastructure, instrumentation, sample transportation, and the test itself. This is especially important with the detection of BoNTs which require on-site detection, often in facilities that have limited laboratory resources. Lastly, the ease of use of an immunoassay correlates with the skills required of the user and the accuracy of the test. This can either entail less training for technical personnel or, in the case of a very simple test, minimal training with dispensation directly to end users. In addition, it is anticipated that the likelihood of operator error is reduced with a simpler test. All of these considerations are also important in medical diagnostic tests where treatment hinges upon rapid detection and widespread use, especially in resource-limited areas, is dependent on the cost.26 In all, the sensitivity and specificity constraints, time requirements, cost, and ease of use must be considered to direct the development of a PON test. 7 1.4 Molecular Diagnostics and Solid-Phase Immunoassays (SPIs) Molecular diagnostic tests, which use molecular biology to analyze the proteome, genome, or microbiome of an individual, have long been at the forefront of personalized medicine.27-30 Traditionally, molecular diagnostic tests analyze markers indicative of a disease (i.e., biomarkers) for patient diagnosis; these tests can serve to provide a unique baseline for biomarker concentration for a particular individual (i.e., personalized medicine). However, these assays can also serve to manage bioterrorist threats through rapid security screening for potential bioterrorism agents. The history of molecular diagnostic tests began in the 1950s when Yalow and Berson reported the detection of insulin by a tracer antibody through competition between free insulin from the patient sample and a known quantity of radiolabeled insulin.31 This report outlined the use of immune complexes (i.e., immunoassay) and the competition between a free and labeled antigen (i.e., competitive-style immunoassay) for biomarker detection. The competitive-style immunoassay described above was translated to a solidphase immunoassay (SPI) by Catt and Tregear in 1967, who simplified the assay format by combining the binding of antigens and separation of bound antigens into a single step.32 The SPI format has been adapted over the years to include a variety of readout mechanisms including fluorescence, electrochemistry, chemiluminescence, and surface plasmon resonance.33-35 With the addition of alternate readout mechanisms came the development of a sandwich-style immunoassay which tagged the captured antigen with a labeled tracer antibody.36 A schematic diagram for a sandwich-style immunoassay is shown in Figure 1.2A. As shown, there are two main components: (1) a capture surface which consists of primary capture antibodies on a solid surface that may or may not use an additional linking 8 Figure 1.2. Schematic of sandwich style immunoassay and exemplar dose response plot. Sandwich-style immunoassay (A) shows capture surface consisting of a solid substrate, linking monolayer, and primary capture antibody that specifically binds to a target antigen (Ag) that is then labeled with a secondary tracer antibody. Dose response plot (B) is constructed from known standards with Ag concentration as a function of readout from label. 9 chemistry for tethering of the primary antibodies, and (2) a secondary tracer antibody that is attached to a label to allow for quantitative readout. The capture surface serves to specifically bind a target antigen that, upon being labeled, can be quantified by a comparison to a dose response plot constructed using standards, as shown in Figure 1.2B. With the advances in biomaterials and sensing techniques, research in biosensor development has continued to advance rapidly.27,29-30 Still, the most notable and widespread SPI is the enzyme-linked immunosorbent assay (ELISA), in which the biomarker is tagged by an enzyme that produces a colorimetric product.37-38 In ELISA, the signal from an antigen binding event is amplified from the production of a large number of chromophores from a single enzyme which enhances the ability to quantitate low antigen concentrations. Though hugely popular since its inception in 1971, there are several documented limitations to ELISA, including environmental stability, cost, and multiplexation challenges that plague the commercialization of a multiplexed test.39-43 Thus, our laboratory has focused on alternative detection platforms using giant magnetoresistance (GMR)44-46 and surface-enhanced Raman scattering (SERS) 47-57 to overcome some of the traditional limitations to ELISA. 1.5 Surface-Enhanced Raman Scattering (SERS) Immunoassay SERS is based on the phenomenon known as Raman scattering which occurs when an incident photon interacts with molecular vibrations, causing excitation to a virtual state.58-59 This phenomenon is represented in Figure 1.3A. Though a majority of the scattered light has a conserved frequency (i.e., elastically scattered light known as Rayleigh scattering), some of the light is inelastically scattered with shifts in frequency due 10 Figure 1.3. Schematic of Raman scattering and surface-enhanced Raman scattering (SERS) on a nanoparticle (NP). Schematic of Raman scattering (A) shows the interaction of incident light (incoming photons) and molecular vibrations between atoms to produce anti-Stokes Raman, Rayleigh, and Stokes Raman scattering. Diagram of SERS (B) shows the localized surface plasmon resonance of metal NPs, the basis for a SERS-based immunoassay. When irradiated by an electric field, the metal NP exhibits a collective oscillation of the electron cloud that enhances the electromagnetic field (E-field) around the NP providing an enhancement in Raman signal.60 11 to a gain (i.e., anti-Stokes Raman scattering) or loss (i.e., Stokes Raman scattering) of energy by the scattered photon.58-59 Raman instrumentation is typically designed to filter the Rayleigh and anti-Stokes Raman scattering to give a Stokes Raman signal intensity in counts per unit time (counts per second - cps) as a function of Raman shift (i.e., frequency in wavenumbers relative to that of the excitation source). The Raman signal intensity, however, is intrinsically weak due to the low number of incident photons that undergo Raman scattering (i.e., 1 in 106).61 In the 1970s, the discovery of SERS by Van Duyne et al. resulted in a renewal in the use of Raman spectroscopy due to increased sensitivity.62-64 SERS is a specific phenomenon in which a localized surface plasmon is generated which can enhance the electromagnetic field of nearby molecules.60,62-65 The SERS enhancement has been determined to be a combination of chemical and electromagnetic factors with practical enhancements up to 1014.66-67 The primary electromagnetic factor arises from the production of a localized surface plasmon resonance (LSPR)68-73 while the minor chemical factor in SERS enhancement is due to the charge-transfer between the adsorbed Ramanactive species and the underlying nanostructured metal.74-76 The LSPR occurs when incident photons induce an oscillation in the electron cloud of a metal nanoparticle (NPs) that enhances the electromagnetic field, as shown in Figure 1.3B. The SERS enhancement is dependent on the incident wavelength and the NP size,77 along with the composition of the underlying substrate and NP,78-79 distance between the NP and the underlying substrate,80 angle of incidence of the incident light,81 and dielectric properties of the medium.82 The SERS phenomenon can be translated to an immunoassay detection strategy by 12 the addition of a Raman-active NP as a label.54,83 The schematic for this assay is shown in Figure 1.4. The extrinsic Raman label (ERL) consists of a 60 nm gold nanoparticle (AuNP) functionalized with a Raman reporter molecule (RRM) and secondary tracer antibody. When the capture antibody layer captures the target antigen, the bound antigen is then recognized by the antibodies on the ERL, with the amount of antigen indirectly quantified by the strength of the SERS signal from the RRM. Our laboratory has previously published on the use of SERS-based immunoassays in the detection of biomarkers such as prostatespecific antigen,50 feline calicivirus,51 porcine parvovirus,84 Mycobacterium avium subsp. paratuberculosis,47,57 vitamin D3,56 and cancer markers MUC4, MMP7, and CA19-9.49,55 The use of SERS as a readout mechanism for immunoassays has several benefits over ELISA.54,85-87 In particular, SERS as an immunoassay readout method has shown promise with regards to increased sensitivity, multiplexation capability, ability to be fielddeployable, and ability to archive samples as described below. First, the sensitivity of a sandwich-style immunoassay is directly related to the ability to detect small quantities of bound antigen. With SERS, the potential for detection of a single antigen binding event has been demonstrated by measurements of single molecules88-90 and individual ERLs.90 Second, the relatively narrow band widths of Raman in comparison to absorption allows for increased differentiation (i.e., due to limited band overlap) between different RRMs. This attribute becomes especially important when considering multiplexation capabilities, which in a SERS-based immunoassay can be achieved by a mixture of ERLs, each functionalized with a different RRM and tracer antibody.48 Further, the multiplexed SERSbased immunoassay can be analyzed by a single excitation source, which allows for enhanced portability (i.e., field-deployability) by simplifying Raman instrumentation.91-92 13 Figure 1.4. Schematic of steps in SERS-based immunoassay. Steps include (A) the preparation of extrinsic Raman labels (ERLs) by functionalization of a gold nanoparticle (AuNP) with a Raman reporter molecule (RRM), dithiobis (succinimidyl nitrobenzoate) (DSNB), and tracer antibody (Ab); (B) the preparation of the capture substrate by formation of a linking monolayer, via the linker molecule dithiobis (succinimidyl propionate) (DSP), and capture antibody (Ab); and (C) the capture of a target antigen (Ag) by the capture substrate with subsequent labeling by ERLs. 14 Finally, the completed assay (i.e., ERLs dried on a gold substrate) is expected to be environmentally stable which would allow for the archiving of SERS-based immunoassays for analysis at a later time. 1.6 Giant Magnetoresistance (GMR) Immunoassay GMR is a phenomenon observed in thin-film structures of alternating magnetic and conductive nonmagnetic layers whose thickness is comparable to the mean free path of an electron. These devices demonstrate a notable resistance change as a function of an applied external magnetic field (Happ).93-97 A schematic for this process is shown in Figure 1.5A. In the absence of a magnetic field (i.e., Happ = 0), the magnetic moments in the layers in the GMR sensor are aligned in an antiparallel manner due to antiferromagnetic exchange coupling.98 In this case, the probability of electron scattering at the layer interfaces is high (i.e., with a decreased electron mean free path, the sensor resistance increases). With the application of a magnetic field (i.e., Happ ≠ 0), the magnetic moments of the layers become aligned, the electron mean free path increases (i.e., electron scattering is less probable), and the device's resistance decreases. When Happ is large enough, the device becomes saturated and no further change in resistance is observed with increasing Happ; this is known as the saturation region. The resistance of a multilayer GMR sensor as a function of Happ - the transfer curve - is shown in Figure 1.5B. As described above, the scattering in a multilayer GMR sensor changes with the magnetic alignment of the layers and the lowest resistance is observed when the magnetic moments of the magnetic layers are aligned in a parallel fashion - the saturation region. The hysteretic and linear regions of the transfer curve occur 15 Figure 1.5. Schematic of GMR phenomenon exhibited by a thin-film structure. Schematic of thin-film structure (A) shows alternating magnetic (grey) and nonmagnetic (orange) metallic layers demonstrating GMR where the resistance decreases with parallel alignment of the magnetic layers and increases with antiparallel alignment of the magnetic layers. The black arrows indicate movement of the electrons (black circles). GMR plot (B) shows resistance of the thin-film structure as a function of applied magnetic field (Happ) with the saturation region showing little or no change in resistance and the linear and hysteretic region showing the highest resistance, due to antiparallel alignment of the magnetic layers in the absence of a magnetic field. 16 at low applied magnetic field (i.e., Happ ~ 0), and demonstrate the highest resistance and greatest sensitivity of the device. Since 1998,99 use of a GMR sensor in an immunoassay format has been a major thrust in biosensor research.100-103 In an immunoassay with GMR-based readout, the captured antigen is labeled with magnetic nanoparticles (MNP), rather than a gold nanoparticle as with SERS-based readout. The device then operates as an MNP-label detector in a static applied magnetic field. The signal of the sensor is perturbed when it comes in close proximity to an MNP label; it is this slight perturbation that is used to register the presence and quantity of MNP labels as a change in measured resistance. The ubiquity of the GMR sensors is due, in part, to their use in the read heads of hard disk drives.98,104 Owing to the ability of GMR sensors to detect the magnetization of nanometric magnetic data bits, GMR-based readout of immunomagnetic assays is extremely sensitive.44-46,105 In addition, GMR-based read rates are extremely fast, a result leveraged from the work performed by the hard disk drive industry, which makes them an excellent detector choice for high-throughput biomarker detection. In this case, however, multiplexation would be achieved by the fast read rates of multiple addresses rather than a single read of one address, as with SERS-based readout. Additionally, the small size of GMR sensors make them ideal for use in the field where portability is key. To this end, our laboratory has demonstrated the potential of GMR-based readout for low-level biomarker detection.44-46 We continue to focus on immunoassay readout by both SERS and GMR systems in our laboratory. This dissertation includes work on magnetic microcapsule (MMC) immunoassay labels to potentially improve upon the magnetic labels we currently use in 17 our GMR assays through enhanced magnetization (i.e., high magnetic signal per binding event) and reduced sedimentation (i.e., decrease in sedimentation-based nonspecific adsorption). 1.7 Protein Immobilization in Immunoassays To achieve low limits of detection, all sandwich-based immunoassays, including those described above using SERS-based and GMR-based readouts, rely on the effective immobilization of a protein to a surface. It is this immobilization that allows for the specific capture of a target antigen from solution. As such, the reproducibility, density, and robustness of this capture layer is critical in the development of more sensitive immunoassays. Thus, this dissertation includes a detailed investigation of the effectiveness of NHS-linking chemistry, which is common practice in our and other laboratories, with methods to improve the efficiency of protein immobilization. We anticipate that by improving the effectiveness of protein immobilization, we can achieve a more robust, reproducible, and sensitive immunoassay platform. Protein immobilization strategies have been developed with aims of achieving high density while maintaining protein activity and stability on the surface.106-108 The immobilization of proteins, however, often affects the activity of the immobilized protein due to steric occlusion and denaturing.109 This reduction in accessibility can reduce the amount of target captured, and thus affect the sensitivity of an assay. Weak attachment of a protein can also be problematic due to the increased possibility for desorption upon the application of sample solutions and during rinsing procedures. Thus, special care must be taken to ensure binding activity and permanent attachment of the immobilized protein. 18 Many methods have been developed to immobilize proteins at a surface by physical adsorption, bioaffinity proteins, or covalent linkages.106-107 Though many of these methods are effective, covalent coupling of a protein is often preferred due to its robustness.108,110114 This method utilizes a spontaneously adsorbed monolayer (SAM) with functional end groups that can react with moieties of the protein to produce stable linkages.115-118 1.8 Protein Immobilization via NHS Chemistry One of the most common covalent linking chemistries involves Nhydroxysuccinimide (NHS) esters.115,119 This approach is shown in Figure 1.6A. Here, an NHS ester-terminated adlayer on gold is accomplished by either direct derivatization with an NHS ester containing thiol or disulfide or generated in situ by activation of a carboxylic acid-terminated adlayer with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDAC). Upon exposure to a capture antibody solution, the NHS ester end group reacts with the primary amines of biomolecules to form an amide linkage. In addition to the immobilization by NHS-based chemistries being widely applied to a variety of platforms, the amide linkages formed are stable to desorption and rely on the innate amine groups of proteins, eliminating the need for further functionalization of the protein.114 Though NHSbased chemistry simplifies the tethering of proteins to surfaces, the hydrolysis of the terminal succinimidyl group competes with amidization.114 Upon hydrolysis, the linking adlayer is deactivated which directly affects the extent of covalently immobilized protein density. Thus, we expect that in the case where proteins are not covalently linked, desorption of the immobilized protein may cause reduced capture of the antigen. Mechanistically, the hydrolysis and aminolysis of NHS-esters has been shown to Figure 1.6. Schematic of antibody (Ab) immobilization using NHS ester-terminated monolayers. Immobilization of Ab (A) is achieved by reaction with an NHS ester-terminated monolayer (left) to form an amide linkage (right). Schematic of simplified BAC2 mechanism (B) for the hydrolysis or aminolysis of NHS esters shows attack of the hydroxide ion or deprotonated amine at the acyl carbon (left) which produces a tetrahedral intermediate (middle) and then either a carboxylic acid or amide linkage (right). 19 20 be base-catalyzed, which means the reaction rate is directly dependent on the solution pH (i.e., the hydroxide ion concentration for hydrolysis and the extent of deprotonation of the primary amine for aminolysis).119 In solution, the pathways for the base hydrolysis and aminolysis of esters follow a similar BAC2 mechanism, shown in Figure 1.6B, where ‘B' represents base-catalysis, the subscript ‘AC' indicates that the bond between the acyl carbon and oxygen of the tetrahedral intermediate breaks as the reaction proceeds, and ‘2' refers to the bimolecular nature of the reaction.120 Though NHS esters are susceptible to hydrolysis, there are several conditions that can increase the relative rate of aminolysis by either promoting aminolysis or hindering hydrolysis. First, increasing the concentration of the protein, thus increasing the primary amine concentration, can diminish the effect of hydrolysis by increasing the rate of aminolysis. Second, regeneration of the NHS ester surface through continued NHS esterification over the course of the reaction can allow for the previously hydrolyzed monolayer to undergo aminolysis. Finally, as previously stated, the pH can affect reaction rates through the deprotonation of the amine reactant in aminolysis or the hydroxide ion concentration in hydrolysis. However, since the rates of aminolysis and hydrolysis are both expected to increase with increasing pH, the ideal solution conditions for protein immobilization are difficult to anticipate and, in addition, must maintain the protein structure and stability.121 Thus far, we have used mostly unrelated experiments (i.e., conditions that do not mimic those used in immunoassays) to extrapolate the effectiveness of NHS chemistry in protein immobilization. Herein, we aim to investigate the reactivity of NHS ester-terminated monolayers on gold under relevant immunoassay conditions. 21 1.9 Interfacial Reactions of Spontaneously Adsorbed Monolayers (SAMs) In the investigation of the reactivity of NHS ester-terminated monolayers on gold, we must consider the structure of a SAM and how that structure may affect the interfacial reactivity. As shown in Figure 1.7, the exposed end group of a SAM allows for reaction or interaction with molecules and ions from the bulk solution.122-123 These interactions between the bulk solution and the end group determine the nature of the interface which can differ in comparison to the bulk solution (i.e., pH, ion concentration, and reactant concentration).124-126 In addition, the density and structure (i.e., inter-chain interactions and defects) of the monolayer can affect the molecular environment, and thus reactivity of the monolayer. In fact, there have been numerous reports detailing the effects that these factors (i.e., end group identity, inter-chain interactions, and defects) have on the reactivity of SAMs.127-128 First, the ease of ionization of a monolayer often changes from that in solution with acids and bases becoming less acidic and less basic, respectively, by approximately 2 to 5 pH units.123,129-130 Thus, the reactivity of immobilized reactants likely varies from those in solution in part due to the change in local concentration in the double layer. Second, the packing density of the adlayer can be affected by a multitude of parameters including methylene chain length and end group identity.131-137 For example, longer methylene chain alkanethiolates (n > 10) have been shown to exhibit a densely packed (i.e., crystalline-like) adlayer, while shorter methylene chains (n < 10) produce a loosely packed (i.e., liquid-like) adlayer, Figure 1.8A.131-132 The packing density can also be affected by the end group size.134-135,138-139 The packing density of a monolayer is directly related to the defect density or type 22 Figure 1.7. Schematic of monolayer on gold formed by the head group (thiolate) interaction with solid gold substrate. The end group interacts with the bulk solution to produce a double layer at the interface (a subset of the bulk solution that often has a different concentration of reactants and ions). 23 Figure 1.8. Schematic of monolayers on gold showing various packing densities and defects based on (A) adsorbate orientation, (B) domain boundaries, and (C) gold terraces or step edges. 24 of defects (i.e., local distortion in monolayer packing). These defects, which can be due to domain boundaries formed by misaligned head/end groups or underlying substrate roughness by terrace size/step edges, are cartooned in Figure 1.8B and Figure 1.8C, respectively.136-137 The impact of defect density of a monolayer is significant with respect to interfacial reactions, with several reports confirming that interfacial reactions originate at defect sites and grow outward.140-144 These reports posit that the type and extent of defects in a monolayer can affect the reaction process and/or reaction rate.145-146 In the case of protein immobilization, the factors mentioned above must all be considered in order to determine the optimal conditions to achieve covalently immobilized proteins. To date, the relationship between reactivity studies of NHS esters and protein immobilization has been limited by conditions that do not mimic those used in immunoassays.128,140-141,147-151 Thus, with the considerations for monolayer properties that can affect interfacial reactivity outlined above, this dissertation includes fundamental studies of hydrolysis and aminolysis kinetics of NHS-terminated monolayers for extension to protein immobilization. 1.10 Dissertation Summary This dissertation will expand on the discussion presented above through the development of a SERS-based immunoassay for BoNT, several fundamental studies on protein immobilization using NHS chemistry, and the synthesis of magnetic microcapsules (MMCs) as a potential label in a GMR-based immunoassay. Chapter 2 presents the development of a SERS-based immunoassay for BoNT serotype A (BoNT-A) and BoNT serotype B (BoNT-B) using a SERS-based readout. These immunoassays give similar 25 detection limits to the currently available methods (i.e., mouse bioassay) in buffer and human serum with the potential for extension into a rapid, field-deployable platform. Chapter 3 discusses protein immobilization using NHS ester chemistry through a fundamental investigation of the competing hydrolysis and aminolysis surface reactions. Specifically, the hydrolysis and aminolysis reaction rate constants of immobilized NHS esters are determined through external reflection infrared spectroscopy (IR-ERS) and compared to ascertain the extent of immobilized protein under typical immunoassay conditions. These results importantly point to the high likelihood of physical adsorption of proteins rather than covalent linking, using NHS chemistry under typical immunoassay conditions. Chapter 4 extends the discussion of protein immobilization using NHS ester chemistry through changes to the monolayer structure and reaction conditions. These extended studies show some improvements in the extent of covalent protein immobilization using a molecule with a longer methylene chain and change in solution conditions. These findings, however, point to the critical need for a solution that eliminates hydrolysis altogether. Chapter 5 presents the synthesis and characterization of MMCs for labels in a GMR-based immunoassay. These MMCs show promise as an improved label compared to those commercially available when used in a GMR-based immunoassay due to reduced sedimentation properties (i.e., decrease in density due to the hollow nature of the microcapsule) and enhanced magnetization (i.e., inclusion of multiple smaller magnetic nanoparticles in a single MMC). Ultimately, this dissertation contributes to the development of sensitive PON tests for bioterrorism and human health with detailed investigations of the hydrolysis and aminolysis of NHS esters for protein immobilization. 26 1.11 References (1) Siegel, R. L.; Miller, K. D.; Jemal, A. CA-Cancer J. Clin. 2015, 65, 5-29. (2) Wardle, J.; Robb, K.; Vernon, S.; Waller, J. Am. Psychol. 2015, 70, 119-133. (3) Nikoleli, G.-P.; Karapetis, S.; Bratakou, S.; Nicolelis, D. P.; Tzamtzis, N.; Psychoyios, V. N.; Psaroudakis, N. Biosensors for Security and Bioterrorism: Definitions, History, Types of Agents, New Trends and Applications. In Advanced Sciences and Technologies for Security Applications, Nikolelis, D. P.; Nikoleli, D.P., Eds.; Springer International Publishing: West Sussex, England, 2016; pp 1-13. (4) Sharma, S.; Zapatero-Rodriguez, J.; Estrela, P.; O'Kennedy, R. Biosensors 2015, 5, 577-601. (5) Chan, C. P. Y.; Mak, W. C.; Cheung, K. Y.; Sin, K. K.; Yu, C. M.; Rainer, T. H.; Renneberg, R. Annu. Rev. Anal. Chem. 2013, 5, 191-211. (6) Saravanan, P.; Rajaseger, G.; Eric, Y. P.-H.; Moochhala, S. Botulinum Toxin: Present Knowledge and Threats. In Biological Toxins and Bioterrorism, Gopalakrishnakone, P.; Balali-Mood, M.; Llewellyn, L.; Singh, R. B., Eds. Springer Netherlands: Dordrecht, 2015; pp 29-42. (7) Gill, D. M. Microbiol. Rev. 1982, 46, 86-94. (8) Lamanna, C. Science 1959, 130, 763-772. (9) Arnon, S. S.; Schechter, R.; Inglesby, T. V.; Henderson, D. A.; Bartlett, J. G.; Ascher, M. S.; Eitzen, E.; Fine, A. D.; Hauer, J.; Layton, M.; Lillibridge, S.; Osterholm, M. T.; O'Toole, T.; Parker, G.; Perl, T. M.; Russell, P. K.; Swerdlow, D. L.; Tonat, K. JAMA 2001, 285, 1059-1070. (10) Khan, A. S.; Levitt, A. M.; Sage, M. J. MMWR 2000, 49, 1-14. (11) Franz, D. R.; Jahrling, P. B.; Friedlander, A. M.; McClain, D. J.; Hoover, D. L.; Bryne, W. R.; Pavlin, J. A.; Christopher, G. W.; Eitzen, E. M. JAMA 1997, 278, 399-411. (12) Zilinskas, R. A. JAMA 1997, 278, 418-424. (13) Dhaked, R. K.; Singh, M. K.; Singh, P.; Gupta, P. Indian J. Med. Res. 2010, 132, 489-503. (14) Huang, W.; Foster, J. A.; Rogachefsky, A. S. J. Am. Acad. Dermatol. 2000, 43, 249-259. 27 (15) Schiavo, G.; Matteoli, M.; Montecucco, C. Physiol. Rev. 2000, 80, 717-766. (16) Segelke, B.; Knapp, M.; Kadkhodayan, S.; Balhorn, R.; Rupp, B. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 6888-6893. (17) Turton, K.; Chaddock, J. A.; Acharya, K. R. Trends Biochem. Sci. 2002, 27, 552558. (18) Lacy, D. B.; Tepp, W.; Cohen, A. C.; DasGupta, B. R.; Stevens, R. C. Nat. Struct. Mol. Biol. 1998, 5, 898-902. (19) Bittner, M. A.; DasGupta, B. R.; Holz, R. W. J. Biol. Chem. 1989, 264, 1035410360. (20) Poulain, B.; Tauc, L.; Maisey, E. A.; Wadsworth, J. D.; Mohan, P. M.; Dolly, J. O. Proc. Natl. Acad. Sci. 1988, 85, 4090-4094. (21) Cheng, L. W.; Land, K. M.; Stanker, L. H. Current Methods for Detecting the Presence of Botulinum Neurotoxins in Food and Other Biological Samples. In Bioterrorism, Morse, S., Ed. Intech: 2012; pp 1-17. (22) Lindström, M.; Korkeala, H. Clin. Microbiol. Rev. 2006, 19, 298-314. (23) Dorner, B. G.; Zeleny, R.; Harju, K.; Hennekinne, J.-A.; Vanninen, P.; Schimmel, H.; Rummel, A. TrAC, Trends Anal. Chem. 2016, Advance online publication. doi: 10.1016/j.trac.2016.05.024. (24) Saah, A. J.; Hoover, D. R. Ann. Intern. Med. 1997, 126, 91-94. (25) Fortina, P.; Surrey, S.; Kricka, L. J. Trends Mol. Med. 2002, 8, 264-266. (26) Granger, J. H.; Schlotter, N. E.; Crawford, A. C.; Porter, M. D. Chem. Soc. Rev. 2016, 45, 3865-3882. (27) Song, Y.; Huang, Y.-Y.; Liu, X.; Zhang, X.; Ferrari, M.; Qin, L. Trends Biotechnol. 2014, 32, 132-139. (28) de Castro, D. G.; Clarke, P. A.; Al-Lazikani, B.; Workman, P. Clin. Pharmacol. Thera. 2012, 93, 252-259. (29) Sin, M. L.; Mach, K. E.; Wong, P. K.; Liao, J. C. Expert Rev. Molec. Diagn. 2014, 14, 225-244. (30) Turner, A. P. Chem. Soc. Rev. 2013, 42, 3184-3196. (31) Yalow, R.; Berson, S. Nature 1959, 184, 1648-1649. 28 (32) Catt, K. J.; Niall, H. D.; Tregear, G. W. Nature 1967, 213, 825-827. (33) Mullett, W. M.; Lai, E. P.; Yeung, J. M. Methods 2000, 22, 77-91. (34) Pei, X.; Zhang, B.; Tang, J.; Liu, B.; Lai, W.; Tang, D. Anal. Chim. Acta 2013, 758, 1-18. (35) Gribnau, T.; Leuvering, J.; Van Hell, H. J. Chromatogr. B: Biomed. Sci. Appl. 1986, 376, 175-189. (36) Wide, L. Solid phase antigen-antibody systems. In Radioimmunoassay methods, Kirkham, K. E., Hunter, W. M., Eds. Churchill Livingstone: London, 1971; Vol. 1, pp 405-412. (37) Engvall, E.; Perlmann, P. Immunochemistry 1971, 8, 871-874. (38) Van Weemen, B.; Schuurs, A. FEBS Lett. 1971, 15, 232-236. (39) Laurentius, L. B.; Owens, N. A.; Park, J.; Crawford, A. C.; Porter, M. D. Expert Rev. Mol. Diagn. 2016, 16, 883-895. (40) Yager, P.; Domingo, G. J.; Gerdes, J. Annu. Rev. Biomed. Eng. 2008, 10, 107-144. (41) Tighe, P. J.; Ryder, R. R.; Todd, I.; Fairclough, L. C. Proteomics:Clin. Appl. 2015, 9, 406-422. (42) Jain, K. K. Expert Rev. Mol. Diagn. 2003, 3, 153-161. (43) Giljohann, D. A.; Mirkin, C. A. Nature 2009, 462, 461-464. (44) Millen, R. L.; Kawaguchi, T.; Granger, M. C.; Porter, M. D.; Tondra, M. Anal. Chem. 2005, 77, 6581-6587. (45) Nordling, J.; Millen, R. L.; Bullen, H. A.; Porter, M. D.; Tondra, M.; Granger, M. C. Anal. Chem. 2008, 80, 7930-7939. (46) Millen, R. L.; Nordling, J.; Bullen, H. A.; Porter, M. D.; Tondra, M.; Granger, M. C. Anal. Chem. 2008, 80, 7940-7946. (47) Yakes, B. J.; Lipert, R. J.; Bannantine, J. P.; Porter, M. D. Clin. Vaccine Immunol. 2008, 15, 235-242. (48) Wang, G.; Park, H.-Y.; Lipert, R. J.; Porter, M. D. Anal. Chem. 2009, 81, 96439650. (49) Granger, J. H.; Granger, M. C.; Firpo, M. A.; Mulvihill, S. J.; Porter, M. D. Analyst 29 2013, 138, 410-416. (50) Grubisha, D. S.; Lipert, R. J.; Park, H. Y.; Driskell, J.; Porter, M. D. Anal. Chem. 2003, 74, 5936-5945. (51) Driskell, J. D.; Kwarta, K. M.; Lipert, R. J.; Porter, M. D.; Neill, J. D.; Ridpath, J. F. Anal. Chem. 2005, 77, 6147-6154. (52) Driskell, J. D.; Lipert, R. J.; Porter, M. D. J. Phys. Chem. B 2006, 110, 1744417451. (53) Driskell, J.; Uhlenkamp, J. M.; Lipert, R. J.; Porter, M. D. Anal. Chem. 2007, 79, 4141-4148. (54) Ni, J.; Lipert, R. J.; Dawson, B.; Porter, M. D. Anal. Chem. 1999, 71, 4903-4908. (55) Wang, G.; Lipert, R. J.; Jain, M.; Kaur, S.; Chakraboty, S.; Torres, M. P.; Batra, S. K.; Brand, R. E.; Porter, M. D. Anal. Chem. 2011, 83, 2554-2561. (56) Dufek, E. J.; Ehlert, B.; Granger, M. C.; Sandrock, T. M.; Legge, S. L.; Herrmann, M. G.; Mieikle, A. W.; Porter, M. D. Analyst 2010, 135, 2811-2817. (57) Yakes, B. J.; Lipert, R. J.; Bannantine, J. P.; Porter, M. D. Clin. Vaccine Immunol. 2008, 15, 227-234. (58) Skoog, D. A.; Holler, F. J.; Nieman, T. A., Principles of Instrumental Analysis. Saunders College Pub.: Philadelphia, PA, 1998. (59) Ingle Jr, J. D.; Crouch, S. R., Spectrochemical analysis. Prentice Hall: Upper Saddle River, 1988. (60) Stiles, P. L.; Dieringer, J. A.; Shah, N. C.; Van Duyne, R. P. Annu. Rev. Anal. Chem. 2008, 1, 601-626. (61) Smith, E.; Dent, G., Modern Raman Spectroscopy - A Practical Approach. John Wiley & Sons Ltd,: West Sussex, England, 2005. (62) Fleischmann, M.; Hendra, P. J.; McQuillan, A. Chem. Phys. Lett. 1974, 26, 163166. (63) Albrecht, M. G.; Creighton, J. A. J. Am. Chem. Soc. 1977, 99, 5215-5217. (64) Jeanmaire, D. L.; Van Duyne, R. P. J. Electroanal. Chem. Interfacial Electrochem. 1977, 84, 1-20. (65) Kneipp, K.; Moskovits, M.; Kneipp, H. Surface-enhanced Raman Scattering: 30 Physics and Applications. Springer Science & Business Media: Berlin, 2006. (66) Le Ru, E.; Blackie, E.; Meyer, M.; Etchegoin, P. G. J. Phys. Chem. C 2007, 111, 13794-13803. (67) Schatz, G. C.; Van Duyne, R. P. Surf. Sci. 1980, 101, 425-438. (68) Moskovits, M. J. Chem. Phys. 1978, 69, 4159-4161. (69) Otto, A.; Timper, J.; Billmann, J.; Kovacs, G.; Pockrand, I. Surf. Sci. Lett. 1980, 92, L55-L57. (70) Schatz, G. C.; Van Duyne, R. P. Electromagnetic Mechanism of Surface‐enhanced Spectroscopy. John Wiley & Sons: Chichester, 2002. (71) Kerker, M. Acc. Chem. Res. 1984, 17, 271-277. (72) Aravind, P. K.; Nitzan, A.; Metiu, H. Surf. Sci. 1981, 110, 189-204. (73) Creighton, J. A. Surf. Sci. 1983, 124, 209-219. (74) Persson, B. Chem. Phys. Lett. 1981, 82, 561-565. (75) Adrian, F. J. J. Chem. Phys. 1982, 77, 5302-5314. (76) Kambhampati, P.; Child, C.; Foster, M. C.; Campion, A. J. Chem. Phys. 1998, 108, 5013-5026. (77) Yoon, J. K.; Kim, K.; Shin, K. S. J. Phys. Chem. C 2009, 113, 1769-1774. (78) Kim, K.; Yoon, J. K. J. Phys. Chem. B 2005, 109, 20731-20736. (79) Kim, K.; Lee, H. S. J. Phys. Chem. B 2005, 109, 18929-18934. (80) Hill, R. T.; Mock, J. J.; Urzhumov, Y.; Sebba, D. S.; Oldenburg, S. J.; Chen, S.-Y.; Lazarides, A. A.; Chilkoti, A.; Smith, D. R. Nano Lett. 2010, 10, 4150-4154. (81) Mubeen, S.; Zhang, S.; Kim, N.; Lee, S.; Krämer, S.; Xu, H.; Moskovits, M. Nano Lett. 2012, 12, 2088-2094. (82) Knight, M. W.; Wu, Y.; Lassiter, J. B.; Nordlander, P.; Halas, N. J. Nano Lett. 2009, 9, 2188-2192. (83) Porter, M. D.; Lipert, R. J.; Siperko, L. M.; Wang, G.; Narayanan, R. Chem. Soc. Rev. 2008, 37, 1001-1011. 31 (84) Driskell, J. D.; Kwarta, K. M.; Lipert, R. J.; Vorwald, A.; Neill, J. D.; Ridpath, J. F.; Porter, M. D. J. Virol. Methods 2006, 138, 160-169. (85) Mulvaney, S. P.; Musick, M. D.; Keating, C. D.; Natan, M. J. Langmuir 2003, 19, 4784-4790. (86) Faulds, K.; Smith, W. E.; Graham, D. Anal. Chem. 2004, 76, 412-417. (87) Graham, D.; Mallinder, B. J.; Whitcombe, D.; Watson, N. D.; Smith, W. E. Anal. Chem. 2002, 74, 1069-1074. (88) Qian, X.-M.; Nie, S. Chem. Soc. Rev. 2008, 37, 912-920. (89) Nie, S.; Emory, S. R. Science 1997, 275, 1102-1106. (90) Park, H.-Y.; Lipert, R. J.; Porter, M. D. In Single-particle Raman Measurements of Gold Nanoparticles Used in Surface-enhanced Raman Scattering (SERS)-based Sandwich Immunoassays, Optics East, International Society for Optics and Photonics: 2004; pp 464-477. (91) Han, Z.; Liu, H.; Meng, J.; Yang, L.; Liu, J.; Liu, J. Anal. Chem. 2015, 87, 95009506. (92) Zheng, J.; Pang, S.; Labuza, T. P.; He, L. Talanta 2014, 129, 79-85. (93) Prinz, G. A. Science 1998, 282, 1660-1663. (94) White, R. L. IEEE Trans. Magn. 1992, 28, 2482-2487. (95) Mapps, D. J. Sens. Actuators, A 1997, 59, 9-19. (96) Johnson, M. J. Phys. Chem. B 2005, 109, 14278-14291. (97) Daughton, J. M.; Bade, P. A.; Jenson, M. L.; Rahmati, M. M. M. IEEE Trans. Magn. 1992, 28, 2488-2493. (98) Fert, A. Rev. Mod. Phys. 2008, 80, 1517-1530. (99) Baselt, D. R.; Lee, G. U.; Natesan, M.; Metzger, S. W.; Sheehan, P. E.; Colton, R. J. Biosens. Bioelectron. 1998, 13, 731-739. (100) Tamanaha, C.; Mulvaney, S.; Rife, J.; Whitman, L. Biosens. Bioelectron. 2008, 24, 1-13. (101) Koh, C.-Y.; Schaff, U. Y.; Piccini, M. E.; Stanker, L. H.; Cheng, L. W.; Ravichandran, E.; Singh, B.-R.; Sommer, G. J.; Singh, A. K. Anal. Chem. 2015, 87, 32 922-928. (102) Zhang, Y.; Zhou, D. Expert Rev. Mol. Diagn. 2012, 12, 565-571. (103) Hall, D.; Gaster, R.; Lin, T.; Osterfeld, S.; Han, S.; Murmann, B.; Wang, S. Biosens. Bioelectron. 2010, 25, 2051-2057. (104) Chappert, C.; Fert, A.; Van Dau, F. N. Nat. Mater. 2007, 6, 813-823. (105) Grochowski, E. Datatech (September 1998), ICG Publishing 1998, 11-16. (106) Rusmini, F.; Zhong, Z.; Feijen, J. Biomacromolecules 2007, 8, 1775-1789. (107) Jonkheijm, P.; Weinrich, D.; Schroder, H.; Niemeyer, C. M.; Waldmann, H. Angew. Chem., Int. Ed. 2008, 47, 9618-9647. (108) Trilling, A. K.; Beekwilder, J.; Zuilhof, H. Analyst 2013, 138, 1619-1627. (109) Hlady, V.; Buijs, J. Curr. Opin. Biotechnol. 1996, 7, 72-77. (110) Buijs, J.; van den Berg, P. A. W.; Lichtenbelt, J. W. T.; Norde, W.; Lyklema, J. J. Colloid Interface Sci. 1996, 178, 594-605. (111) Malmsten, M. J. Colloid Interface Sci. 1998, 207, 186-199. (112) Bhakta, S. A.; Evans, E.; Benavidez, T. E.; Garcia, C. D. Anal. Chim. Acta 2015, 872, 7-25. (113) Makaraviciute, A.; Ramanaviciene, A. Biosens. Bioelectron. 2013, 50, 460-471. (114) Hermanson, G. T. Bioconjugate Techniques. Academic press: Boston, 2013. (115) Frey, B. L.; Corn, R. M. Anal. Chem. 1996, 68, 3187-3193. (116) Mateo, C.; Torres, R.; Fernandex-Lorente, G.; Ortiz, C.; Fuentes, M.; Hidalgo, A.; Lopez-Gallego, F.; Abian, O.; Palomo, J. M.; Betancor, L.; Pessela, B. C. C.; Guisan, J. M.; Fernandez-Lafuente, R. Biomacromolecules 2003, 4, 772-777. (117) Wong, L. S.; Khan, F.; Micklefield, J. Chem. Rev. 2009, 109, 4025-4053. (118) Camarero, J. A. Peptide Science 2007, 90, 450-458. (119) Mädler, S.; Bich, C.; Touboul, D.; Zenobi, R. J. Mass Spectrom. 2009, 44, 694706. (120) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry Part A: Structure and 33 Mechanisms. 4 ed.; Kluwer Academic/Plenum Publishers: New York, 2000. (121) Anjaneyulu, P. S. R.; Staros, J. V. Int. J. Pept. Protein Res. 1987, 30, 117-124. (122) Ulman, A. An Introduction to Ultrathin Organic Films: From Langmuir-Blodgett to Self-Assembly. Academic Press: New York, 1991. (123) Ulman, A. Chem. Rev. 1996, 96, 1533-1554. (124) Kane, V.; Mulvaney, P. Langmuir 1998, 14, 3303-3311. (125) Rowe, G. K.; Creager, S. E. J. Phys. Chem. 1994, 98, 5500-5507. (126) White, H. S.; Peterson, J. D.; Cui, Q.; Stevenson, K. J. J. Phys. Chem. B 1998, 102, 2930-2934. (127) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103-1170. (128) Chechik, V.; Crooks, R. M.; Stirling, C. J. M. Adv. Mater. 2000, 12, 1161-1171. (129) Bryant, M. A.; Crooks, R. M. Langmuir 1993, 9, 385-387. (130) Betts, J. J. P., B. A. Trans. Faraday Soc. 1956, 52, 1581-1589. (131) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321-335. (132) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559-3568. (133) Wagner, P.; Hegner, M.; Kernen, P.; Zaugg, F.; Semenza, G. Biophys. J. 1996, 70, 2052-2066. (134) Walczak, M. M.; Popenoe, D. D.; Deinhammer, R. S.; Lamp, B. D.; Chung, C.; Porter, M. D. Langmuir 1991, 7, 2687-2693. (135) Walczak, M. M.; Alves, C. A.; Lamp, B. D.; Porter, M. D. J. Electroanal. Chem. 1995, 396, 103-114. (136) Chidsey, C. E. D.; Loiacono, D. N. Langmuir 1990, 6, 682-691. (137) Creager, S. E.; Hockett, L. A.; Rowe, G. K. Langmuir 1992, 8, 854-861. (138) Wagner, P.; Hegner, M.; Kernen, P.; Zaugg, F.; Semenza, G. Biophys. J. 1996, 70, 2052-2066. 34 (139) Wink, T.; Van Zuilen, S.; Bult, A.; Van Bennekom, W. Analyst 1997, 122, 43R50R. (140) Schönherr, H.; Chechik, V.; Stirling, C. J.; Vancso, G. J. J. Am. Chem. Soc. 2000, 122, 3679-3687. (141) Schönherr, H.; Feng, C.; Shovsky, A. Langmuir 2003, 19, 10843-10851. (142) Dordi, B.; Pickering, J. P.; Schonherr, H.; Vancso, G. J. Surf. Sci. 2004, 570, 5766. (143) Dordi, B.; Pickering, J. P.; Schönherr, H.; Vancso, G. J. Eur. Polym. J. 2004, 40, 939-947. (144) Neogi, P.; Neogi, S.; Stirling, C. J. M. J. Chem. Soc., Chem. Commun. 1993, 11341136. (145) Leopold, M. C.; Black, J. A.; Bowden, E. F. Langmuir 2002, 18, 978-980. (146) Leopold, M. C.; Bowden, E. F. Langmuir 2002, 18, 2239-2245. (147) Dordi, B.; Schonherr, H.; Vancso, G. J. Langmuir 2003, 19, 5780-5786. (148) Xiao, S.-J.; Brunner, S.; Wieland, M. J. Phys. Chem. B 2004, 108, 16508-16517. (149) Cline, G. W.; Hanna, S. B. J. Am. Chem. Soc. 1987, 109, 3087-3091. (150) Cline, G. W.; Hanna, S. B. J. Org. Chem. 1988, 53, 3583-3586. (151) Chechik, V.; Stirling, C. J. M. Langmuir 1997, 13, 6354-6356. CHAPTER 2 SERS-BASED DETECTION OF CLOSTRIDIUM BOTULINUM NEUROTOXINS A AND B TOWARD THE DEVELOPMENT OF A BIODEFENSE POINT-OF-NEED PLATFORM 2.1 Introduction Due to high levels of lethality, botulinum neurotoxins (BoNTs) are classified as category A pathogens by the Center for Disease Control and Prevention (CDC).1 Produced by the bacterium Clostridium botulinum, BoNTs cause extreme and potentially fatal muscle paralysis.1 Even with immediate treatment, recovery requires weeks to months of palliative care.2 The LD50 for BoNTs, or the lethal dose that would result in the death of 50% of an exposed population, is 1.3 to 2.1 ng/kg intravenously, 1 μg/kg when ingested, or 10 to 13 ng/kg when inhaled.3 Assuming an average adult body weight of 70 kg, this means a single gram of BoNT, if effectively aerosolized and dispersed, could kill nearly a million people.2,4 Past use of BoNTs in biowarfare include a failed attempt in Japan in the 1930s and more recently, 1985-1998, the discovery of large quantities of BoNTs in Iraq for the possible use in biowarfare.2,5 These factors collectively place the development of methods for the rapid detection of BoNTs as a critical need for national security and public health.12,6-7 Today, the diagnosis of botulism is based largely on clinical suspicion based on the 36 patient's history and symptoms.1,8-9 However, there are a number of diseases that have similar symptoms, including Guillain-Barre syndrome, strokes, and myasthenia which complicates differential diagnosis.2 The mouse bioassay can be used to identify BoNTs by the inoculation of mice with the suspect sample, followed by close monitoring for symptoms. The mouse bioassay has a limit of detection (LoD) of ~20 pg/mL, but requires up to 7 days to complete.10 In addition, the mouse bioassay is only a qualitative "yes-no" test, and does not differentiate between toxins that produce similar symptoms (i.e., paralysis). Ultimately, the time required for the mouse bioassay and the use of live mice obviously prevents its use as a rapid screening test. The detection of BoNTs is further complicated by its existence as seven serotypes, designated A through G.3 All BoNTs are ~150 kDa zinc-dependent proteases that consist of a heavy and light chain linked together by a disulfide bridge.3 The ~100 kDa heavy chain is responsible for binding and translocating the toxin into the cell, allowing the enzymatic, ~50 kDa light chain to block the release of neurotransmitters like acetylcholine.11-13 However, the structure of BoNT serotypes can differ by as much as 64%, with additional subtypes differing by as much as 32%.4,14-17 It is known that BoNT serotypes A, B, E, and F cause botulism in humans and that the A (BoNT-A) and B (BoNT-B) serotypes account for the majority of cases.3 Research in BoNT diagnostic tests has focused heavily on overcoming the limitations of the mouse bioassay (i.e., time, cost, and ease of use).8,18-20 To date, several laboratories have reached comparable LoDs by using enzyme-linked immunosorbent assays (ELISAs),21-25 nucleic acid amplification tests (NAATs),26 and a few other approaches.18,27-29 Despite achieving a high analytical sensitivity, these tests fall short of 37 meeting the attributes of a point-of-need (PON) test (i.e., time, cost, ease of use, and field deployability), which are central to the on-site screening of suspected biowarfare samples and/or rapid diagnosis of patient populations during a bioterrorism attack.3 Past work in our laboratory focused on the development of highly sensitive immunoassays for disease markers (e.g., prostrate-specific antigen, Mycobacterium avium subsp. paratuberculosis, and cancer markers MUC4, MMP7, and CA19-9) using an extensible surface-enhanced Raman scattering (SERS) detection strategy.30-38 The workflow for this immunoassay is generalized in Figure 2.1A. The immunoassay uses extrinsic Raman labels (ERLs), which consist of gold nanoparticles that are coated with a Raman reporter molecule (RRM) and are subsequently modified with a layer of antibodies (Abs), and a capture substrate, which consists of a layer of antibodies deposited on a smooth gold substrate. The presence of captured antigen (Ag) is therefore recognized by the layer of Abs on the ERLs. Taking advantage of the SERS phenomenon, which enhances an electromagnetic field at the surface of nanostructured materials due to surface plasmon resonance, the amount of Ag can be indirectly quantified by the strength of the Raman signal from the RRM. In addition to the high sensitivity of SERS, the recent development of low-cost, portable Raman systems enables field deployability (i.e., PON test).39-40 This paper describes the development of SERS-based sandwich immunoassays for inactivated botulinum neurotoxins, BoNT-A and BoNT-B. First, we present the selection of antibody pairs (i.e., capture and tracer antibodies) by assessing the detection of BoNT antigens in a SERS-based immunoassay. This included testing polyclonal (pAb) and monoclonal (mAb) antibodies to assess the importance of single versus multiple epitope binding and for BoNT-A, antibodies raised against the heavy or light chain. The dose 38 Figure 2.1. Schematic of SERS-based sandwich immunoassay on 96-well Array-It© plate. (A) A gold address serves as the capture substrate after modification with dithiobis succinimidyl propionate (DSP) and BoNT-A or BoNT-B specific antibodies (Ab). After capture, the antigen (Ag) is labeled with extrinsic Raman labels (ERLs), which consist of AuNPs coated with a layer of a Raman reporter molecule, dithiobis succinimidyl nitrobenzoate (DSNB), and layer of BoNT-A or BoNT-B specific antibodies. (B) A 96-well Array-It© plate defines gold addresses (1 mm dia.) on a glass slide. 39 response plots for the optimized immunoassays demonstrate LoDs in the low to sub-pM (pg/mL) range for both BoNT-A and BoNT-B in buffer (i.e., simulating the analysis of ‘suspect' powder samples that can be dispersed in a buffer matrix for detection) and human serum (i.e., simulating a test for patient diagnosis). Ultimately, these results demonstrate the low-level detection of BoNTs using a SERS-based immunoassay that we believe can realistically be translated to a PON test for combatting biowarfare. 2.2 Experimental 2.2.1 Reagents and Materials Dithiobis (succinimidyl propionate) (DSP, 95%), methanol (≥99.9%) and acetonitrile (ACN) were purchased from Sigma Aldrich; gold shot (99.999%), from Alfa Aesar; sodium chloride and Tween-20, from Fisher Scientific; 200-proof ethanol (EtOH, ACS grade), from Pharmco-AAPER; modified Dulbecco's phosphate-buffered saline packs (PBS, 0.10 M, pH 7.8) and borate buffer packs (BB, 50 mM, pH 8.5), from Thermo Scientific; bovine serum albumin (BSA), from Jackson ImmunoResearch Laboratories, Inc.; gold nanoparticles (AuNP, 60 nm), from BBI; human serum standards (Acusera), from Randox Laboratories; C. botulinum BoNT-B Light Chain antibody (AF5420, α-BoNT-B pAb), from R&D Systems, Inc.; and polyclonal anti-botulinum neurotoxin type A (730A, α-BoNT-A pAb, chicken IgY), monoclonal anti-botulinum neurotoxin type A (731A, αBoNT-A mAb, mouse IgG), botulinum neurotoxin type A toxoid from Clostridium botulinum (133L, BoNT-A toxoid), and recombinant botulinum neurotoxin type B light chain (620A, BoNT-B), from List Biological Laboratories. The synthesis of the Raman reporter molecule, 5-5'-dithiobis (succinimidyl-2-nitrobenzoate) (DSNB) has previously 40 been described.30 Components that were received lyophilized were reconstituted and stored per manufacturer recommendations. All other chemicals were used as received. All aqueous solutions were prepared using water purified by passage through a Barnstead water polishing system to obtain ASTM type 1 water at a resistivity of 18.2 MΩ. 2.2.2 Capture Substrate Preparation Gold addresses (1 mm diameter) were fabricated by vapor deposition of a 200-nm coating of gold with a ~10-nm chromium adhesion layer on glass slides (25 mm × 75 mm). Before gold deposition, glass slides were cleaned by piranha etch (3:1 H2SO4 (95.098.0%):H2O2 (30%)), rinsed with methanol, and dried with nitrogen. An aluminum mask was used to produce the 1 mm gold addresses (3 × 8 array) on glass slides that form the bottom of an Array-It© plate as shown in Figure 2.1B. A DSP monolayer was formed on each gold address by immersion of the slide in an ethanolic solution of 0.10 mM DSP for 16 h. After chemical modification, the slides were thoroughly rinsed with ethanol, dried carefully with nitrogen, and fitted in the bottom of the Array-It© platform to form 96 separate gold addresses in an 8 × 12 array. Each of the following incubations used a sealing film (R&D systems) to reduce evaporation. First, a capture antibody layer was formed by the addition of 50 µL of 10 µg/mL α-BoNT-A mAb or α-BoNT-B pAb (PBS buffer, pH 7.4) to each well. After a 3 h incubation, the wells were rinsed 3 times by reverse pipet (i.e., addition of 100 µL of 10 mM PBS (pH 7.4) with 0.1% Tween-20 (PBS-T) followed by aspiration). The wells were then blocked with 100 µL of 1% (w/v) BSA in PBS to reduce nonspecific adsorption of the antigen calibrants on the gold address and well wall. After 1 h, the wells were rinsed as 41 described above with PBS-T. 2.2.3 Extrinsic Raman Label (ERL) Preparation The stock AuNP solution (60 nm) was adjusted to pH ~8.5 by the addition of 40 µL of 50 mM BB to 1 mL of AuNPs. After pH adjustment, the ERLs were functionalized with 10 µL of 1 mM DSNB in ACN and mixed on a rotating plate for 3 h at room temperature. The DSNB monolayer serves to couple tracer antibodies and acts as an RRM for Raman readout. A tracer antibody layer was formed by the addition of 5 µg of α-BoNT-A mAb or 2.5 µg of α-BoNT-B pAb with rotation for 3 h at room temperature. For stability and to decrease nonspecific adsorption, a 100 µL aliquot of 10% (w/v) BSA in 2 mM BB (pH 8.5) was added to each ERL solution and mixed on a rotating plate for an additional 30 min at room temperature. Finally, the ERLs were rinsed three times by centrifugation at 2029 g for 10 min where the supernatant was removed and the pellet was resuspended in 1.0 mL of 1% BSA in 2 mM BB for the first two rinses and 500 μL of 1% BSA in 2 mM BB for the final resuspension. The ionic strength of the solution was adjusted by the addition of 50 μL of 1.71 M NaCl to obtain 150 mM NaCl. The concentration of the ERLs was verified by UV-Vis to be ~4.9 × 1010 particles/mL; this is ~2× the stock concentration of ~2.60 × 1010 particles/mL.41 2.2.4 Immunoassay Procedure All assay steps following preparation of the capture surface and ERLs were done in a Biosafety Level 2 (BSL2) laboratory in a class III biosafety cabinet. Calibration standards were prepared by serial dilution of stock recombinant inactivated forms of BoNT- 42 A or BoNT-B in PBS or untreated, undiluted human serum. The standards (50 μL) were applied to the prepared capture surface with samples containing no BoNT as the blank standard. After a 3 h incubation at room temperature, the wells were rinsed as noted in Section 2.2.2 with 2 mM BB (150 mM NaCl, pH 8.5) with 0.1% Tween-20. An as-prepared ERL solution (50 µL) was added to each well for 16 h incubation at room temperature. The plate was rinsed following the rinse procedure in Section 2.2.2 using 2 mM BB (10 mM NaCl, pH 8.5) with 0.1% Tween-20. After allowing the plate to fully dry (> 1 h), the slide was removed from the Array-It© assembly and analyzed via Raman spectroscopy. 2.2.5 Raman Instrumentation Raman analysis was performed using a DXR Raman microscope (Thermo Scientific) equipped with a 632.8 nm HeNe laser. The spectra were collected in an array fashion (6 × 6 array consisting of 36 spectra equally spaced by 50 μm) using a 10x objective, 5.0 ± 0.1 mW laser power, 50 µm slit aperture, and 2 replicate scans with a 1.0 s integration time. Analysis included baseline correction before peak height determinations of the symmetric nitro stretch (νs(NO2)) at 1336 cm−1 from DSNB for quantification of the bound Ag. 2.3 Results and Discussion The following sections present the development of SERS-based immunoassays for the detection of two botulinum neurotoxin serotypes, BoNT-A and BoNT-B, in PBS and human serum. The two different sample matrices reflect the need to measure BoNT in powder samples by dissolution in a buffer to screen suspicious materials and in patient 43 samples with respect to medical diagnosis post-exposure. Section 2.3.1 describes the initial development of the SERS-based immunoassays in PBS, including the identification of the most effective pairs of antibodies and their levels of immobilization for antigen capture and labeling. Section 2.3.2 presents the dose response plots for detection of BoNT-A and BoNT-B in PBS. Section 2.3.3 extends this study by detailing the results of measurements for the two markers when spiked into human serum. This paper concludes with a brief discussion of work planned to extend these immunoassays toward the development of a PON test for live toxin. 2.3.1 Assay Tuning Developing a SERS-based sandwich immunoassay often requires that the steps in a procedure be fine-tuned to realize a high level of performance, including the preparation of the capture substrate and ERLs. This section details the optimization of assay components for the detection of BoNT-A and BoNT-B antigens by: (1) antibody screening, (2) capture antibody concentration and diluent composition, and (3) tracer antibody concentration. Initial screening of the capture and tracer antibodies is used to assess the ability of different immobilized antibodies to effectively capture and label the BoNT antigens. After identifying the most effective antibody pairs, the next step is to determine the requisite levels of each when used as immobilized reagents. For the optimization studies, the signals for a blank (i.e., 0 μg/mL) and high antigen concentration (i.e., 1 μg/mL BoNT) are compared and used in the calculation of a signal/blank ratio (RS/B). The RS/B provides an indication of the sensitivity of the assay, and is calculated by dividing the signal from a high antigen concentration (i.e., 1 μg/mL BoNT) 44 by the signal from a blank (i.e., 0 μg/mL BoNT) sample. A representative SERS spectra from the analysis of BoNT-A spiked in PBS is shown in Figure 2.2. Characteristic Raman bands for the nitro group and aromatic ring of the RRM, DSNB, are present: the symmetric nitro stretch (νs(NO2)) at ~1336 cm−1, nitro bending mode (δs(NO2)) at ~848 cm−1, and aromatic ring modes at ~1556 and ~1068 cm−1.30,38 The baseline-corrected peak height of νs(NO2) is used to indirectly quantify the amount of captured antigen. In a sandwich immunoassay, the antigen is both captured and tagged by specific antibodies (i.e., capture and tracer Ab, respectively). Antibodies for BoNTs can be specifically raised against the heavy or light chain. To this end, there are two types of antibodies for BoNT-A (a monoclonal antibody (mAb) that is raised against the light chain of BoNT-A and a polyclonal antibody (pAb) that is raised against the heavy chain of BoNTA) that are investigated as potential capture/tracer antibody pairs (mAb/pAb, pAb/pAb, mAb/mAb, and mAb/pAb). The pAb, by definition, can recognize multiple epitopes, while the mAb often imparts more specificity due to a comparatively strong monovalent affinity (i.e., it strongly recognizes only one epitope).42 To assess the capability of the Ab pairs (capture and tracer) to detect the BoNT Ag, the SERS signal was analyzed from a completed assay following similar procedures to those outlined in Section 2.2.2. The results from this study are shown in Figure 2.3A. In Figure 2.3A, use of the pAb as a tracer antibody improves detection of BoNT-A when compared to use of the mAb as a tracer antibody. When the pAb is used as a tracer antibody, the signals for BoNT-A are equal to ~3300 and ~ 2600 cps for the mAb and pAb as a capture antibody, respectively, while the signal is decreased to ≤ 400 cps when using 45 Figure 2.2. Representative Raman spectra of completed immunoassay (0 and 1 μg/mL BoNT-A in PBS). The spectra for 1 μg/mL BoNT-A shows the characteristic peaks of Raman reporter molecule, DSNB, from the ERL while the spectra for 0 μg/mL BoNT-A shows the instrument noise for a true blank. Both spectra are shown with a red dotted lines representing the baseline. The samples were prepared by the mAb as capture and mAb (0 μg/mL BoNT-A) or pAb as tracer (1 μg/mL BoNT-A). 46 Figure 2.3. SERS signal (νs(NO2)) for a completed assay for a PBS blank and PBS spiked with 1 μg/mL of (A) BoNT-A or (B) BoNT-B with antibody pairs (capture/tracer): monoclonal antibody (mAb) and polyclonal antibody (pAb). For BoNT-A, the mAb is raised against the light chain and the pAb is raised against the heavy chain. For BoNT-B, both the pAb and mAb are raised against the light chain. 47 the mAb as a tracer antibody regardless of the capture antibody. The increase in signal when using the pAb tracer antibody is likely due to more effective labeling of captured antigens due to multiple epitope availability (i.e., a steric effect). Using pAbs as a capture antibody is also likely to increase the signal due to an increase in the amount of captured Ag by the recognition of multiple epitopes. These data demonstrate this trend; however, the highest signal occurs with the mAb as a capture antibody and the pAb as a tracer antibody likely due to the recognition of more epitopes (i.e., recognition of epitopes in both the heavy (by the pAb) and light chains (by the mAb)). Use of the pAb as a tracer antibody also produces larger signals for the blank samples (i.e., 0 μg/mL BoNT-A or as-prepared PBS) when compared to use of the mAb as a tracer antibody. The signals for the blank samples in which the pAb serves as a tracer antibody are ~860 and ~100 cps with a pAb capture antibody and a mAb capture antibody, respectively. These blank signals are notably larger than those for the blank samples with the mAb tracer antibody (i.e., ~11 and ~9 cps with the pAb or the mAb as a capture antibody, respectively, which are on par with the instrument measurement noise of 18 ± 3 cps that is used in the RS/B analysis). The increased signal for the blank with the pAb as the tracer antibody follows past observations in that pAbs are more susceptible to nonspecific adsorption.42-43 A normalized comparison of the sensitivity of the antibody pairs can be evaluated by examining the RS/B values, calculated to 32 ± 6, 3 ± 0, 7 ± 2, and 22 ± 5 for the capture/detection Ab pairs of mAb/pAb, pAb/pAb, mAb/mAb, and pAb/mAb, respectively. The use of the pAb for both the capture and tracer Ab results in the lowest RS/B value, with a high degree of nonspecific adsorption dominating the response. While the pAb/mAb 48 antibody pair produces a higher RS/B value, the signal for BoNT-A is low compared to the mAb/pAb antibody pair. Thus, the mAb/pAb pair will be used for all subsequent optimization studies. Similar studies were carried out for BoNT-B with a pAb and mAb raised against the light chain of BoNT-B. The results, shown in Figure 2.3B, indicate that the use of the pAb as both a capture and tracer antibody performs the best, with a RS/B value of 183. All other capture/tracer Ab combinations had much smaller RS/B values (< 20). In addition to antibody selection, the capture Ab concentration, capture Ab diluent, and tracer Ab concentration were also fine-tuned. As mentioned previously, Ab concentrations are optimized to ensure adequate capture and labeling of the BoNT antigen. Although all of the Ab concentrations used in the coating process are in theoretical excess (> 3×), it is important to determine the relative amounts of reagent immobilized needed for efficient capture and labeling of the target antigen. We also examined the addition of a nonionic surfactant, Tween-20, in the diluent (PBS buffer) for the capture Ab. This follows reports showing a reduction in protein adsorption (e.g., 96-well plate walls, centrifuge tubes, pipet tips) by the addition of a nonionic surfactant.44 The fixed conditions used for each test are outlined below, to demonstrate the impact of each assay parameter on overall performance. First, the results for BoNT-A assays which used four different Ab concentrations in PBS to prepare the capture substrate and a fixed tracer Ab concentration of 5 μg/mL, are shown in Figure 2.4A. The RS/B values were calculated as 33 ± 7, 40 ± 11, 32 ± 6, and 21 ± 3 for 2.5, 5, 10, and 20 μg/mL capture Ab, respectively. As evident, the SERS signal for both the sample and blank increases with increasing capture Ab concentration. We 49 Figure 2.4. SERS signal (νs(NO2)) for 0 and 1 μg/mL BoNT-A. (A) capture Ab concentrations, (B) capture Ab diluents, and (C) tracer Ab concentrations. These experiments used the mAb/pAb (capture/tracer) pair. 50 expect an increase in the SERS signal intensity as capture Ab concentration increases until reaching a level in which the densest layer of capture Abs is formed. However, we also observe an increase in the degree of measured nonspecific adsorption with increases in the capture Ab concentration. We suspect the increase in nonspecific adsorption is due to changes in the orientation of the immobilized antibody that increases interactions with the ERL. Nonetheless, this results in an RS/B value peaking at a capture Ab concentration of 5 μg/mL with decreases for both the lower and higher Ab concentrations investigated. Thus, the optimal capture Ab concentration was set at 5 μg/mL. Second, the results for assays using PBS and PBS with Tween-20 (0.1%) as diluents for the solution used to prepare the capture Ab are summarized in Figure 2.4B. These experiments used a capture Ab concentration of 10 μg/mL and tracer Ab concentration of 5 μg/mL. The RS/B values were calculated as 13 ± 1 and 26 ± 4 for PBS and PBS-T, respectively. With the largest SERS signal and RS/B with the dilution of the capture Ab in PBS-T, it is evident that Tween-20 reduces nonspecific adsorption likely by inhibiting hydrophobic interactions between the underlying DSP-based monolayer and the capture antibody.45-48 Finally, results for a completed BoNT-A assay with differing tracer Ab concentrations are shown in Figure 2.4C, using a capture Ab concentration of 10 μg/mL in PBS. The RS/B values were calculated to be 20 ± 3, 32 ± 6, and 16 ± 3, for the tracer Ab concentrations of 2.5, 5, and 10 μg/mL, respectively. There is an increase in SERS signal intensity as the tracer Ab concentration increases, similar to the trend observed with increasing capture Ab concentration. We also find that the signal for the blank (0 μg/mL) increases as the tracer Ab concentration increases, which decreases the RS/B value at 10 51 μg/mL tracer Ab. We speculate that the increased nonspecific adsorption seen with the highest tracer Ab concentration is potentially due to the formation of multilayers of tracer Ab on the surface. The optimum tracer Ab concentration, with the largest RS/B, is obtained with 5 μg/mL tracer Ab. Similar experiments were performed for the assay of BoNT-B, which, to summarize, yielded the best RS/B value at a capture Ab concentration of 5 μg/mL and tracer Ab concentration of 2.5 μg/mL. 2.3.2 Dose Response Plots for BoNT-A and BoNT-B in PBS With the fine tuning studies complete, the guidance from the results can be combined to create dose response plots for BoNT-A and BoNT-B in PBS in order to test the capabilities of the SERS assay for both antigens. Dilutions of both BoNT-A and BoNTB in PBS were prepared for antigen concentrations of 0 to 50 ng/mL. The resulting SERS spectra and dose response plots using the baseline corrected SERS peak intensity of νs(NO2) relative to analytical concentration of BoNT-A or BoNT-B present are shown in Figure 2.5. The increase in the strength of νs(NO2) with increasing BoNT concentration is clearly evident in Figure 2.5A. These data were used to construct the dose response plot shown in Figure 2.5B. The LoD, which correlates with the lowest concentration of captured antigen that can be detected at a specified statistical level, is calculated from the signal from the blank plus three times the standard deviation from the blank and the slope (i.e., the sensitivity) from a least-squares linear fit. The LoDs are calculated as 2.9 ng/mL (19 pM) for BoNT-A in PBS and 84 pg/mL (0.6 pM) for BoNT-B in PBS. With the most conservative toxic dose of BoNTs (i.e., intravenous exposure with an LD50 of 1.3 ng/kg), we estimate a toxic 52 Figure 2.5. Plots from completed SERS-immunoassays for BoNT-A and BoNT-B in PBS. (A) Representative Raman spectra of completed immunoassays showing trend of signal strength of νs(NO2) with BoNT-B concentration. (B) Dose response plots for BoNT-A and BoNT-B antigens by SERS signal (νs(NO2)). The BoNTA immunoassay was performed with the BoNT-A mAb (5 μg/mL) as a capture Ab and BoNT-A pAb (5 μg/mL) as a tracer Ab. The BoNT-B immunoassay was performed with the BoNT-B pAb (5 μg/mL) as a capture Ab and BoNT-B pAb (2.5 μg/mL) as a tracer Ab. Error bars represent average of 3 samples. 53 amount of ~91 ng for an average adult of 70 kg. The detection of BoNTs in this sample matrix (i.e., PBS) is applicable to biosafety screening of powder samples which, for example, may involve dispersion in a simple buffer. Given this, the toxic amounts for relevant detection following dispersion in 150 μL (i.e., the volume needed for the immunoassay herein, run in triplicate) translates to ~600 ng/mL. Therefore, these LoDs are well within acceptable limits for relevant BoNT detection. Nevertheless, we expect that the use of an active form of BoNT-A versus the formaldehyde-inactivated form (i.e., toxoid) used herein will further improve the LoD.4952 In fact, there are numerous reports detailing the reduction in antigenicity of a toxoid after exposure to formaldehyde. These reports show that antibody recognition decreases with formaldehyde inactivation due to the loss of certain epitopes by major conformational changes. We imagine that, at the very least, the use of an active BoNT toxin will improve our LoD to ~0.5 ng/mL, given results from the manufacturer using the same Abs that showed a reduction in BoNT antigenicity by 5×.49 With the potential for field-deployable SERS detection, the simplification of BoNT detection to an immunoassay rather than live animal testing, and the reduction in time (~1 day), we expect these SERS-based immunoassays for BoNT-A and BoNT-B can fulfill the need for a PON test. Next, we show the results for a dose response plot in a more complex matrix, human serum, for potential use in the diagnosis of exposed patients. 54 2.3.3 Dose Response Plots for BoNT-A and BoNT-B in Human Serum The dose response plots from dilutions of BoNT-A and BoNT-B spiked in human serum from 0 to 25 ng/mL are shown in Figure 2.6A. The dose response plots, analyzed as described previously, give LoDs of 5 ng/mL (33 pM) for BoNT-A in human serum and 91 pg/mL (0.6 pM) for BoNT-B in human serum. These LoDs are nearly the same as those with PBS as a sample matrix, indicating that there is little interaction between serum components and the underlying capture surface or ERL. The relevant detection limit for BoNT in human serum taking the most conservative toxic dose (i.e., intravenous exposure with an LD50 of 1.3 ng/kg), the blood volume and weight of an average adult (i.e., 5 L and 70 kg), and potential degradation in the body is ≤ 20 pg/mL. Thus, detection of BoNT in human samples may require lower LoDs than even the mouse bioassay can achieve. Nevertheless, an immunoassay for BoNT-A using 5% milk in PBS as the blocking agent, versus 1% BSA in PBS that was used in all previous assays, resulted in a reduction of the LoD to 417 pg/mL (3 pM) BoNT-A in human serum, Figure 2.6B. This LoD represents a 10× reduction from that obtained in PBS. In addition, the LoD for the BoNTA immunoassay in human serum will likely decrease to ~75 pg/mL upon the use of an active form of BoNT-A, as previously described, due to an increase in antigenicity.49-53 Additional studies of blocking agents may reduce the LoD even further. 2.4 Conclusions This paper details the development of a SERS-based immunoassay for the detection of two BoNT serotypes, BoNT-A and BoNT-B. After optimization of the 55 Figure 2.6. Dose response plots from completed SERSimmunoassays for BoNT-A and BoNT-B in human serum. (A) Dose response plots for BoNT-A and BoNT-B antigens by SERS signal (νs(NO2)). These samples used 1% BSA in PBS as the blocking agent. (B) Dose response plot for BoNT-A using 5% milk in PBS as the blocking agent. Error bars represent average of 3 samples. 56 preparation of the capture substrate (i.e., capture antibody and concentration) and ERL (i.e., tracer antibody and concentration), the dose response plots gave LoDs of 2.9 ng/mL and 84 pg/mL for BoNT-A and BoNT-B spiked in PBS, respectively. With the BoNTs spiked in undiluted, untreated serum, the dose response plots gave LoDs of 5 ng/mL and 91 pg/mL for BoNT-A and BoNT-B, respectively. Further improvements with the use of 5% milk in PBS as a blocking agent resulted in a reduction of the LoD to 417 pg/mL BoNT-A in human serum. We expect the LoD for the BoNT-A immunoassay to decrease to ~75 pg/mL upon use of the activated form of BoNT-A as the antigen, given observations of reduced antigenicity with formaldehyde-inactivation BoNTs, as used herein.49 Thus, we anticipate achieving pg/mL (low to sub-pM) detection limits for BoNT-A and BoNT-B in both a simple sample matrix, PBS, for powder testing in biosafety screenings, and in a complex sample matrix, human serum, for patient diagnosis. These results show the low-level detection of the potential bioterrorism agent, BoNT, using SERS-based immunoassays. In addition, with recent advancements in the development of portable Raman instrumentation, these immunoassays have potential for rapid detection in the field (i.e., PON test) which is critical in the fight against biowarfare. The potential for translation of these SERS-based immunoassays to PON tests overcomes the major drawback (i.e., field deployability) with the mouse bioassay and others (e.g., ELISAs).25,54-57 With further development with the active forms of the BoNTs including multiplexation (i.e., alternate RRMs for simultaneous but differentiable detection of both BoNT-A and BoNT-B in a single well), we presume SERS-based detection can be an improvement to the current gold standard, the mouse bioassay, with regards to cost, time, and field deployability. 57 2.5 References (1) Botulism, Centers for Disease Control http://www.cdc.gov/botulism/ (accessed Nov 23, 2016). and Prevention. (2) Arnon, S. S.; Schechter, R.; Inglesby, T. V.; Henderson, D. A.; Bartlett, J. G.; Ascher, M. S.; Eitzen, E.; Fine, A. D.; Hauer, J.; Layton, M.; Lillibridge, S.; Osterholm, M. T.; O'Toole, T.; Parker, G.; Perl, T. M.; Russell, P. K.; Swerdlow, D. L.; Tonat, K. JAMA 2001, 285, 1059-1070. (3) Cheng, L. W.; Land, K. M.; Stanker, L. H. Current Methods for Detecting the Presence of Botulinum Neurotoxins in Food and Other Biological Samples. In Bioterrorism, Morse, S., Ed. Intech: 2012; pp 1-17. (4) Hill, K. K.; Smith, T. J.; Helma, C. H.; Ticknor, L. O.; Foley, B. T.; Svensson, R. T.; Brown, J. L.; Johnson, E. A.; Smith, L. A.; Okinaka, R. T.; Jackson, P. J.; Marks, J. D. J. Bacteriol. 2007, 189, 818-832. (5) Zilinskas, R. A. JAMA 1997, 278, 418-424. (6) Dhaked, R. K.; Singh, M. K.; Singh, P.; Gupta, P. Indian J. Med. Res. 2010, 132, 489-503. (7) Bigalke, H.; Rummel, A. Toxicology 2005, 214, 201-220. (8) Lindström, M.; Korkeala, H. Clin. Microbiol. Rev. 2006, 19, 298-314. (9) Solomon, H. M.; Lilly, T. J. Clostridium botulinum. In Bacteriological Analytical Manual, 8 ed.; U.S. Food & Drug Administration: 2001. (10) Ferreira, J. L.; Eliasberg, S. J.; Edmonds, P.; Harrison, M. A. J. Food. Prot. 2004, 67, 203-206. (11) Schiavo, G.; Matteoli, M.; Montecucco, C. Physiol. Rev. 2000, 80, 717-766. (12) Schiavo, G.; Benfenati, F.; Poulain, B.; Rossetto, O.; de Laureto, P. P.; DasGupta, B. R.; Montecucco, C. Nature 1992, 359, 832-835. (13) Swaminathan, S.; Eswaramoorthy, S. Nat. Struct. Biol. 2000, 7, 693-699. (14) García, M. T.; Gracia, I.; Duque, G.; de Lucas, A.; Rodríguez, J. F. Waste Management 2009, 29, 1814-1818. (15) Jacobson, M. J.; Lin, G.; Tepp, W.; Dupuy, J.; Stenmark, P.; Stevens, R. C.; Johnson, E. A. Appl. Environ. Microbiol. 2011, 77, 4217-4222. 58 (16) Smith, T. J.; Lou, J.; Geren, I. N.; Forsyth, C. M.; Tsai, R.; Laporte, S. L.; Tepp, W. H.; Bradshaw, M.; Johnson, E. A.; Smith, L. A.; Marks, J. D. Infec. Immun. 2005, 73, 5450-5470. (17) Smith, T. J.; Hill, K. K.; Foley, B. T.; Detter, J. C.; Munk, A. C.; Bruce, D. C.; Doggett, N. A.; Smith, L. A.; Marks, J. D.; Xie, G.; Brettin, T. S. PLoS One 2007, e1271. (18) Grate, J. W.; Ozanich, R. M., Jr.; Warner, M. G.; Marks, J. D.; Bruckner-Lea, C. J. Trends Anal. Chem. 2010, 29, 1137-1156. (19) Scarlatos, A.; Welt, B. A.; Cooper, B. Y.; Archer, D.; DeMarse, T.; Chau, K. V. J. Food Sci. 2005, 70, 121-130. (20) Sharma, S.; Zapatero-Rodriguez, J.; Estrela, P.; O'Kennedy, R. Biosensors 2015, 5, 577-601. (21) Szílagyi, M.; Rivera, V. R.; Neal, D.; Merrill, G. A.; Poli, M. A. Toxicon 2000, 38, 381-389. (22) Ekong, T. A. N.; McLellan, K.; Sesardic, D. J. Immunological Methods 1995, 180, 181-191. (23) Sharma, S. K.; Ferreira, J. L.; Eblen, B. S.; Whiting, R. C. Appl. Environ. Microbiol. 2006, 72, 1231-1238. (24) Scotcher, M. C.; Cheng, L. W.; Stanker, L. H. PLoS ONE 2010, 5, 1-10. (25) Dorner, B. G.; Zeleny, R.; Harju, K.; Hennekinne, J.-A.; Vanninen, P.; Schimmel, H.; Rummel, A. TrAC, Trends Anal. Chem. 2016, Advance online publication. doi: 10.1016/j.trac.2016.05.024. (26) Fach, P.; Micheau, P.; Mazuet, C.; Perelle, S.; Popoff, M. J. Applied Microbiology 2009, 107, 465-472. (27) Cheng, L. W.; Stanker, L. H. J. Agric. Food. Chem. 2013, 61, 755-760. (28) Koh, C.-Y.; Schaff, U. Y.; Piccini, M. E.; Stanker, L. H.; Cheng, L. W.; Ravichandran, E.; Singh, B.-R.; Sommer, G. J.; Singh, A. K. Anal. Chem. 2015, 87, 922-928. (29) Lee, J.; Brennan, M. B.; Wilton, R.; Rowland, C. E.; Rozhkova, E. A.; Forrester, S.; Hannah, D. C.; Carlson, J.; Shevchenko, E. V.; Schabacker, D. S.; Schaller, R. D. Nanoletters 2015, 15, 7161-7167. (30) Grubisha, D. S.; Lipert, R. J.; Park, H. Y.; Driskell, J.; Porter, M. D. Anal. Chem. 59 2003, 74, 5936-5945. (31) Grubor, N. M.; Shinar, R.; Jankowiak, R.; Porter, M. D.; Small, G. J. Biosens. Bioelectron. 2004, 19, 547-556. (32) Park, H.-Y.; Driskell, J. D.; Kwarta, K. M.; Lipert, R. J.; Porter, M. D.; Schoen, C.; Neill, J. D.; Ridpath, J. F. Top. Appl. Phys. 2006, 103, 427-446. (33) Driskell, J. D.; Kwarta, K. M.; Lipert, R. J.; Porter, M. D.; Neill, J. D.; Ridpath, J. F. Anal. Chem. 2005, 77, 6147-6154. (34) Driskell, J. D.; Lipert, R. J.; Porter, M. D. J. Phys. Chem. B 2006, 110, 1744417451. (35) Ni, J.; Lipert, R. J.; Dawson, B.; Porter, M. D. Anal. Chem. 1999, 71, 4903-4908. (36) Yakes, B. J.; Lipert, R. J.; Bannantine, J. P.; Porter, M. D. Clin. Vaccine Immunol. 2008, 15, 227-234. (37) Wang, G.; Lipert, R. J.; Jain, M.; Kaur, S.; Chakraboty, S.; Torres, M. P.; Batra, S. K.; Brand, R. E.; Porter, M. D. Anal. Chem. 2011, 83, 2554-2561. (38) Dufek, E. J.; Ehlert, B.; Granger, M. C.; Sandrock, T. M.; Legge, S. L.; Herrmann, M. G.; Mieikle, A. W.; Porter, M. D. Analyst 2010, 135, 2811-2817. (39) Han, Z.; Liu, H.; Meng, J.; Yang, L.; Liu, J.; Liu, J. Anal. Chem. 2015, 87, 95009506. (40) Zheng, J.; Pang, S.; Labuza, T. P.; He, L. Talanta 2014, 129, 79-85. (41) Haiss, W.; Thanh, N. T. K.; Aveyard, J.; Fernig, D. G. Anal. Chem. 2007, 79, 42154221. (42) Lipman, N. S.; Jackson, L. R.; Trudel, L. J.; Weis-Garcia, F. ILAR J. 2005, 46, 258268. (43) Ramos-Vara, J. A. Vet. Pathol. 2005, 42, 405-426. (44) Kenny, G. E.; Dunsmoor, C. L. Isr. J. Med. Sci. 1987, 23, 732-734. (45) Qu, Y.; Berghman, L. R.; Vandesande, F. Anal. Biochem. 1998, 259, 167-175. (46) Jenkins, S. H.; Heineman, W. R.; Halsall, H. B. Anal. Biochem. 1988, 168, 292299. (47) Steinitz, M. Anal. Biochem. 2000, 282, 232-238. 60 (48) Mohammad, K.; Esen, A. J. Immunol. Meth. 1989, 117, 141-145. (49) List Biological Laboratories, Inc. (2016). Botulinum Neurotoxin Type A Toxoid. https://www.listlabs.com/mediafiles/c_of_a/1333A1-927.pdf. (50) Anderson, J. H., Jr.; Lewis, G. E., Jr. Clinical Evaluation of Botulinum Toxins. In Biomedical Aspects of Botulism, Academic Press, Inc.: New York, 1987; pp 233246. (51) Jones, R. G. A.; Liu, Y.; Rigsby, P.; Sesardic, D. J. Immunol. Meth. 2008, 337, 4248. (52) Wright, G. G.; Duff, J. T.; Fiock, M. A.; Devlin, H. B.; Soderstrom, R. L. J. Immunol. 1960, 84, 384-389. (53) Torii, Y.; Tokumaru, Y.; Kawaguchi, S.; Izumi, N.; Maruyama, S.; Mukamoto, M.; Kozaki, S.; Takahashi, M. Vaccine 2002, 20, 2556-2561. (54) Lawn, S. D.; Peter Mwaba; Bates, M.; Piatek, A.; Alexander, H.; Marais, B. J.; Cuevas, L. E.; McHugh, T. D.; Zijenah, P. L.; Kapata, N.; Abubakar, I.; McNerney, R.; Hoelscher, P. M.; Memish, P. Z. A.; Migliori, P. G. B.; Kim, P.; Maeurer, P. M.; Schito, M.; Zumla, P. A. The Lancet Infectious Diseases 2013, 13, 349-361. (55) Yager, P.; Domingo, G. J.; Gerdes, J. Annu. Rev. Biomed. Eng. 2008, 10, 107-144. (56) Tighe, P. J.; Ryder, R. R.; Todd, I.; Fairclough, L. C. Proteomics:Clin. Appl. 2015, 9, 406-422. (57) Giljohann, D. A.; Mirkin, C. A. Nature 2009, 462, 461-464. CHAPTER 3 SUCCINIMIDYL ESTER SURFACE CHEMISTRY: IMPLICATIONS OF THE COMPETITION BETWEEN AMINOLYSIS AND HYDROLYSIS ON COVALENT PROTEIN IMMOBILIZATION Reprinted with permission from Lim, C. Y.; Owens, N. A.; Wampler, R. D.; Ying, Y.; Granger, J. H.; Porter, M. D.; Takahashi, M.; Shimazu, K., Succinimidyl Ester Surface Chemistry: Implications of the Competition between Aminolysis and Hydrolysis on Covalent Protein Immobilization. Langmuir 2014, 30 (43), 12868-12878. Copyright 2014 American Chemical Society. This material is available under the terms of the ACS AuthorChoice license. http://pubs.acs.org/doi/abs/10.1021/la503439g 3.1 Introduction N-hydroxysuccinimide (NHS) and other activated esters are often used as coupling agents to covalently tether antibodies, enzymes, peptides, and other biomaterials to surfaces for use in bioanalytical sensors.1-3 There are two common strategies for fabricating NHS-functionalized surfaces: (1) reaction of surface carboxylate groups with NHS and Nethyl-N-(3-dimethylamino)propyl carbodiimide (EDAC);4 and (2) direct derivatization of gold, silicon, and other surfaces with an NHS-containing coating.5 In both cases, coupling 62 is achieved by treating the NHS-activated surface with a solution of a primary aminecontaining reactant. This step (i.e., aminolysis) forms amide linkages with the sterically accessible amines of the reactant. We have used this methodology to construct antibody-modified surfaces for the selective capture and labeling of markers in immunoassays for the detection of infectious diseases, cancer, and nutrient deficiencies.6-7 Scheme 3.1 summarizes our two-step procedure. Step 1 forms the active ester surface by immersion of a gold substrate in a 0.10 mM solution of 3,3'-dithiobis (succinimidyl) propionate (DSP), i.e., Lomant's Reagent, in ethanol.8 This step chemisorbs a monolayer of the gold-bound thiolate of DSP, 3-Nhydroxysuccinimidyl propane thiolate.9 Step 2 reacts the NHS-activated substrate in a buffered, protein-containing solution. We typically carry out this step in borate buffer (pH 8.50, 50 mM), which, as recently reviewed,3 is similar to the conditions commonly recommended for this reaction. This step is designed to immobilize proteins on the surface via an amide linkage. Along with steric effects,10 there are two more factors to consider when using aminolysis for surface immobilization. First, the amine groups of the biological reactant act as nucleophiles, viz. deprotonated amines.11 Thus, the pH of the coupling solution controls the nucleophile concentration. Second, the coupling conditions should not alter the inherent reactivity of the biomolecule (e.g., the binding affinity of an antibody to an antigen). As such, these reactions are typically carried out in an aqueous solution at a pH and ionic strength close to those found physiologically (e.g., pH ~6-9 and 150 mM NaCl) in order to preserve the tertiary structure of proteins.12-13 This issue is central in the immobilization of antibodies in that structural denaturation may reduce the affinity for Scheme 3.1. Method for covalent coupling of capture antibody layer in immunoassay using an NHS terminated monolayer and primary amines in antibodies. Step 1 forms the NHS-terminated monolayer on gold by chemisorption of DSP. Step 2 reacts the NHS-terminated monolayer on gold with an antibody in aqueous buffer (aminolysis); the competing reaction with hydroxide ions (hydrolysis) is shown in parallel. 63 64 antigen binding.10 The use of physiological reaction conditions, however, can induce the hydrolysis of the activated ester group, which would reduce the number of reactive sites for aminolytic coupling and decrease the efficiency of the linking chemistry.3,14 Thus, the coupling conditions must strike a balance between the rates of the two reactions, aminolysis and hydrolysis, both of which are affected by pH, buffer composition, reactant concentration, temperature, and other factors.1,3 There is a large body of evidence, including that from our laboratory,15 which supports the effectiveness of Step 2 in Scheme 3.1. Much of this evidence is based on measurements of the biological or chemical activity of the immobilized species. Examples include the colorimetric16-17 and electrochemical detection18 of the activity of an immobilized enzyme or a tethered redox molecule. More so, infrared external reflection spectroscopy (IR-ERS)19 and X-ray photoelectron spectroscopy (XPS)20 have been used in efforts to detect the formation of an amide linkage after protein immobilization. However, while supporting the presence of a protein layer, a clear interpretation of these measurements is compromised by the difficulty in identifying the presence of the aminolytically formed amide linkage vis-à-vis the large number of amide groups inherent in the protein itself.21 As a consequence, studies have used monoamines (e.g., lysine22 and ethylamine23) as reactive mimics to simplify the spectroscopic analysis and, by inference, assess the effectiveness of the aminolysis reaction when using proteins. Nonetheless, the extension of these results to an exacting determination of protein tethering is subject to the inability to rule out the presence of proteins due to nonspecific adsorption.24 This paper reports on the results of an investigation of the effectiveness of Step 2 in Scheme 3.1 by an examination of the competitive rates of interfacial aminolysis and base 65 hydrolysis reactions of the DSP-based monolayer on gold substrates. The heterogeneous base hydrolysis reaction rate was examined using borate buffer (pH 8.50, 50 mM). The rate of aminolysis measured in the same buffer system used ethylamine as a mimic of the reactivity with respect to protein immobilization. These findings are presented and discussed after we describe the results of characterization of the DSP-based monolayer on gold by IR-ERS, XPS, electrochemical techniques, and contact angle measurements. Ultimately, these data are used to estimate the effectiveness of the aminolytic coupling reaction in Scheme 3.1 as it applies to immobilizing proteins on a biosensor surface. 3.2 Experimental 3.2.1 Reagents and Materials Dithiobis (succinimidyl propionate) (DSP, > 95%), ethylamine hydrochloride, 1,4dioxane, and borate buffer (50 mM, pH 8.50) were obtained from Fisher Scientific; 200 proof ethanol from AAPER Pharmco; potassium bromide (KBr, 99+%, IR grade) and tetramethylammonium chloride (TMAC, > 98%) from Acros Organics; N-(benzoyloxy) succinimide (NBS) and gold shot (99.995%) from Alfa Aesar; and 1-octadecane-d37-thiol from C/D/N Isotopes. 1,4-Dioxane was dried over molecular sieves (EMD) prior to use. All other chemicals were used as received. Aqueous solutions were prepared using water purified by passage through a Barnstead water polishing system to obtain water at a resistivity of 18.2 MΩ. 66 3.2.2 Monolayer Preparation Substrates used to prepare the DSP-based monolayers were formed on glass slides (25 x 75 mm). The glass slides were first cleaned by immersion in piranha etch (3:1 H2SO4:H2O2 (30%)) for 10 min, followed by rinsing with copious amounts of high purity water and subsequently drying under a stream of nitrogen. Caution: Piranha etch reacts violently with most organic materials and must be used and handled with extreme care. The glass slides were then coated by the vapor deposition of a 20 nm chromium adhesion layer and a 200 nm gold film. The gold substrates were then immersed in a 0.10 mM ethanolic solution of DSP. After 16 h, the slides were rinsed with high purity ethanol and dried under a stream of nitrogen. 3.2.3 Infrared Spectroscopy (IRS) IR spectra were obtained using a Nicolet Magna 850 Fourier transform infrared spectrometer equipped with a liquid nitrogen-cooled mercury cadmium telluride detector. Spectra were collected in a nitrogen atmosphere by co-adding 512 scans at a resolution of 4 cm−1. Transmission spectra for powdered samples were taken after dispersion in high purity KBr pellets. External reflection spectra (IR-ERS) used p-polarized light incident at 82° from the surface normal and are reported as -log R/Ro, where R is the spectrum of the sample and Ro is that of an octadecanethiolate-d37 reference monolayer on gold. 3.2.4 X-ray Photoelectron Spectroscopy (XPS) XPS measurements used a Kratos Axis Ultra DLD XPS with a monochromatic Al X-ray source at an incidence angle of 60°. Spectra were obtained with a 700 x 300 μm 67 hybrid slot size, 12 mA emission current, and 15 kV anode potential; the instrument pressure ranged from 10−9 to 10−10 torr. Survey scans were collected at a 160 eV pass energy and 1 eV step size. High resolution scans used a 40 eV pass energy and 0.1 eV step size, with a dwell time of 300 ms in the C(1s) and O(1s) spectral regions and 1.2 s in the N(1s) and S(2p) spectral regions. All binding energies are reported with respect to the Au(4f7/5) emission band at 84.0 eV.25 Band shapes were modeled using Gaussian-Lorentzian profiles and a linear or Shirley background subtraction.26 For spectral deconvolution, fits in O(1s), C(1s), and N(1s) regions were constrained only for band shape; the fits were set to find the minimum number of bands that gave an overall residual of less than 3%. The S(2p) bands were constrained to full widths at half-maximum of 0.9 to 1.3 eV, an integrated S(2p1/2) band intensity twice that of the S(2p3/2) band, and separation between the two bands of 1.2 eV.27 3.2.5 Electrochemistry Electrochemical measurements used a three-electrode cell with the gold substrates as the working electrode, platinized platinum foil as the auxiliary electrode, and Ag/AgCl (sat'd KCl) as the reference electrode; all potentials are reported against this electrode. The geometric area of the working electrode was 0.65 cm2 at a roughness factor of 1.40.28-29 Voltammetric scans were performed in 0.5 M KOH after sparging with argon (~30 min). 3.2.6 UV-Vis Spectroscopy UV-Vis measurements were performed with a Cary 5000 UV-VIS-NIR spectrophotometer at room temperature using a 1.00 cm quartz cuvette. Scans were 68 collected between 350 and 220 nm with an integration time of 0.10 s and a spectral bandwidth of 2 nm. Kinetic measurements recorded the solution absorbance at 260 nm at 0.10 s intervals; the first data point was collected ~10 s after solution mixing. 3.2.7 Contact Angle Measurements Contact angles were measured with a Dataphysics OCA15EC instrument at room temperature. Deionized water was used as the probe liquid. The advancing, θa, and receding, θr, contact angles were measured by increasing or decreasing the volume of the droplet, respectively. 3.2.8 Adlayer Surface Concentration by XPS The surface concentration for sulfur was determined using a previously documented procedure.30-31 Briefly, the system was modeled as a uniform film with a thickness (τ) determined from signal attenuation. For film thickness calculations, the signal from the Au(4f) bands of an unmodified gold substrate was used as the reference response 0 (𝐼𝐼𝐴𝐴𝐴𝐴 ), with the intensity of the signal gold sample (𝐼𝐼𝐴𝐴𝐴𝐴 ) given by: 0 𝐼𝐼𝐴𝐴𝐴𝐴 = 𝐼𝐼𝐴𝐴𝐴𝐴 𝑒𝑒𝑒𝑒𝑒𝑒 − 𝜏𝜏 𝐿𝐿𝐴𝐴𝐴𝐴 (3.1) where 𝐿𝐿𝐴𝐴𝐴𝐴 , the effective attenuation length, was determined using the NIST Standard Reference Database 82 software.32 With the calculated film thickness, the elemental concentration of sulfur, 𝑁𝑁𝑆𝑆 , was found from: 𝜏𝜏 𝐼𝐼𝑋𝑋 𝑇𝑇𝐴𝐴𝐴𝐴 𝜎𝜎𝐴𝐴𝐴𝐴 𝐿𝐿𝑄𝑄𝐴𝐴𝐴𝐴 𝑒𝑒𝑒𝑒𝑒𝑒 − 𝐿𝐿𝐴𝐴𝐴𝐴 𝑁𝑁𝑆𝑆 = 𝑁𝑁𝐴𝐴𝐴𝐴 𝐼𝐼𝐴𝐴𝐴𝐴 𝑇𝑇𝑆𝑆 𝜎𝜎𝑆𝑆 𝐿𝐿𝑄𝑄𝑆𝑆 1 − 𝑒𝑒𝑒𝑒𝑒𝑒 − 𝜏𝜏 𝐿𝐿𝑆𝑆 (3.2) 69 In Equation 3.2, 𝑁𝑁𝐴𝐴𝐴𝐴 is the atomic density of gold (19.28 g/cm3), 𝑇𝑇𝐴𝐴𝐴𝐴 and 𝑇𝑇𝑆𝑆 are the respective analyzer transmission functions, 𝜎𝜎𝐴𝐴𝐴𝐴 and 𝜎𝜎𝑆𝑆 are the respective photoelectric 𝑄𝑄 𝑄𝑄 cross sections, and 𝐿𝐿𝐴𝐴𝐴𝐴 and 𝐿𝐿𝑆𝑆 are the appropriate effective attenuation lengths. The value of 𝑁𝑁𝑆𝑆 is then used to calculate the surface concentration (𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁,𝑋𝑋𝑋𝑋𝑋𝑋 ) as: 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁,𝑋𝑋𝑋𝑋𝑋𝑋 = 𝑁𝑁𝑆𝑆 𝜏𝜏 (3.3) 3.2.9 Interfacial Kinetics For the studies of hydrolysis kinetics, DSP-based monolayers on gold-coated glass slides were immersed in a large volume (~60 mL) of 50 mM borate buffer (pH 8.50, buffer capacity of 15 mM) for different periods of time at room temperature. At the end of each time period, each slide was thoroughly rinsed with distilled water, dried with nitrogen, and analyzed via IR-ERS; the rinsing and drying steps collectively required ~20 s. It is assumed that this process quenches the reaction in less than 20 s. The slides were then reimmersed in the borate buffer solution until the end of the next time segment and the rinsing, drying, and analysis procedures were repeated. The same procedure and conditions were used for aminolysis, after the addition of 500 mM ethylamine. 3.3 Results and Discussion This paper presents findings from an investigation of the reactivity (base hydrolysis and aminolysis) of the NHS-terminated monolayer formed by the spontaneous adsorption of DSP on gold. It describes results from: (1) characterizations of the as-formed monolayer by IR-ERS, XPS, electrochemistry, and contact angle measurements to establish the architecture of the adlayer; and (2) IR-ERS reaction rate studies of the adlayer in borate buffer (pH 8.50) with and without ethylamine to serve as a basis for a comparison of the 70 respective rates of the base hydrolysis and aminolysis of the terminal NHS group of the adlayer. The reaction rate data, which include measurements of the homogeneous base hydrolysis of the adsorbate precursor (DSP) in aqueous base, are then examined within the context of establishing the effectiveness of this pathway to protein tethering. 3.3.1 Characterization of As-Formed Adlayer The composition and surface concentration of the as-prepared adlayer were characterized via IR-ERS, XPS, and electrochemistry. IR spectra of DSP and NHS dispersed as a powder in a KBr pellet and of the as-formed DSP-based monolayer on gold are shown in Figure 3.1 between 2000 and 1000 cm−1. Band assignments are listed in Table 3.1. The spectral features most characteristic of NHS are the symmetric carbonyl stretch (νs(C=O)) at 1780 cm−1 and the broader, much stronger envelope of asymmetric carbonyl stretches (νa(C=O)) between ~1750 and 1675 cm−1; this envelope reflects contributions from hydrogen bonding and other electronic interactions in the solid phase.22,33 Bands at lower energy (e.g., the asymmetric C-N-C stretch (νa(C-N-C)) at 1219 cm−1 and the C-O stretch (ν(C-O)) at 1078 cm−1) are additional succinimidyl group identifiers.22 All of the above bands are present in the spectra for DSP in KBr and the resulting adlayer. Importantly, the highest energy feature in these two spectra, which is assigned to the carbonyl stretch (ν(C=O)) of the ester linkage between NHS and the alkyl chain of DSP, is expectedly absent in the spectrum for NHS. There are also barely detectable features at much higher energy that correspond to the asymmetric (νa(CH2)) and symmetric (νa(CH2)) stretches of the methylene groups in the alkyl chain at 2922 and 2852 cm−1, respectively (data not shown).22 There is little evidence (i.e., clear differences between the widths of the 71 Figure 3.1. Infrared spectra for NHS and DSP dispersed in KBr and for the DSP-based monolayer chemisorbed on gold. 72 Table 3.1. Infrared spectral peak positions and band assignments for DSP and NHS dispersed in KBr and for the DSP-based adlayer on gold.22,33 Mode Assignment Description ν(C=O) νs(C=O) νa(C=O) δ (CH2) νs(C-N-C) νa(C-N-C) ν(C-O) carbonyl stretch of ester symmetric carbonyl stretch of NHS asymmetric carbonyl stretch of NHS methylene scissors deformation symmetric CNC stretch of NHS asymmetric CNC stretch of NHS N-C-O of succinimide Peak Position (cm−1) NHS-KBr DSP-KBr DSP/Au 1814 1820 1780 1788 1787 1750-1675 1748 1426 1433 1464 1307 1373 1378 1219 1216 1215 1078 1075 1074 73 two KBr spectra with respect to that for the DSP- based adlayer) of interactions between neighboring NHS groups in the adlayer. These results confirm the presence of the DSPbased adlayer on gold. The XPS results (spectra a in Figure 3.2 and Table 3.2) provide further evidence for the presence of the DSP-derived adlayer. In the O(1s) binding energy region, the two observable bands can be assigned to the C=O (532.1 eV) and C-O (534.4 eV) groups of the NHS terminus.34 The two bands present in the C(1s) binding energy region at 284.8 and 288.7 eV are assigned to the methylene and carbonyl carbons, respectively.34-35 The asymmetry in the methylene band is ascribed to the methylene carbon adjacent to the carbonyl carbon.35 In the S(2p) binding region, there are two bands for the S(2p) couplet: S(2p3/2) at 161.9 eV and S(2p1/2) at 163.1 eV. The positions of the two bands in the S(2p) couplet confirm the presence of the gold-bound thiolate formed in the chemisorption of thiols and disulfides on gold.36 Finally, the N(1s) binding energy region has a single band at 401.6 eV, which is diagnostic of the NHS nitrogen.20,34 The remaining spectra in Figure 3.2 are discussed in the next section. The surface concentration, 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁,𝑋𝑋𝑋𝑋𝑋𝑋 , of the DSP-based monolayer was calculated by using the spectral intensities from the XPS S(2p) region and the procedures accompanying Equations 3.1-3.3.30-31 This analysis gives a value for 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁,𝑋𝑋𝑋𝑋𝑋𝑋 of 6.44 ± 0.97 × 10−10 moles/cm2 (n = 5). For comparison, an electrochemical determination of 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁 , 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁,𝐸𝐸𝐸𝐸 , was carried out by measuring the charge required for the one-electron reductive desorption of the adlayer in 0.50 M KOH (aq).28,38-40 A representative linear sweep voltammogram at 0.100 V/s from these measurements is shown in Figure 3.3. Two large cathodic waves are evident, a sharper feature with a peak-current maximum at 74 Figure 3.2. XPS spectra of: (a) as-formed DSP-based adlayer on gold; (b) hydrolyzed DSP-based adlayer on gold after overnight (greater than 16 hours) immersion in 50 mM borate buffer (pH 8.50), and (c) aminolyzed DSP-based adlayer on gold after immersion in 500 mM ethylamine in 50 mM borate buffer (pH 8.50). All band intensities (counts per second - CPS) have been normalized to the Au (4f)7/2 band. The residuals (not shown) from the deconvolution analysis for S(2p), C(1s), N(1s), and O(1s) are 1.2, 2.6, 1.8, and 2.2%, respectively. 75 Table 3.2. XPS band assignments and positions for as-prepared and reacted DSPbased monolayers.20,34-37 Band Position (eV)a Core Level Assignment asafter after prepared hydrolysis aminolysis O(1s) carbonyl oxygen 532.1 532.0 531.1 O(1s) NHS ester oxygen 534.4 C(1s) methylene carbonb 284.6 284.5 284.5 C(1s) methylene carbon next to carbonyl carbon 285.3 285.4 286.0 C(1s) carbonyl carbon 288.7 288.6 287.5 S(2p3/2) gold-bound thiolate 161.9 162.0 162.0 S(2p1/2) gold-bound thiolate 163.1 163.3 163.2 N(1s) succinimidyl nitrogen 401.6 N(1s) amide nitrogen 399.5 a The uncertainty in the band positions is ±0.1 eV in the S(2p) region and up to 0.4 eV in the O(1s), C(1s) and N(1s) regions. bAssigned to methylene groups but unable to distinguish between those in the alkyl chains and the NHS group. 76 Figure 3.3. Linear voltammetric sweep (scan rate: 0.100 V/s) in 0.50 M KOH (aq) for the reductive desorption of the DSP-based adlayer on gold. 77 −0.82 V and a broader feature at −1.05 V. The presence of multiple waves is consistent with differences in the sorptive strength of the thiolate to different crystallographic binding sites on a polycrystalline gold surface.41 Using a linear baseline approximation to estimate the contribution of the double layer charging current, integration of the area under the current-potential curves for 7 different samples yields an average desorption charge of 69.7 ± 5.4 μC/cm2. After accounting for a roughness factor of 1.40,28-29 this translates to a surface area normalized charge of 49.8 ± 3.9 μC/cm2 or a 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁,𝐸𝐸𝐸𝐸 of 5.16 ± 0.40 × 10−10 mol/cm2. This value agrees well with the roughness factor corrected value for 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁,𝑋𝑋𝑋𝑋𝑋𝑋 of 5.86 ± 0.88 × 10−10 mol/cm2. These two values differ by less than 15% from those reported by other laboratories.5,18 Furthermore, both values are lower than that expected for the (√3x√3)R30° adlayer formed by n-alkanethiols on Au(111), which reflects the difference in the packing density of the bulky NHS terminal group.42-43 3.3.2 Compositional Analysis of Reacted Adlayers XPS was used to confirm the identity of the surface-bound reaction products for hydrolysis and aminolysis (spectra b and c in Figure 3.2 and Table 3.2). Immersion of the adlayer for ~16 h in either borate buffer (pH 8.50, 50 mM) or 500 mM ethylamine in the same borate buffer resulted in the following differences in the adlayer. First, the bands associated with the NHS ester (534.4 eV in the O(1s) and 401.6 eV in the N(1s) spectral regions) are no longer detectable after either treatment, indicative of removal of the NHS group. Second, the presence of the aminolysis reaction product is indicated by the appearance of the N(1s) band at 399.5 eV, which is diagnostic of an amide nitrogen.20,34,37 The two bands in the C(1s) binding energy region, representative of methylene and 78 carbonyl carbons, are still present after each of the treatments, but their relative intensities have changed in accordance with the expected reaction products. Lastly, the strength of the S(2p) bands remains unchanged, confirming the stability of the gold-bound adlayer under the reaction conditions used herein. These results verify the presence of the expected surface reaction products and are supported by the IR-ERS data in the next section. 3.3.3 Adlayer Base Hydrolysis The rate of the alkaline hydrolysis for the NHS-activated ester monolayer was monitored as a function of immersion time in borate buffer (pH 8.50, 50 mM) by IR-ERS. As shown in Figure 3.4, the temporal evolution of the spectra is indicative of the progression in the hydrolytic loss of the NHS group. The νa(C=O) at 1748 cm−1, for example, decreases in strength by more than 50% in less than 480 s; this feature is virtually undetectable after 14 min. Other spectral features (e.g., N-C-O band at 1074 cm−1 and the carboxylate vibration band at 1265 cm−1) follow this trend but are too weak in strength to be used for kinetic analysis. The spectra in Figure 3.4 were analyzed to more fully characterize the hydrolysis kinetics by determining the temporal decrease in the strength of νa(C=O) of the adlayer. The bimolecular reaction rate for the base hydrolysis of the adlayer can be written as: 𝑑𝑑 Γ𝑁𝑁𝑁𝑁𝑁𝑁 = − 𝑘𝑘ℎ Γ𝑁𝑁𝑁𝑁𝑁𝑁 [𝑂𝑂𝐻𝐻 − ] 𝑑𝑑𝑑𝑑 (3.4) where Γ𝑁𝑁𝑁𝑁𝑁𝑁 is the surface concentration of the NHS reactive group (moles/cm2), 𝑘𝑘ℎ is the second-order heterogeneous rate constant for hydrolysis (M−1 s−1), [𝑂𝑂𝐻𝐻 − ] is the hydroxide ion concentration in bulk solution (M), and 𝑡𝑡 is time (s). Assuming that the conditions for 79 Figure 3.4. Infrared spectra of the DSP-based adlayer after different immersion times in 50 mM borate buffer (pH 8.50). 80 a pseudo first-order reaction can be applied by use of a buffered alkaline solution, Equation 3.4 can be simplified to: 𝑑𝑑Γ𝑁𝑁𝑁𝑁𝑁𝑁 = −𝑘𝑘ℎ′ Γ𝑁𝑁𝑁𝑁𝑁𝑁 𝑑𝑑𝑑𝑑 (3.5) where 𝑘𝑘ℎ′ is the pseudo first-order heterogeneous reaction rate constant (s−1). Integration with the appropriate limits yields: ′ (3.6) Γ𝑁𝑁𝑁𝑁𝑁𝑁 (𝑡𝑡) = Γ𝑁𝑁𝑁𝑁𝑁𝑁(𝑡𝑡=0) 𝑒𝑒 −𝑘𝑘ℎ𝑡𝑡 The value of 𝑘𝑘ℎ′ can be determined from the slope of a plot of −ln(1-x) versus 𝑡𝑡 where x represents the extent of reaction, which is calculated from Γ Γ𝑁𝑁𝑁𝑁𝑁𝑁 (t) . This analysis assumes 𝑁𝑁𝑁𝑁𝑁𝑁 (t=0) that Γ𝑁𝑁𝑁𝑁𝑁𝑁 is directly proportional to the strength of νa(C=O) at 1748 cm−1. Due to the socalled "infrared metal surface selection rule," this proportionality holds in IR-ERS only if the orientation of the transition dipole moment for this vibrational mode is constant throughout the reaction.44 We have invoked this assumption in the analysis of these data.45 The results of this analysis are shown in Figure 3.5, which can be used to determine the time required for 50% conversion (t50%) of 380 ± 40 s. In contrast to the linear relationship expected for a pseudo first-order reaction, this plot has a nonlinear shape. There are a few possible interpretations for this dependence, including a difference in reaction order that can change with time and a reaction with two different rates that are connected in series. The former can be ruled out as the hydroxide ion is fixed by the use of 50 mM borate buffer (Experimental Section), which maintains the conditions of a pseudo first-order reaction. Thus, the shape of the kinetic plot is considered to originate from a progression of two reactions in series. An interpretation of this type of rate profile, known as the Avrami or Johnson-MehlAvrami-Kolmogorov (JMAK) theorem, was originally proposed to describe the phase 81 Figure 3.5. Kinetic plot for hydrolysis of the DSP-based monolayer in 50 mM borate buffer (pH 8.50). The dotted annotation between the experimental data points serve only as a guide to the eye. Some of the error bars are close to the size of the data point. 82 transitions in solids where a reaction at a surface can be divided into the three kinetic regimes depicted in Figure 3.6.46-48 The three regimes are: (1) a slow reaction rate at short times (t1 and t2) due primarily to the formation of reacting nuclei (i.e., the initiation stage); (2) a reaction interval at intermediate times (t3 and t4) with a more rapid and relatively constant rate in which previously formed nuclei grow in size and eventually form overlapping domains of reacted material (i.e., the bulk transformation stage); and (3) a period at long times (t5) in which the rate slows as the reaction approaches completion. Indeed, the intermediate region (i.e., bulk transformation stage) in JMAK theory spans the extent of reaction from 0.15 < x < 0.8, which corresponds to the time period from roughly 360 s to 660 s in Figure 3.5.49 We view the kinetic plot in Figure 3.5 to be comprised of three overlapping kinetic regimes, but with limited data to fully characterize the final regime due to a decrease in signal to noise such that the peak is no longer quantified with great certainty. In the initial stage (0 to ~180 s), the reaction proceeds slowly, reflecting the role of a nucleation type of process in which the hydrolytic removal of NHS groups reduces steric barriers to the attack of hydroxide ions on the acyl carbon of esters at the edge of the nuclei. As time and the size of these growing domains increase (roughly 360 to 660 s), the rate of reaction undergoes an increase due to a greater number of accessible surface reactants. During this stage, the rate of the reaction becomes close to constant, which is indicative of an immeasurable change in the number of NHS groups at the domain boundaries. In the third and final regime (> 720 s), the rate slows as the surface reactant is exhaustively consumed. Based on the above interpretation, the reactivity of the adlayer was analyzed for our purposes via a pseudo first-order rate law in the bulk transformation stage. This analysis Figure 3.6. Schematic of an Avrami transformation from unreacted material at t0 in which the rate is slow at small times (t1 and t2) due to the formation and initial growth of nuclei (black dots), i.e., the initiation stage; (2) more rapid at times t3 and t4 due to reacting nuclei (grey), i.e., the bulk transformation stage; and (3) low at long times (t5) due to decreased amount of starting material (white). Dotted outlines show progression of reaction from t3 to t5. 83 84 used a linear fit of the data between 360 and 660 s and gave a pseudo first-order reaction rate constant, 𝑘𝑘ℎ′ , of 4.6 ± 0.3 × 10−3 s−1, and a second-order reaction rate constant, 𝑘𝑘ℎ , of 1.5 ± 0.1 × 103 M−1s−1 (n = 3). This value for 𝑘𝑘ℎ is much higher than that previously reported for this system in aqueous alkaline (1.00 × 10−3 M NaOH) solutions (6.1 ± 1.1 × 10−1 M−1s−1).50 While the origin(s) of the larger value found herein is presently unclear, we suspect that it reflects, at least in part, a difference in the number and/or size of structural defects in the adlayer, which is supported by wettability data.35,51-53 For this adlayer system, Dordi et al. previously reported advancing (θa) and receding (θr) contact angles for water of 60 ± 2° and 39 ± 2°, respectively. We measured a comparable θa of 59 ± 2° (n = 6), but a much lower θr of 29 ± 5° (n = 6) which suggests that our adlayer is not as tightly packed as that in the earlier work. Experiments are now being designed to test for additional possible origins of this difference. 3.3.4 Homogeneous Base Hydrolysis For comparative purposes, we measured the rates of the base hydrolysis for DSP and NBS (N-(benzoyloxy)succinimide) in aqueous solution. NBS served as a model for connection to the earlier work by Cline and Hanna which investigated the base hydrolysis of several types of NHS esters in both aprotic and aqueous solutions.54 These UV-Vis measurements monitored the reaction by following the appearance of the NHS anion with time. This anion adsorbs in the UV spectral region and has an absorbance maximum at 260 nm with a molar absorptivity of 9700 M−1cm−1;55 the neutral form of NHS absorbs much deeper in the UV region. By monitoring the reaction of NBS under the conditions used by Cline and Hanna (20% 1,4-dioxane and an ionic strength of 1.0 M through the addition of 85 tetramethylammonium chloride, (TMAC)), we determined a second-order homogeneous reaction rate constant, 𝑘𝑘ℎ,𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 for NBS in 20% dioxane, of 9.2 ± 0.7 × 101 M−1s−1 (n = 6), which is in good agreement with the 8.7 × 101 M−1s−1 value reported earlier.54 The reaction conditions used in subsequent experiments were analogous to those described previously for the interfacial experiments, borate buffer (50 mM, pH 8.50) with 1% 1,4-dioxane added for DSP solubility. Figure 3.7 shows the spectrophotometric data and the extent of reaction analysis for a 0.10 mM solution of DSP under these conditions. The absorbance initially increases rapidly, slowing to a limiting value as the reaction nears completion (~20 min). To quantify the reaction rate, the absorbance at 260 nm was monitored at 0.10 s increments. These data are shown in the inset of Figure 3.7. The plot exhibits the expected linear dependence for a pseudo first-order reaction. The analysis of this data gives a value of 𝑘𝑘ℎ,𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 for DSP of 8.6 ± 0.5 × 102 M−1s−1 (n = 18). For comparison, the value for 𝑘𝑘ℎ,𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 for NBS under these same conditions was 5.7 ± 0.2 × 102 M−1s−1 (n = 9); this reaction rate constant is larger than that reported earlier due to the inclusion of TMAC in that work, which is known to slow such reactions.56 Previous studies have shown that the rates for the heterogeneous base hydrolysis of DSP-based monolayers and several other NHS esters are markedly retarded (100 to 1000x) in comparison to those found for the analogous homogeneous reactions.50,57-58 In our case, however, the heterogeneous rate constant, 𝑘𝑘ℎ , for the DSP-based adlayer is nearly twice that of the value found in bulk solution. This finding indicates that the interfacial effects (i.e., sterics, polarity, etc.) often considered as slowing interfacial reaction rates do not play a significant role in the data reported in Figure 3.4. 86 Figure 3.7. UV-Vis absorption spectra for hydrolysis of 0.10 mM DSP in 50 mM borate buffer (pH 8.50) and 1% 1,4-dioxane. The spectrum of the blank (dotted line) is that for the reaction mixture without DSP present, which has been self-normalized. The first spectrum of the reaction mixture was obtained after mixing the reaction solution and collecting a spectrum, a time span of ~25 s. All subsequent spectra are displayed at intervals of ~100 s from the start of reaction. The inset is a pseudo first-order kinetic plot for the hydrolysis reaction based on solution absorbance at 260 nm. Some of the error bars are close to the size of the data point. 87 3.3.5 Adlayer Aminolysis For aminolysis reactivity studies, this investigation used ethylamine as a small molecule mimic of the amines in the lysine residues of immunoglobulin (IgG) proteins. Proteins like IgG can contain between 80 to 90 lysines per molecule,59-60 and those located at the periphery of the protein structure have acid strengths (pKa ~10.3)61 similar to ethylamine (pKa ~10.8).62 The IR-ERS spectra for the temporal evolution of the reaction of the adlayer with 500 mM ethylamine (50 mM borate buffer, pH 8.50) are consistent with amide formation. Initially, there is a rapid decrease of the νa(C=O) for NHS at 1748 cm−1, which at longer times is accompanied by the appearance of much weaker bands at 1665, 1556, and 1265 cm−1 (Figure 3.8 and Table 3.3) that can be assigned to amide I, amide II and amide III vibrational modes, respectively.63-65 Interestingly, the strength of νa(C=O) decreases by ~50% in 120 ± 30 s (t50%) (n = 3), whereas the same decrease for immersion in borate buffer only (Figure 3.4) required nearly 400 s. The more rapid loss of the NHS group reflects contributions from both the base hydrolysis and aminolysis reactions. The kinetic plot for the data in Figure 3.8 is shown in Figure 3.9. Due to the weakness of amide bands, this analysis tracked the decrease in the NHS carbonyl mode at 1748 cm−1. To obtain the heterogeneous aminolysis reaction rate constant, the combined second-order rate laws for the two reactions occurring in parallel can be written as: 𝑑𝑑 Γ𝑁𝑁𝑁𝑁𝑁𝑁 = − 𝑘𝑘ℎ Γ𝑁𝑁𝑁𝑁𝑁𝑁 [𝑂𝑂𝐻𝐻 − ]− 𝑘𝑘𝑎𝑎 Γ𝑁𝑁𝑁𝑁𝑁𝑁 [𝑁𝑁𝐻𝐻2 ] 𝑑𝑑𝑑𝑑 (3.7) where 𝑘𝑘𝑎𝑎 is the second-order heterogeneous rate constant for aminolysis (M−1s−1) and [𝑁𝑁𝐻𝐻2 ] is the deprotonated amine concentration in bulk solution (M). The corresponding integrated rate law is given by Equation 3.8: Γ𝑁𝑁𝑁𝑁𝑁𝑁(𝑡𝑡) = Γ𝑁𝑁𝑁𝑁𝑁𝑁(𝑡𝑡=0) 𝑒𝑒 −(𝑘𝑘ℎ[𝑂𝑂𝑂𝑂 − ]+𝑘𝑘 𝑎𝑎 [𝑁𝑁𝑁𝑁2 ])𝑡𝑡 (3.8) 88 Figure 3.8. Infrared spectra of the DSP-based monolayer after immersion in 500 mM ethylamine in 50 mM borate buffer (pH 8.50) for 24 hours (top) and various time steps (bottom). 89 Table 3.3. Infrared spectral peak positions and band assignments for aminolysis reaction products of the DSP-based monolayer.63-65 Mode Assignment νs(C=O) 80% ν(C=O) 60% δ(N-H), 40% ν(C-N) δ(CH2) 40% ν(C-N), 30% δ(N-H), 20% ν(CH3-C) νas(C-C,C-N) Mode Description free carboxylic acid amide I amide II methylene scissors deformation amide III CN, CC of NHCH2CH3 Peak Position (cm−1) 1742 1665 1556 1456 1265 1107 90 Figure 3.9. Kinetic plot for aminolysis of the DSP-derived adlayer in 500 mM ethylamine in 50 mM borate buffer (pH 8.50). 91 For comparative purposes, we analyzed the plot in Figure 3.9 at the time interval from 60 to 300 s, with the assumption that the rate in this time interval is representative of the bulk transformation stage for the two competing reactions per Avrami analysis. Thus, the time interval of 60 to 300 s yields a 𝑘𝑘𝑎𝑎 of 9.4 ± 2.8 × 10−1 M−1s−1. In comparison, analyses using different time intervals in Figure 3.9 (0 to 600 s, 0 to 300 s, or 120 to 300 s) yielded 𝑘𝑘𝑎𝑎 values of 0.09, 6.9, and 7.7 × 10−1 M−1 s−1, respectively. These values differ at most by an order of magnitude in comparison to that for the 60 to 300 s interval. Hence, the analysis in the time interval of 60 to 300 s, which yields the largest value for 𝑘𝑘𝑎𝑎 (i.e., the best case scenario for the effectiveness of the aminolysis reaction), can be used for a generalized comparison with the rate to hydrolysis. 3.3.6 Implications of the Kinetic Measurements on the Aminolytic Immobilization of Protein The two reaction rate experiments show that the base hydrolysis reaction ( 𝑘𝑘ℎ = 1.5 ± 0.1 × 103 M−1 s−1) is inherently much faster than the aminolysis reaction ( 𝑘𝑘ℎ = 9.4 ± 2.8 × 10−1 M−1 s−1) at the DSP-based adlayer. To qualitatively estimate the impact of this difference, the relative rates of the two processes in the bulk transformation stage can be compared using the ratio, 𝑟𝑟𝑎𝑎/ℎ , expressed as: 𝑟𝑟𝑎𝑎/ℎ = 𝜈𝜈𝑎𝑎 𝑘𝑘𝑎𝑎 Γ𝑁𝑁𝑁𝑁𝑁𝑁 [𝑁𝑁𝑁𝑁2 ] 9.4 x 10−1 [𝑁𝑁𝑁𝑁2 ] [𝑁𝑁𝑁𝑁2 ] = = = 6.3 x 10−4 − 3 − 𝜈𝜈 ℎ 𝑘𝑘ℎ Γ𝑁𝑁𝑁𝑁𝑁𝑁 [𝑂𝑂𝑂𝑂 ] 1.5 x 10 [𝑂𝑂𝑂𝑂 ] [𝑂𝑂𝑂𝑂 − ] (3.9) where 𝜈𝜈𝑎𝑎 is the reaction rate of aminolysis and 𝜈𝜈 ℎ is the reaction rate of hydrolysis. Equation 3.9 clearly points to the dominance of the base hydrolysis reaction in borate buffer (pH 8.50, 50 mM). Per Step 2 in Scheme 3.1, we typically use 100 μg/mL of antibody for 92 immobilization,66-68 which for IgG proteins (150 kDa) translates to ~0.7 μM. Using this value for the amine concentration in Equation 3.9, 𝑟𝑟𝑎𝑎/ℎ then equals 1.4 × 10−4. In this case, the removal of the NHS terminal groups is completed (i.e., 99.9% conversion) in ~210 s; however, less than 0.02% of the conversion is due to aminolysis. We can extend this projection by recognizing that the number of lysines sterically accessible at the periphery of an IgG protein ranges from 10 to 26.59-60 If, neglecting the role of pH and the concomitant acceleration of the rate of base hydrolysis, we increase the apparent amine concentration by 26, 𝑟𝑟𝑎𝑎/ℎ increases to 3.6 × 10−3, and the conversion of Γ𝑁𝑁𝑁𝑁𝑁𝑁 due to aminolysis, while increasing to ~0.4%, remains insignificant. Moreover, we have yet to take into account the deprotonation state of the accessible amine groups. At a solution pH of 8.50, less than 2% of the sterically accessible amines are deprotonated, which works against improvements in the effectiveness of aminolysis (i.e., 𝑟𝑟𝑎𝑎/ℎ decreases to 5.6 × 10−5 and the conversion due to aminolysis drops below 0.01%). In projecting the implications of Equation 3.9 further, one could ask "what reactive amine concentration would be needed in order to make an argument in favor of protein tethering via aminolysis?" If, for example, the goal would be to achieve a value of 𝑟𝑟𝑎𝑎/ℎ of unity, the reactive amine (deprotonated) concentration would need to approach 5 mM. This concentration translates to an IgG protein level, assuming that 2% of the 26 sterically accessible amides are deprotonated at pH 8.50, of ~50 g/mL, which is, of course, not feasible. In other words, these kinetic data lead to the conclusion that the mechanism for protein coupling under the conditions in Scheme 3.1 has a minimal contribution (if any) to the formation of a layer of capture antibodies and that another pathway, i.e., adsorption, dominates the preparation process. 93 3.4 Conclusions This work has endeavored to shed light on the use of NHS-ester monolayers in the immobilization of proteins, such as capture antibodies, in biosensor systems through IRERS reactivity studies of the desired aminolysis and competing hydrolysis reactions under common immunoassay conditions (borate buffer, pH 8.50). In contrast to the expected pseudo first-order linear reaction rate, the competing reaction of hydrolysis at the DSPbased monolayer surface gave evidence for a reaction in series with multiple reaction rates. An extension of JMAK theory was applied to describe these multiple kinetic regimes: an initiation stage with formation of reacting nuclei; a bulk transformation stage with rapid growth of the nuclei; and a final stage with slow growth. Applying an interpretation based on JMAK theory to the kinetic plots resulted in a tremendous difference in reaction rates for the competing hydrolysis and the desired aminolysis reactions. Moreover, these large differences in reaction rate constants clearly show that hydrolysis is the dominant reaction under conditions in which coupling agents are immersed in low protein concentrations and buffers of near physiological pH. Thus, this kinetic data points to the case in which the vast majority of the proteins immobilized per Scheme 3.1 are present due to electrostatic, hydrogen bonding, and van der Waals interactions, rather than by covalent linkages. There are a large number of diagnostic test platforms used in today's healthcare system that employ NHS-based chemistry for protein immobilization. The data herein suggests the possibility that adsorption may play a more important role in such processes than originally thought and that a reexamination of the immobilization chemistry may, in some cases, improve metrics of performance (e.g., reproducibility, limits of detection, etc.) To this end, future work will investigate the competing reactions of hydrolysis and aminolysis in NHS- 94 based chemistries under various reaction conditions with the anticipation that further understanding of these competing reactions will help elucidate a means to achieve covalent coupling of proteins. These experiments will be reported elsewhere. 3.5 References (1) Jonkheijm, P.; Weinrich, D.; Schroder, H.; Niemeyer, C. M.; Waldmann, H. Angew. Chem., Int. Ed. 2008, 47, 9618-9647. (2) Rusmini, F.; Zhong, Z.; Feijen, J. Biomacromolecules 2007, 8, 1775-1789. (3) Hermansen, G. T. Bioconjugate Techniques. 2nd ed.; Academic Press, Inc.: New York, 2008. (4) Patel, N.; Davies, M. C.; Hartshorne, M.; Heaton, R. J.; Roberts, C. J.; Tendler, S. J. B.; Williams, P. M. Langmuir 1997, 13, 6485-6490. (5) Wagner, P.; Hegner, M.; Kernen, P.; Zaugg, F.; Semenza, G. Biophys. J. 1996, 70, 2052-2066. (6) Porter, M. D.; Lipert, R. J.; Siperko, L. M.; Wang, G.; Narayanan, R. Chem. Soc. Rev. 2008, 37, 1001-1011. (7) Dufek, E. J.; Ehlert, B.; Granger, M. C.; Sandrock, T. M.; Legge, S. L.; Herrmann, M. G.; Mieikle, A. W.; Porter, M. D. Analyst 2010, 135, 2811-2817. (8) Lomant, A. J.; Fairbanks, G. J. Mol. Biol. 1976, 104, 243-261. (9) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481-4483. (10) Xu, H.; Lu, J. R.; Williams, D. E. J. Phys. Chem. B 2006, 110. (11) Mädler, S.; Bich, C.; Touboul, D.; Zenobi, R. J. Mass Spectrom. 2009, 44, 694706. (12) Arakawa, T.; Timasheff, S. N. Biochemistry 1982, 21, 6545-6552. (13) Chi, E. Y.; Krishnan, S.; Randolph, T. W.; Carpenter, J. F. Pharm. Res. 2003, 20, 1325-1336. (14) Anjaneyulu, P. S. R.; Staros, J. V. Int. J. Pept. Protein Res. 1987, 30, 117-124. 95 (15) Grubor, N. M.; Shinar, R.; Jankowiak, R.; Porter, M. D.; Small, G. J. Biosens. Bioelectron. 2004, 19, 547-556. (16) Wang, Q.; Kromka, A.; Houdkova, J.; Babchenko, O.; Rezek, B.; Li, M.; Boukherroub, R.; Szunerits, S. Langmuir 2012, 28, 587-592. (17) Darder, M.; Takada, K.; Pariente, F.; Lorenzo, E.; Abruna, H. D. Anal. Chem. 1999, 71, 5530-5537. (18) Cabrita, J. F.; Abrantes, L. M.; Viana, A. S. Electrochim. Acta 2005, 50, 21172124. (19) Veiseh, M.; Zareie, M. H.; Zhang, M. Langmuir 2002, 18, 6671-6678. (20) Jiang, L.; Glidle, A.; Griffith, A.; McNeil, C. J.; Cooper, J. M. Bioelectrochem. Bioenerg. 1997, 42, 15-23. (21) Grdadolnik, J.; Marechal, Y. Biopolymers 2000, 62, 40-53. (22) Frey, B. L.; Corn, R. M. Anal. Chem. 1996, 68, 3187-3193. (23) Dordi, B.; Pickering, J. P.; Schonherr, H.; Vancso, G. J. Surf. Sci. 2004, 570, 5766. (24) Norde, W.; MacRitchie, F.; Nowicka, G.; Lyklema, J. J. Colloid Interface Sci. 1986, 112, 447-156. (25) Seah, M. P.; Smith, G. C.; Anthony, M. T. Surf. Interface Anal. 1990, 15, 293-308. (26) Shirley, D. A. Phys. Rev. B 1972, 5, 4709-4714. (27) Moulder, J. F. Handbook of X-ray Photoelectron Spectroscopy. Physical Electronics, Inc.: Eden Prairie, MN, USA, 1992. (28) Walczak, M. M.; Popenoe, D. D.; Deinhammer, R. S.; Lamp, B. D.; Chung, C.; Porter, M. D. Langmuir 1991, 7, 2687-2693. (29) Angerstein-Kozlowska, H. A.; Conway, B. E.; Hamelin, A.; Stoicoviu, L. J. Electroanal. Chem. Interfacial Electrochem. 1987, 228, 429-453. (30) Petrovykh, D. Y.; Kimura-Suda, H.; Whitman, L. J.; Tarlov, M. J. J. Am. Chem. Soc. 2003, 125, 5219-5226. (31) Petrovykh, D. Y.; Kimura-Suda, H.; Tarlov, M. J.; Whitman, L. J. Langmuir 2004, 20, 429-440. 96 (32) Powell, C. J.; Jablonski, A. NIST Electron Effective-Attenuation-Length Database, Version 1.0 (SRD-82), U. S. Department of Commerce, National Institute of Standards and Technology: Gaithersburg, MD., 2001. (33) McKittrick, P. T.; Katon, J. E. Appl. Spectrosc. 1990, 44, 812-817. (34) Delamarche, E.; Sundarababu, G.; Biebuyck, H.; Michel, B.; Gerber, C.; Sigrist, H.; Wolf, H.; Ringsdorf, H.; Xanthopoulos, N.; Mathieu, H. J. Langmuir 1996, 12, 1997-2006. (35) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321-335. (36) Bain, C. D.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1989, 5, 723-727. (37) Cho, Y.; Ivanisevic, A. J. Phys. Chem. B 2005, 109, 6225-6232. (38) Laredo, T.; Leitch, J.; Chen, M.; Burgess, I. J.; Dutcher, J. R.; Lipkowski, J. Langmuir 2007, 23, 6205-6211. (39) Schneider, T. W.; Buttry, D. A. J. Am. Chem. Soc. 1993, 115, 12391-12397. (40) Yang, D. F.; Wilde, C. P.; Morin, M. Langmuir 1997, 13, 243-249. (41) Zhong, C. J.; Zak, J.; Porter, M. D. J. Electroanal. Chem. 1997, 421, 9-13. (42) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559-3568. (43) Nuzzo, R. G.; Fusco, F. A.; Allara, D. L. J. Am. Chem. Soc. 1987, 109, 2358-2368. (44) Porter, M. D. Anal. Chem. 1988, 60, 1143A-1155A. (45) We view this assumption as reasonable since there is no detectable change in the absolute strengths of C-H the stretching modes between 2800 and 3000 cm-1 (data not shown). (46) Avrami, M. J. Chem. Physics 1940, 8, 212-224. (47) Avrami, M. J. Chem. Physics 1941, 9, 177-184. (48) Vinokurov, I. A.; Morin, M.; Kankare, J. J. Phys. Chem. B 2000, 104, 5790-5796. (49) Fatemi, N.; Whitehead, R.; Price, D.; Dollimore, D. Thermochim. Acta 1986, 104, 93-100. 97 (50) Dordi, B.; Schonherr, H.; Vancso, G. J. Langmuir 2003, 19, 5780-5786. (51) Ulman, A. Chem. Rev. 1996, 96, 1533-1554. (52) Schönherr, H.; Chechik, V.; Stirling, C. J.; Vancso, G. J. J. Am. Chem. Soc. 2000, 122, 3679-3687. (53) A rigorous analysis of the reaction rate for the base hydrolysis of esters includes an accounting of the reaction with H2O. We did not detect any conversion for the homogeneous and heterogeneous reactions of DSP in water after 24 h. (54) Cline, G. W.; Hanna, S. B. J. Org. Chem. 1988, 53, 3583-3586. (55) Miron, T.; Wilchek, M. Anal. Biochem. 1982, 126, 433-435. (56) Baltzly, R.; Berger, I. M.; Rothstein, A. A. J. Am. Chem. Soc. 1950, 72, 4149-4152. (57) Chechik, V.; Stirling, C. J. M. Langmuir 1997, 13, 6354-6356. (58) Chechik, V.; Crooks, R. M.; Stirling, C. J. M. Adv. Mater. 2000, 12, 1161-1171. (59) Southwick, P. L.; Ernst, L. A.; Tauriello, E. W.; Parker, S. R.; Mujumdar, R. B.; Mujumdar, S. R.; Clever, H. A.; Waggoner, A. S. Cytometry 1990, 11, 418-430. (60) Nakagawa, Y.; Capetillo, S.; Jirgensons, B. J. Biol. Chem. 1972, 247, 5703-5708. (61) Zhang, M.; Vogel, H. J. J. Biol. Chem. 1993, 268, 22420-22428. (62) Lyde, D. R. CRC Handbook of Chemistry and Physics 2004-2005: A ReadyReference Book of Chemical and Physical Data. CRC Press: 2004. (63) Miyazawa, T.; Shimanouchi, T.; Mizushima, S. I. J. Chem. Phys. 1956, 24, 408418. (64) Miyazawa, T.; Shimanouchi, T.; Mizushima, S. I. J. Chem. Phys. 1958, 29, 611616. (65) Zeroka, D.; Jensen, J. O.; Samuels, A. C. J. Mol. Struct. 1999, 465, 119-139. (66) Driskell, J. D.; Kwarta, K. M.; Lipert, R. J.; Porter, M. D.; Neill, J. D.; Ridpath, J. F. Anal. Chem. 2005, 77, 6147-6154. (67) Grubisha, D. S.; Lipert, R. J.; Park, H. Y.; Driskell, J.; Porter, M. D. Anal. Chem. 2003, 74, 5936-5945. 98 (68) Yakes, B. J.; Lipert, R. J.; Bannantine, J. P.; Porter, M. D. Clin. Vaccine Immunol. 2008, 15, 227-234. CHAPTER 4 HYDROLYSIS AND AMINOLYSIS OF SUCCINIMIDYL ESTER MONOLAYERS: IMPACT OF METHYLENE CHAIN-LENGTH AND SOLUTION COMPOSITION ON EFFICIENT PROTEIN IMMOBILIZATION 4.1 Introduction Coatings composed of N-hydroxysuccinimide (NHS) esters are frequently used to covalently immobilize antibodies and other biological materials on surfaces through the formation of amide linkages.1-6 However, given the typical conditions used for the coupling reaction (e.g., pH 6.5 to 8.5), competition from hydrolysis can limit aminolysis efficiency.1,7 Along these lines, we recently reported that the rate of the hydrolysis of the NHS terminus of a monolayer prepared on a gold surface from the chemisorption of dithiobis (succinimidyl propionate) (DSP) outpaces aminolysis under the conditions often used for the linking reaction (e.g., a reaction solution of immunoglobulin G (IgG) at 100 µg/mL buffered at pH 8.5).8 We therefore concluded that it is much more likely that the antibodies are simply adsorbed on the surface by hydrophobic or electrostatic interactions.1,9 This paper builds on that study by investigating conditions (i.e., changes to the methylene chain length (n) of the NHS adsorbate, solution composition, and pH) that could potentially improve the efficiency of the aminolytic coupling of proteins via NHS 100 esters.8 Monolayer structure characteristics such as packing density can significantly affect interfacial reaction rates due to differences in accessibility to the immobilized reactant.10 Several studies have investigated the effect of n on reactivity using NHS ester-terminated monolayers, but they differ in their conclusions.11-12 For example, Patel et al. report that shorter chains increase aminolytic coupling due to increased amine accessibility from the disordered monolayer while Wagner et al. state that monolayers derived from longer chains (n = 10) improve coupling efficiency due to steric protection of the more densely packed bulky end group, which hinders hydrolysis.11,13 The formation of the adlayers in these two studies differs, however, with the former involving an additional activation step using NHS and N-ethyl-N-(3-dimethylaminopropyl) carbodiimide (EDAC) to convert the carboxylic acid groups of the as-formed adlayer to an NHS ester terminus. Nonetheless, these reports agree that steric accessibility plays a role in interfacial reaction rates, but to what extent n influences aminolysis and/or hydrolysis of NHS monolayers, and thus protein coupling, remains in question. The structural changes that affect adlayer reactivity (i.e., monolayer structure, hydrophobicity, lateral interactions between chemisorbed molecules) can be further investigated through an examination of the thermodynamic activation parameters: entropy of activation (Δ𝑆𝑆 ‡ ), enthalpy of activation (Δ𝐻𝐻 ‡ ), free energy of activation (Δ𝐺𝐺 ‡ ), activation energy (𝐸𝐸𝐴𝐴 ), and the Arrhenius prefactor (𝐴𝐴) of the transition state.14-15 For adlayers specifically, an increase in Δ𝑆𝑆 ‡ indicates a more favorable transition state, an increase in 𝐸𝐸𝐴𝐴 indicates an increase in rotational mobility, and an increase in Δ𝐻𝐻 ‡ indicates an increase in lateral methylene interactions between neighboring molecules.10,15 101 Changes to the solution (i.e., buffer composition, concentration, and pH) may also affect the rate of hydrolysis.12,16 In general, protocols suggest the use of neutral to slightly basic buffer solutions (pH 7 to 8.5) that maintain physiological conditions (e.g., ionic strength to preserve protein stability)17-18 and facilitate deprotonation of the amine groups, which aids in the amide linkage formation.1 This, of course, comes with the caveat that the competing hydrolysis reaction is also base catalyzed (i.e., increases with increasing pH). To combat this, procedures often recommend the use of high amine concentrations to tip the reaction to favor aminolysis over base-catalyzed hydrolysis. However, high protein concentrations can be cost prohibitive and limited by solubility.18 Thus, to effectively immobilize proteins, one should carefully consider solution conditions that minimize hydrolysis and maximize aminolysis while maintaining a practical protein concentration, ionic strength, and pH. This report investigates the effects of methylene chain length (n) and solution composition on the efficiency of NHS ester-based protein immobilization. This investigation includes the characterization of four dithiobis succinimidyl monolayers on gold with varying n (2 < n < 10) by external reflection infrared spectroscopy (IR-ERS), Xray photoelectron spectroscopy (XPS), electrochemical reductive desorption, and contact angle measurements. Hydrolysis and aminolysis reactivity of the four NHS-terminated monolayers in sodium hydroxide and borate buffer was determined by IR-ERS with and without the amine source, ethylamine. Additionally, thermodynamic activation parameters (Δ𝑆𝑆 ‡ , Δ𝐻𝐻 ‡ , Δ𝐺𝐺 ‡ , 𝐸𝐸𝐴𝐴 , and 𝐴𝐴) were determined from the hydrolysis experiments at varying temperatures to reveal the origins of differences in monolayer reactivity. The aminolysis and hydrolysis kinetic results are used to assess the efficiency of protein immobilization 102 with these different systems (i.e., varying n and solution composition) and provide insight into assay conditions that will promote covalent protein immobilization using NHS chemistry. 4.2 Experimental 4.2.1 Reagents and Materials Dithiobis (succinimidyl propionate) (DSP), ethylamine hydrochloride (EA), and sulfuric acid (ACS grade) were purchased from Sigma Aldrich; sodium nitrate, from EMD Millipore; gold shot (99.999%), from Alfa Aesar; borate buffer (BB, 50 mM, pH 8.50) packs, from Thermo Scientific; hydrogen peroxide (ACS, 30%) and sodium hydroxide, from Fisher Scientific; dithiobis (succinimidyl hexanoate) (DSH), dithiobis (succinimidyl octanoate) (DSO), and dithiobis (succinimidyl undecanoate) (DSU), from Dojindo Molecular Technologies; 200 proof ethanol (EtOH), from Pharmco-AAPER; and octadecanethiolate-d37 from CDN isotopes. The EtOH was dried over a fresh charge of molecular sieves for two 24 hour periods before use. All other chemicals were used as received. Aqueous solutions were prepared with water purified using a Barnstead water polishing system that ensured a resistivity of 18.2 MΩ. 4.2.2 Adlayer Preparation Monolayers were formed on gold-coated glass slides (25 mm × 75 mm). Glass slides were cleaned via piranha etch (3:1 H2SO4:H2O2 (30%)) for 10 min followed by thorough rinsing with deionized water. Caution: Piranha etch reacts violently with most organic materials and must be used and handled with extreme care. The slides were then 103 rinsed with methanol, dried under a stream of nitrogen, and coated with ~15 nm of a chromium adhesion layer and ~200 nm of a gold film via vapor deposition. Monolayers were prepared by immersing Cr/Au-coated glass slides in a 0.10 mM ethanolic solution of DSP, DSH, DSO, or DSU. After 16 h, the slides were rinsed with ethanol and dried under a stream of nitrogen. 4.2.3 Infrared Spectroscopy (IR) Instrumentation External reflection infrared spectra (IR-ERS) were collected with a Nicolet Magna 850 Fourier transform infrared spectrometer with a mercury cadmium telluride detector and p-polarized light at an incidence angle of 80°. Spectra were acquired by co-addition of 512 scans at 2 cm−1 resolution and reported as -log R/R0, where R represents the reflection spectrum from the sample monolayer and R0 the reflection spectrum from an octadecanethiolate-d37 reference monolayer. Further characterization of the four adlayers by X-ray photoelectron spectroscopy, electrochemistry, and contact angle measurements is described in detail in the Appendix. 4.2.4 Interfacial Kinetic Studies These studies were performed by immersion of adlayers on gold in ~60 mL of the running solution at varied time segments. Temperature control was achieved through a water-filled heating bath (BÜCHI Labortechnik AG B-491, ± 2°C) in which a glass housing, containing ~60 mL of the running solution, was allowed to equilibrate (~1 h) before starting the adlayer reaction. After a specific time segment, the adlayer was rinsed thoroughly with DI water and gently dried with nitrogen.19 IR-ERS spectra were collected 104 immediately. For hydrolysis, the running solutions were 50 mM BB (pH 8.50), 10 mM PBS (pH 7.4), 1 × 10−3 M NaOH (aq, pH ~11), or 2 × 10−7 M NaOH (aq, pH ~8.5). Sodium nitrate was added to fix the ionic strength, μ, from 0.001 to 0.4 for the NaOH solutions. Aminolysis studies were performed using the same procedure with the addition of 500 mM ethylamine in the running solutions. Reaction rate constants are calculated from 3 replicate adlayers. 4.3 Results and Discussion We recently reported that, under typical conditions used for protein coupling in immunoassays (e.g., 100 µg/mL IgG in 50 mM borate buffer, pH 8.5), the NHS terminus of a monolayer formed by the spontaneous adsorption of DSP hydrolyzes at a rate well beyond that of aminolysis.8 In building on that work, this paper investigates the effects of methylene chain length (n) and solution composition on the rates of these two competing chemistries. We first characterized the structure and wettability of four different NHSterminated monolayers on gold (DSP (n = 2); DSH (n = 5); DSO (n = 7); and DSU (n = 10)) by IR-ERS, XPS, electrochemistry, and contact angle measurements. We then carried out kinetic studies by monitoring the loss of the succinimidyl leaving group upon hydrolysis or aminolysis of the different monolayers by IR-ERS. These experiments examined the: (1) hydrolysis of each monolayer in borate buffer (50 mM, pH 8.50), (2) hydrolysis of DSP- and DSU-based monolayers in borate buffer at three different temperatures to determine thermodynamic activation parameters, (3) hydrolysis of the DSP-based monolayers in phosphate buffer and sodium hydroxide, and (4) aminolysis of DSP- and DSU-based monolayers in sodium hydroxide. Finally, these hydrolysis and 105 aminolysis data were applied to protein immobilization as part of general guidelines for the use of NHS chemistry to covalently link proteins to a solid surface. 4.3.1 Characterization of NHS-terminated Monolayers by IR-ERS IR-ERS spectra of the as-formed monolayers from DSP, DSH, DSO, and DSU on gold are shown in Figure 4.1A from 2000 to 1000 cm−1 and in Figure 4.1B from 3000 to 2800 cm−1. Band assignments are given in Table 4.1. As expected, the spectral features for the 4 monolayers are similar and match those of the expected composition. In the low energy region, the band assignments are: ν(C=O) at ~1816 cm−1, νs(C=O) at ~1786 cm−1, νa(C=O) at ~1748 cm−1, νs(C-N-C) at ~1380 cm−1, νa(C-N-C) at ~1218 cm−1, and ν(C- O) at ~1075 cm−1.16 These spectra can also be used to qualitatively examine the packing density of the different adlayers. This analysis assumes that the NHS terminus for the 4 adlayers is conserved.20 The small differences (< 4%) in the strength of νa(C=O) for the 3 longer chain monolayers therefore points to comparable packing densities, whereas the decrease in the relative strength of this band by more than 30% for the shortest chain monolayer indicates a lower packing density. The reduction in packing density for DSP likely also represents a loss in order for the DSP-based monolayer compared to the other 3 adlayers. This packing difference is supported by an analysis of the spectra in Figure 4.1B. The two strongest bands in these spectra are from the asymmetric, νa(CH2), and symmetric, νs(CH2), methylene stretches at ~2928 and ~2857 cm−1, respectively.21 These bands can also be used to qualitatively assess the density of chain packing by recognizing that the positions of the two modes are sensitive to the packing density of the 106 Figure 4.1. IR-ERS spectra of NHS ester based monolayers on gold. IR-ERS spectra of DSP(black), DSH- (red), DSO- (blue), and DSU-based (green) adlayers are shown in (A) the fingerprint region from 2000 cm−1 to 1000 cm−1, and (B) the C−H region from 3000 cm−1 to 2800 cm−1. 107 Table 4.1. IR spectral peak positions and band assignments for DSP-, DSH-, DSO-, and DSU-based adlayers on gold. mode assignment description ν(C═O) νs(C═O) νa(C═O) δ(C═O) νs(C−N−C) νa(C−N−C) ν(C−O) ν(CH2) νa(CH2) νs(CH2) carbonyl stretch of NHS ester symmetric carbonyl stretch of NHS asymmetric carbonyl stretch of NHS methylene scissors deformation symmetric C−N−C stretch of NHS asymmetric C−N−C stretch of NHS N−C−O stretch of NHS methylene stretch of NHS asymmetric methylene stretch symmetric methylene stretch peak position (cm−1) DSP DSH DSO DSU 1814 1815 1815 1816 1788 1787 1787 1786 1748 1748 1748 1749 1433 1433 1433 1433 1373 1380 1381 1381 1215 1217 1217 1217 1077 1074 1076 1078 2962 2961 2965 2964 2929 2928 2928 2925 2857 2857 2857 2856 108 alkyl chains.22-23 For hydrocarbons in a solid form, the νa(CH2) band is at 2918 cm−1 and the νs(CH2) band is at 2851 cm−1.23 These positions change to slightly higher energies as the packing density of the chains decreases, with those for liquid hydrocarbons, 2924 cm−1 for νa(CH2) and 2855 cm−1 for νs(CH2). The subtle, but measureable, shift in the νa(CH2) position is consistent with a change in packing density predicted from the analysis of the NHS strength changes. The ν(CH2) band positions indicate more loosely packed monolayers than those reported by Dordi et al. (i.e., νa(CH2) at 2922 and νs(CH2) at 2852 cm−1 for DSU). Each of the monolayers was also analyzed by XPS and contact angle measurements (data presented in the appendix). The XPS data, like the IR-ERS data, match those of the expected composition of the monolayer.24-27 However, surface concentration (ΓNHS) determinations by XPS (ΓNHS,XPS(DSP): 6.0 ± 0.8 × 10−10; ΓNHS,XPS(DSH): 5.3 ± 0.2 × 10−10; ΓNHS,XPS(DSO): 5.4 ± 0.3 × 10−10; ΓNHS,XPS(DSU): 5.3 ± 0.8 × 10−10 mol/cm2) gave similar ΓNHS for all four monolayers, indicating a similar coverage density.28-29 The advancing contact angle measurements were consistent with literature (θa(DSP):59 ± 1°; θa(DSH):60 ± 0°; θa(DSO):61 ± 1°; θa(DSU):63 ± 1°) whereas the receding contact angle measurements (θr(DSP):33 ± 0°; θr(DSH):50 ± 0°; θr(DSO):49 ± 1°; θr(DSU):55 ± 0°) increased with increases in n.30 These contact angles translate to hysteresis (H) values, the difference between θa and θr, that decrease with increases in n (H(DSP):25 ± 1°; H(DSH):11 ± 1°; H(DSO):12 ± 1°; H(DSU):8 ± 1°). This reduction in θr, and consequential increase in H for the DSP-based adlayer indicates an increase in disorder and hydrophilicity (i.e., wettability) over the other three adlayers.31-32 The decrease in θr between DSP and DSU is larger than that reported by Dordi et al. 109 (θr(DSP):39 ± 2° and θr(DSU):43 ± 2°) and indicates the DSP-based monolayer is more disordered and the DSU-based monolayer is less disordered than those of Dordi et al. With literature indicating that loosely packed monolayers exhibit more rotational freedom of the end group with increased accessibility to reactants,33-34 we posit the more loosely-packed DSP-based monolayer will translate to increased accessibility of the NHS end group, which will likely increase the rate of reaction in comparison to the DSU-based monolayer. 4.3.2 Reaction Products for the Hydrolysis and Aminolysis of NHSterminated Monolayers IR-ERS spectra for 4 different adlayers before and after 1 h of immersion in either 50 mM BB, pH 8.50 for hydrolysis or 500 mM EA, pH 8.50 for aminolysis are shown in Figure 4.2 with band assignments in the appendix. The aminolysis studies used EA as a small molecule mimic of the lysine residues that lie on the outer periphery of proteins. We note that the acid strength of EA (pKa: 10.3) is close to that of a lysine residue (pKa: 10.8).1,35-36 These conditions approximate a "best case scenario" for protein immobilization. Upon hydrolysis and aminolysis, the spectral features originating from the NHS ester (e.g., ν(C═O), νs(C═O), and νa(C═O)) decrease in strength, with the temporal decrease, for example, of νa(C═O) being dependent on n. That is, the hydrolysis of the DSP-based monolayer appears to go to completion within 1 h, whereas the strength of νa(C═O) decreases by ~40%, ~60%, and ~70% for a DSU-, DSO-, and DSH-based adlayers, respectively. The increased extent of reaction with the shorter chain systems indicates the more liquid-like packing of the DSP system does translate to increased 110 Figure 4.2. IR spectra of DSP- (black), DSH- (red) , DSO- (blue) , and DSU-based (green) monolayers after hydrolysis in 50 mM BB (pH 8.50) and aminolysis in 500 mM EA (NaOH, pH 8.50) for 1 h. 111 reactant accessibility, as expected. However, the IR-ERS spectra with aminolysis show a minimal difference in reactivity as a function of n for all monolayers. With hydrolysis, the appearance of bands due to the free carboxylic acid (νs(C═O) at ~1742 cm−1) and carboxylate ion (ν(COO−) at ~1607, ~1550, and ~1264 cm−1) indicate a hydrolyzed monolayer with carboxylic acid terminal groups.8,16 With aminolysis, bands indicative of the amide linkage such as amide I and amide II at 1659 and 1553 cm−1, respectively, arise indicating covalent linkage of ethylamine with the succinimidyl ester.3739 IR-ERS spectra of the C−H region, from 3000 cm−1 to 2800 cm−1, before and after hydrolysis and aminolysis are shown in the appendix. These spectra are commensurate with expected reaction products (e.g., additional CH3 group after reaction with EA - ν(CH3) at 2965 cm−1) while indicating a slight increase in disorder of all four monolayers after reaction (i.e., increase in strength of CH2 bands). Overall, the IR-ERS spectrum of the reacted monolayers indicate the expected products and demonstrate a change in hydrolysis between the different chain lengths by differences in the intensity of νs(C═O) after 1 h immersion. 4.3.3 Hydrolysis Reaction Kinetics of DSP-based (short, n = 2) and DSU-based (long, n = 10) Monolayers by IR-ERS To examine more closely the differences in the hydrolysis of these systems, reaction rate studies for the shortest and longest chain length systems were carried out by means of IR-ERS at different immersion times. An example of these data is given in the appendix which was analyzed as detailed elsewhere, by following the temporal dependence of the NHS band, νa(C═O) at 1748 cm−1.8 Briefly, the hydrolysis reaction can be described by the 112 second-order rate law in Equation 4.1: 𝑑𝑑ΓNHS = −𝑘𝑘ℎ ΓNHS [OH − ] 𝑑𝑑𝑑𝑑 (4.1) where 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁 represents the surface concentration of NHS groups, 𝑘𝑘ℎ is the second-order heterogeneous rate constant for hydrolysis, and [𝑂𝑂𝑂𝑂 − ] is the hydroxide ion concentration in bulk solution. This rate law can be simplified to a pseudo first-order rate law owing to the use of a buffered alkaline solution: 𝑑𝑑ΓNHS = −𝑘𝑘ℎ′ ΓNHS 𝑑𝑑𝑑𝑑 (4.2) where 𝑘𝑘ℎ′ represents the first-order heterogeneous rate constant for hydrolysis. Integration of Equation 4.2 with the appropriate limits gives: ′ ΓNHS(𝑡𝑡) = ΓNHS(𝑡𝑡=𝑂𝑂) 𝑒𝑒 −𝑘𝑘ℎ 𝑡𝑡 (4.3) By assuming the 𝛤𝛤𝑁𝑁𝑁𝑁𝑁𝑁 is directly proportional to the strength of νa(C═O), the extent of reaction can be determined as a function of immersion time, which can then be used to calculate 𝑘𝑘ℎ′ from a slope of plot of -ln(1−x) versus time, where x represents the extent of reaction (i.e., ΓNHS(𝑡𝑡) /ΓNHS(𝑡𝑡 = 𝑂𝑂) ). This analysis also assumes that the orientation of the transition dipole moment of νa(C═O) remains unchanged throughout the time course of the reaction, aka the so-called infrared metal surface selection rule.20 The results of the kinetic analysis for the 2 different monolayers are summarized in Figure 4.3. The shape of the two kinetic plots in Figure 4.3A are similar to that reported by us recently for the DSP-derived system (i.e., a slow initial rate at short times (t < 240 s)), a more rapid rate as time increases (300 < t < 660 s), and, again, a slower rate as the reaction approaches completion (t > 660 s).8 The shape is ascribed to the differences in the rates of three reactions connected in series: a slow "initiation stage" in which reaction 113 Figure 4.3. Hydrolysis kinetics for DSP- and DSU-based monolayers. (A) Kinetic plot for hydrolysis of the DSP- (black) and DSU- (green) based monolayers in 50 mM borate buffer (pH 8.50), with the inset showing and enlarged version of the kinetic plot for the DSU-based monolayer. (B) Histogram of the second-order heterogeneous reaction rate constants, kh, for the three reaction stages: stage 1 (0 < t < 240 s), stage 2 (300 < t < 660 s), and stage 3 (t > 660 s). 114 nuclei are created, a more rapid "bulk transformation stage" in which the nuclei grow in size and begin to overlap, and a slower "final stage" in which the accessibility of the immobilized reactants decreases due to the loss of inter-chain interactions that support the adlayer. These stages will be referred to as: Stage 1 (i.e., initiation stage) with 𝑘𝑘ℎ1 , Stage 2 (i.e., bulk transformation stage) with 𝑘𝑘ℎ2 , and Stage 3 (i.e., final stage) with 𝑘𝑘ℎ3 . The time periods for each stage were determined from the transition points through the second and third derivatives of the kinetic plots shown in the appendix. In contrast, an initiation stage is reported by Dordi et al. only when n > 10, rationalized to stem from the increase in order with longer chain lengths. Though the IR and contact angle measurements indicate the DSP-based monolayer herein is less ordered than that of Dordi et al., we believe the shape of the kinetic plot arises from differences in the monolayer defects which cannot be probed by these surface techniques. It is clear from the 𝑘𝑘ℎ values shown in Figure 4.3B that the rate of hydrolysis depends on methylene chain length. The reaction rates differ by one order of magnitude with 𝑘𝑘ℎ2 of 1.5 ± 0.1 × 103 M−1s−1 for DSP-based monolayers and 2.2 ± 0.2 × 102 M−1s−1 for DSU-based monolayers. The increased 𝑘𝑘ℎ2 with DSP which exhibited more liquid-like packing behavior follows the trend of increased accessibility to reactants observed with liquid-like monolayers. In addition to the changes in 𝑘𝑘ℎ , there are several important differences in the transition points (i.e., the time and extent of reaction) between these two systems. The extent of reaction at the end of stage 1 and 𝑘𝑘ℎ1 decrease for the DSU-based monolayer in comparison to that derived from DSP (𝑘𝑘ℎ1 = 210 M−1s−1 for DSP and 40 M−1s−1 for DSU, x = ~0.15 for DSP ~0.05 for DSU). These differences could stem from 115 either (1) a change in the reaction of the defect sites or (2) a decrease in the density of reaction nuclei. Due to the constant time period for the initiation stage (0 < t < 240 s) between the two adlayers, we believe that the differences in 𝑘𝑘ℎ1 and x stem from the latter. The decrease in density of reaction nuclei may be due to the increased order of the DSUbased monolayers that limits reaction nuclei to specific defect sites. Similar to Stage 1, Stage 2 occurs over similar time scales for both adlayers ranging from 360 to 660 s and 300 to 600 s for the DSP-based and DSU-based monolayers, respectively. Again, 𝑘𝑘ℎ2 and the extent of reaction at the end of this stage decreases for the DSU-based monolayer (𝑘𝑘ℎ2 = 1500 M−1s−1 for DSP and 220 M−1s−1 for DSU, x = ~0.85 for DSP and ~0.30 for DSU). The relative increase of 𝑘𝑘ℎ2 over 𝑘𝑘ℎ1 , though, does not significantly change for the two monolayers, with an increase of ~7× and ~6× for the DSPand DSU-based adlayers, respectively. Again, we propose the decrease in x for the DSUbased monolayer indicates a decrease in the density of the reaction nuclei. With a decrease in the density of reaction nuclei for the DSU-based monolayer, the rapid growth to overlapping reactive sites may not occur in Stage 2 resulting in a transition to a slower rate earlier than in the DSP-based monolayer. The end of Stage 2 is indicated by the presence of Stage 3. With both adlayers, the apparent rate slows to 𝑘𝑘ℎ3 that is similar to 𝑘𝑘ℎ1 (𝑘𝑘ℎ3 = 260 M−1s−1 for DSP and 30 M−1s−1 for DSU). The DSU-based monolayer follows 𝑘𝑘ℎ3 until completion at ~2 h. Thus, the majority of the reaction of the DSU-monolayer is in the final stage with a 𝑘𝑘ℎ3 value of 3.0 ± 0.9 × 101 M−1s−1. The origin of the decreased reactivity for the DSU-based monolayers will be elaborated on by the determination of thermodynamic activation parameters in the next section. 116 4.3.4 Thermodynamic Activation Parameters of DSP- and DSU-based Monolayers The thermodynamic activation parameters for the transition state (i.e., the Arrhenius pre-exponential factor (𝐴𝐴) and activation energy (𝐸𝐸𝐴𝐴 ), from the Arrhenius equation40 and the enthalpy of activation (Δ𝐻𝐻 ‡ ), entropy of activation (Δ𝑆𝑆 ‡ ), and free energy of activation (Δ𝐺𝐺 ‡ ) from the Eyring equation40-41) provide insight into how the confined environment of a monolayer can influence the rate of a reaction at a liquid-solid interface.15,42-43 For the hydrolysis of an NHS ester, the transition state is the tetrahedral intermediate depicted in Figure 4.4. This state is reached by attack of the OH− nucleophile at the electropositive carbonyl carbon of the ester linkage, and is the rate limiting step for hydrolysis in alkaline solution.44 To this end, the hydrolysis reaction rate constants for the bulk phase were determined as previously described at three different temperatures (T = 20, 30, and 40°C). An example of these results, given as plots of -ln(1−x) versus time, is shown in Figure 4.5A for the DSP-based monolayer. Activation parameters were then determined using the linearized forms of the Arrhenius equation (Equation 4.4) and the Eyring equation 1 (Equation 4.5). By a least-squared fit to a plot of ln 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 versus 𝑇𝑇, one can determine 𝐴𝐴 and 𝐸𝐸𝐴𝐴 [J mol−1] by the Arrhenius equation: ln 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 = ln 𝐴𝐴 − 𝐸𝐸𝐴𝐴 𝑅𝑅𝑅𝑅 (4.4) where, in this case, 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 again represents the bimolecular heterogeneous reaction rate constant for bulk phase base hydrolysis (i.e., 𝑘𝑘ℎ2 for DSP and 𝑘𝑘ℎ3 for DSU), 𝑅𝑅 is the gas constant [J mol−1 K−1], and 𝑇𝑇 is the absolute temperature [K].45 117 Figure 4.4. Schematic of NHS ester-terminated monolayer where attack of the acyl carbon during hydrolysis or aminolysis produces a tetrahedral intermediate transition state (red). With a more densely packed (crystalline-like) monolayer, as with the DSU-based monolayer, the transition state is less accessible and "tighter" due to the increase in order and interchain interactions. 118 Figure 4.5. Plots for the determination of thermodynamic activation parameters. (A) Kinetic plot showing differences in bulk transformation stage of hydrolysis of DSP-based monolayers on gold after immersion in 50 mM BB, pH 8.50 at 20˚C (solid line), 30˚C (dotted line), and 40˚C (dashed line). (B) Thermodynamic plot for the hydrolysis of DSP-based (black) and DSU-based (green) monolayers on gold by immersion in 50 mM BB, pH 8.50. 119 Similarly, by a least-squared fit to a plot of ln 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑇𝑇 1 versus 𝑇𝑇, one can determine Δ𝐻𝐻 ‡ [kJ mol−1] and Δ𝑆𝑆 ‡ [J mol−1 K−1] by the Eyring equation: ln 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 −Δ𝐻𝐻 ‡ 1 𝑘𝑘𝐵𝐵 Δ𝑆𝑆 ‡ = + 𝑙𝑙𝑙𝑙 + 𝑇𝑇 𝑅𝑅 𝑇𝑇 ℎ 𝑅𝑅 (4.5) where 𝑘𝑘𝐵𝐵 is the Boltzmann constant [J K−1] and ℎ is Plank's constant [J s].46 Eyring plots from these measurements are shown in Figure 4.5B for the hydrolysis of the DSP- and DSU-based monolayers in 50 mM BB (pH 8.50); the thermodynamic activation parameters are summarized in Table 4.2. The results of a student's t-test analysis, which examines the differences in activation parameters between these two adlayers, is given in the Appendix. The data in Table 4.2 show that there are notable differences in the values of Δ𝐻𝐻 ‡ and 𝐸𝐸𝐴𝐴 for the two monolayers but that the values for 𝐴𝐴 and Δ𝑆𝑆 ‡ are identical. However, the most significant difference between the activation parameters calculated here and those previously reported for similar systems is between the 𝐴𝐴 and Δ𝑆𝑆 ‡ values.10 Whereas our ‡ ‡ = −111 J mol−1 K−1 and Δ𝑆𝑆𝐷𝐷𝐷𝐷𝐷𝐷 = analysis gave similar Δ𝑆𝑆 ‡ values for both adlayers (Δ𝑆𝑆𝐷𝐷𝐷𝐷𝐷𝐷 −112 J mol−1 K−1), Schönherr et al. report a Δ𝑆𝑆 ‡ value of −176 J mol−1 K−1 for the hydrolysis of a DSU-based adlayer on gold.10 Similarly, the 𝐴𝐴 values herein (𝐴𝐴𝐷𝐷𝐷𝐷𝐷𝐷 = 2.9 × 107 M−1 s−1 and 𝐴𝐴𝐷𝐷𝐷𝐷𝐷𝐷 = 3.0 × 107 M−1 s−1) are similar while that reported by Schönherr et al. is 2.1 × 104 M−1 s−1. The negative values of Δ𝑆𝑆 ‡ for all systems reflect the loss of entropy with formation of the tetrahedral intermediate transition state (i.e., the transition state is more ordered than the reactants) especially due to the tight packing around the reaction centers (i.e., carbonyl carbon), Figure 4.4.14,47 This loss of entropy for the transition state when compared to the reactants is often identified as a "tight" transition 120 Table 4.2. Activation parameters for the hydrolysis of DSP- and DSU-based monolayers on gold by immersion in 50 mM BB (pH 8.50). adlayer DSP DSU 𝐴𝐴 (M s ) 2.9 ± 0.3 × 107 3.0 ± 1.5 × 107 −1 −1 𝐸𝐸𝐴𝐴 (kJ mol−1) 24 ± 4 34 ± 6 Δ𝑆𝑆 ‡ (J mol−1 K−1) −111 ± 13 −111 ± 18 Δ𝐻𝐻 ‡ (kJ mol−1) 21 ± 4 31 ± 6 Δ𝐺𝐺 ‡ @T = 298 K (kJ mol−1) 54 ± 12 64 ± 16 121 state, where the transition state is more restricted and less susceptible to reaction.47-49 However, the less negative value of Δ𝑆𝑆 ‡ for our system compared to that of Schönherr et al. is attributed to either (1) less ordering of the transition state or (2) more ordered reactants in the monolayers reported herein.50 The speculation for more ordered reactants is supported by the increase in strength of the CH2 bands in our IR analysis in comparison ‡ ‡ to these other works.10,51 As such, the Δ𝐻𝐻 ‡ values (Δ𝐻𝐻𝐷𝐷𝐷𝐷𝐷𝐷 = 21 kJ mol−1 and Δ𝐻𝐻𝐷𝐷𝐷𝐷𝐷𝐷 = 31 kJ mol−1) and 𝐴𝐴 values also do not support the explanation for a less ordered transition state herein, due to a similar Δ𝐻𝐻 ‡ value of 30 kJ mol−1 and a lower 𝐴𝐴 value reported by Schönherr et al.48,50 Thus, we believe the reactants (i.e., as-prepared monolayer) are more ordered initially when compared to those prepared by Schönherr et al. The more notable changes in the activation parameters between the two ‡ ‡ monolayers are represented in Δ𝐻𝐻 ‡ (Δ𝐻𝐻𝐷𝐷𝐷𝐷𝐷𝐷 = 22 kJ mol−1 and Δ𝐻𝐻𝐷𝐷𝐷𝐷𝐷𝐷 = 31 kJ mol−1) and 𝐸𝐸𝐴𝐴 (𝐸𝐸𝐴𝐴,𝐷𝐷𝐷𝐷𝐷𝐷 = 24 kJ mol−1 and 𝐸𝐸𝐴𝐴,𝐷𝐷𝐷𝐷𝐷𝐷 = 34 kJ mol−1). The increases in Δ𝐻𝐻 ‡ and 𝐸𝐸𝐴𝐴 of the DSU-based monolayer are suspected to originate from two sources, both of which indicate a "tighter" transition state for DSU. One, the DSU-based monolayer exhibits increased interchain interactions due to the contributions from the additional methylene groups; this is similar to many other studied alkanethiolate derived monolayers.23 The increase in lateral molecular interactions likely stabilizes the DSU-based monolayer introducing an enthalpic penalty and a higher energy barrier (i.e., increase in 𝐸𝐸𝐴𝐴 ) for reaction of the DSU- based system to the transition state.10,52 Two, the DSU-based monolayer is more hydrophobic than the DSP-based monolayer (shown in the appendix) which translates to a decrease in the local dielectric constant of the DSU-based monolayer. The more hydrophobic local environment of the DSU-based monolayer is less favorable to the polar 122 tetrahedral intermediate transition state, providing an additional enthalpic penalty for the hydrolysis of the DSU-based monolayer.49-50,53 As an aside, the decrease in hydrophobicity may also be less favorable to hydroxide ion penetration at the DSU-based monolayer which would decrease the reaction rate. Ultimately, the "tighter" transition state for the DSU-based monolayer over the DSP-based monolayer is suspected to lead to the observed decrease in 𝑘𝑘ℎ .48 4.3.5 Hydrolysis of DSP-based Monolayers in NaOH, Borate Buffer, and Phosphate Buffer The solution can also affect interfacial reaction rates by playing a role in the double layer.54 Past reports on the reactivity of NHS esters in solution have not found a dependence on counter ions (e.g., phosphate, borate, and carbonate) at a fixed ionic strength, µ.55 Nonetheless, it is possible that the conditions used in our studies of base hydrolysis (i.e., BB at pH 8.5) and others (i.e., NaOH at pH 11) may contribute to the observed differences in reactivity.30 A summary of the values of 𝑘𝑘ℎ found using a range of solution reaction conditions is given in Table 4.3. The reaction rate constant for the hydrolysis of DSP-based adlayers in NaOH (1 × 10−3 M) has previously been reported by Dordi et al. as 0.61 M−1 s−1 and was replicated in our lab with a 𝑘𝑘ℎ1 of 0.68 M−1s−1 and 𝑘𝑘ℎ2 of 4.3 M−1s−1.10 The occurrence of 𝑘𝑘ℎ2 is suspected to be due to differences in the monolayer structure such as defects, as Dordi et al. reported a linear reaction profile for the entirety of the reaction. Nonetheless, the hydrolysis of a DSP-based monolayer in NaOH gave a similar rate constant regardless of ionic strength, µ, with a 𝑘𝑘ℎ2 of 5.4 ± 0.5 M−1s−1 when increasing µ from 0.001 to 0.4 M. 123 Table 4.3. Heterogeneous hydrolysis rate constants, 𝑘𝑘ℎ , for the hydrolysis of DSP-based adlayers in sodium hydroxide (NaOH), borate buffer (BB), and phosphate-buffered saline (PBS). solution 𝑘𝑘ℎ , M−1s−1 stage ionic strength, µ 1 × 10 M NaOH 0.68 Initiation 0.001 1 × 10−3 M NaOH 4.3 Bulk 0.001 1 × 10−3 M NaOH 5.4 Bulk 0.4 2 × 10−7 M NaOH −* −* 0.001 2 × 10−7 M NaOH −* −* 0.4 5 × 10−2 M BB 1500 Bulk 0.4 1 × 10−2 M PBS 6000 Bulk 0.16 DI H2O −* −* −* *Rate was immeasurable during time segment (0 < t < 7200 s) −3 pH 11 11 11 8.5 8.5 8.5 7.4 7.4 124 Decreasing the pH by immersion in 0.2 µM NaOH resulted in negligible reaction of a DSPbased adlayer (less than 2% in 2 h) for µ = 0.001 and µ = 0.4 M. Moreover, hydrolysis of a DSP-based monolayer in phosphate buffer (pH 7.4, µ = 0.16) gave an even higher rate constant than borate buffer with a 𝑘𝑘ℎ2 of 6.0 ± 0.5 × 103 M−1s−1, ~ 4× increase over that of borate buffer. Because 𝑘𝑘ℎ2 was significantly increased with the addition of borate and phosphate, we suspect the ion identity likely plays a role (i.e., stabilization of the tetrahedral intermediate) in the hydrolysis reaction kinetics at a surface.54 Still, with the difference in μ and pH for the BB and PBS systems, further studies are needed. Regardless of the 𝑘𝑘ℎ value used, whether 𝑘𝑘ℎ1 or 𝑘𝑘ℎ2 , the 𝑘𝑘ℎ values for the hydrolysis of DSP-based adlayers in borate buffer (50 mM, pH 8.50, µ = 0.4) are markedly larger than those in NaOH. Even increasing the ionic strength to match that in borate buffer does not appreciably change the rate constant in NaOH solutions in comparison to those observed in borate buffer. Additionally, decreasing the NaOH concentration to match that of the hydroxide concentration in borate buffer, which should produce indistinguishable rate constants given the constant hydroxide ion concentration, results in a negligible reaction. Thus, with NaOH or DI H2O seeming to suppress hydrolysis of the NHSterminated adlayer, we presume this would increase the extent of covalently bound proteins in aminolysis, given protein-compatibility with these solutions. 4.3.6 Aminolysis of NHS-terminated Monolayers To extend these kinetic studies to gain insight into protein immobilization, the kinetics of the aminolysis reaction were determined using EA via the parallel reaction rate 125 law shown in Equation 4.6: 𝑑𝑑ΓNHS = −𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 ΓNHS [𝑂𝑂𝑂𝑂 − ] − 𝑘𝑘𝑎𝑎 ΓNHS [𝑁𝑁𝑁𝑁2 ] 𝑑𝑑𝑑𝑑 (4.6) where 𝑘𝑘𝑎𝑎 represents the heterogeneous aminolysis rate constant, [𝑁𝑁𝑁𝑁2 ] is the deprotonated amine concentration, and 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 is the heterogeneous hydrolysis rate constant in the bulk phase obtained for the same solution devoid of EA. When the rate of hydrolysis is insignificant, a pseudo first order rate law can be used where the term for hydrolysis in Equation 4.6 (𝑘𝑘ℎ ΓNHS [𝑂𝑂𝑂𝑂 − ]) is ignored, which is the case in NaOH (0.2 µM, pH 8.50) which showed little to no reaction over the time period analyzed (< 2% over the course of 2 h), as stated earlier. The 𝑘𝑘𝑎𝑎 values obtained through either a pseudo first-order or parallel rate law by immersion in 500 mM ethylamine in 50 mM borate buffer (pH 8.50) or 500 mM ethylamine in NaOH (pH 8.50), respectively, will be compared to validate the calculation of the aminolysis rate constant using Equation 4.6 with and without the hydrolysis term (𝑘𝑘ℎ ΓNHS [𝑂𝑂𝑂𝑂 − ]). The results of the kinetic analyses for aminolysis of DSP-based monolayers after immersion in each of these solutions are shown in Figure 4.6. Analysis of the kinetic plot in Figure 4.6A using the parallel reaction rate law in Equation 4.6 gives the reaction rate constants shown in Figure 4.6B. Similar to past results, the aminolysis kinetic plots in Figure 4.6A do not show an initiation stage, unlike the hydrolysis kinetic plots.8 The linear trend for aminolysis throughout the entire reaction suggests that surface reactant accessibility (i.e., packing) is not a factor with aminolysis, potentially due to the difference in charge of the two nucleophiles (i.e., EA is positively charged while OH is negatively charged). Nonetheless, there are two important comparisons to make in Figure 4.6. First, both aminolysis results for the DSP-based monolayer (n = 2) with 50 mM 126 Figure 4.6. Aminolysis kinetics for DSP-, DSH, DSO-, and DSU-based monolayers. (A) Kinetic plot for aminolysis of the DSP- (black), DSH- (red), DSO(blue), and DSU- (green) based monolayers in 500 mM ethylamine in NaOH at pH 8.50 with inset showing fit to DSP and DSU plots. A comparison to aminolysis of the DSP-based monolayer in 50 mM borate buffer at pH 8.50 is shown (open circles). (B) Corresponding second-order heterogeneous reaction rate constants, 𝑘𝑘𝑎𝑎 , as a function of methylene chain length, n. 127 borate buffer (black circles) and NaOH (open circles) give similar rate constants (𝑘𝑘𝑎𝑎,𝐵𝐵𝐵𝐵 = 0.94 ± 0.28 and 𝑘𝑘𝑎𝑎,𝑁𝑁𝑁𝑁𝑁𝑁𝐻𝐻 = 0.51 ± 0.23 M−1 s−1) in Figure 4.6B, which validates the use of Equation 4.6 with (parallel reaction rate law) and without the hydrolysis term (pseudo first- order reaction rate law). Unlike hydrolysis, which increases by three orders of magnitude with borate buffer in comparison to NaOH, the aminolysis rate constants are not statistically different. We suspect that the nucleophilic attack by ethylamine, which is larger and positively charged compared to the smaller, negatively charged hydroxide ion, could account for the lack of difference in the rate constants between the borate buffer and NaOH solutions. In this case, the proposed effects of borate (i.e., stabilization of the tetrahedral intermediate) may not affect the slower rate of aminolysis. Second, the inset plot in Figure 4.6A shows the "bulk transformation stage" for both the DSP-based and DSU-based monolayers. It is evident that the two monolayers have different reaction rates. In contrast to the hydrolysis experiments, for systems where only aminolysis is occurring, the time to reach 50% reacted is 560 ± 50 s for DSP-based adlayers and 940 ± 400 s for DSU-based adlayers. Thus, the aminolysis results show similar trends to those obtained by hydrolysis where the shortest methylene chain system, DSP, reacts faster than the longest methylene chain system, DSU. However, the reaction rate constants shown in Figure 4.6B do not demonstrate this methylene chain length trend to the same extent as hydrolysis with 𝑘𝑘𝑎𝑎 values of 5.0 ± 2.3 × 10−1 M−1s−1 and 2.8 ± 1.5 × 10−1 M−1s−1 for DSP-based and DSU-based monolayers, respectively. This difference of ~ 2× is, again, attributed to the larger nucleophile, EA, that may not have access to the same nucleation sites (i.e., defect sites) as the smaller hydroxide ion. In addition, the adlayer increases in hydrophobicity in comparison to the as-formed adlayer upon aminolysis, whereas 128 hydrolysis results in a decrease in hydrophobicity.10 This may cause a reduction in reactivity for the DSP-based monolayer, which upon aminolysis decreases in hydrophobicity to a greater extent than the more hydrophobic DSU-based monolayer. 4.3.7. Kinetic Implications of Hydrolysis Conditions on Protein Immobilization The heterogeneous hydrolysis and aminolysis reaction rate constants can be used to project the efficiency of protein immobilization at the short and long chain length monolayers. The aminolysis rate constants for the two different monolayers are similar (a 2× difference), while the hydrolysis rate constant for the DSP-based monolayer is ~50× that of the DSU-based monolayer. Taking a ratio of 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 to 𝑘𝑘𝑎𝑎 gives ~3000 and ~100 for DSP-based and DSU-based adlayers, respectively, which indicates a DSU-based adlayer is more likely to favor aminolysis over DSP. Still, an analysis including the limitations posed through the use of relevant protein concentrations can reveal the implications of hydrolysis on protein immobilization. For this analysis, a typical protein concentration around 100 µg/mL was used (i.e., or ~0.7 μM given MW of IgG). For DSP-based monolayers with the determined 𝑘𝑘ℎ,𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 (1.5 × 103 M−1 s−1) and 𝑘𝑘𝑎𝑎 (5.0 × 10−1 M−1 s−1) values, the extent of aminolysis is calculated by Equation 4.6 to give 0.003%, taking into account the amines available for conjugation; the amine concentration is based on exposed lysine residues at the IgG periphery that are deprotonated, depending on the solution pH. Conversely, for DSU-based monolayers with the determined 𝑘𝑘ℎ,𝑏𝑏𝑢𝑢𝑙𝑙𝑙𝑙 (3.0 × 101 M−1 s−1) and 𝑘𝑘𝑎𝑎 (2.8 × 10−1 M−1 s−1) values, the extent of aminolysis is 0.08%, nearly a 30× increase over that of DSP. These findings indicate that the efficiency of aminolytic coupling may be increased by the use of the longer DSU-based 129 monolayers over DSP-based monolayers. Still, the extent of aminolysis is small in buffer conditions that facilitate the competing reaction of hydrolysis (50 mM borate buffer, pH 8.50). 4.4 Conclusions Immobilization of proteins, whether covalent or through adsorption, remains an integral step in the fabrication of biosensors. We previously reported that the competing reaction of hydrolysis dominated aminolysis under common immunoassay conditions (50 mM borate buffer, pH 8.50) when using DSP-based monolayers. This finding suggested that proteins are likely not covalently linked on the surface but rather adsorbed via electrostatics, hydrogen bonding, or van der Waals interactions, which increases the probability for desorption thus potentially limiting biosensor performance. Here, we report that the use of a longer methylene chain system, DSU, results in a higher degree of amidization, nearly 30× greater than using a DSP-based monolayer in standard immunoassay conditions (50 mM borate buffer, pH 8.50). We attributed this to the increased packing density of the DSU-based monolayers that impedes the hydrolysis reaction through steric hindrance, as well as an increase in inter-chain interactions and hydrophobicity. However, because the protein concentration used in immunoassays is relatively low, the calculated extent of amidization with DSU is still low (0.08%) in comparison to hydrolysis. To minimize hydrolysis and increase amidization, we found that solutions of NaOH and DI H2O caused little to no hydrolysis compared to borate or phosphate buffers which may be the best option for increasing protein coupling efficiency given increasing the protein concentration is often cost-prohibitive. Ultimately, our 130 findings indicate that longer methylene chains like the DSU-based monolayer will likely increase covalent linking of proteins to the surface, especially when coupled with solution conditions that minimize hydrolysis. These studies are ongoing. 4.5 References (1) Hermansen, G. T. Bioconjugate Techniques. 2nd ed.; Academic Press, Inc.: New York, 2008; pp 139-140. (2) Rusmini, F.; Zhong, Z.; Feijen, J. Biomacromolecules 2007, 8, 1775-1789. (3) Nakano, K.; Taira, H.; Maeda, M.; Takagi, M. Anal. Sci. 1993, 9, 133-136. (4) Jonkheijm, P.; Weinrich, D.; Schroder, H.; Niemeyer, C. M.; Waldmann, H. Angew. Chem., Int. Ed. 2008, 47, 9618-9647. (5) Lomant, A. J.; Fairbanks, G. J. Mol. Biol. 1976, 104, 243-261. (6) Mädler, S.; Bich, C.; Touboul, D.; Zenobi, R. J. Mass Spectrom. 2009, 44, 694706. (7) Cuatrecasas, P., Indu, P. Biochemistry 1972, 11, 2291-2299. (8) Lim, C. Y.; Owens, N. A.; Wampler, R. D.; Ying, Y.; Granger, J. H.; Porter, M. D. Langmuir 2014, 30, 12868-12878. (9) Buijs, J.; van den Berg, P. A. W.; Lichtenbelt, J. W. T.; Norde, W.; Lyklema, J. J. Colloid Interface Sci. 1996, 178, 594-605. (10) Dordi, B.; Pickering, J. P.; Schonherr, H.; Vancso, G. J. Surf. Sci. 2004, 570, 5766. (11) Patel, N.; Davies, M. C.; Hartshorne, M.; Heaton, R. J.; Roberts, C. J.; Tendler, S. J. B.; Williams, P. M. Langmuir 1997, 13, 6485-6490. Van Ryswyk, H.; Turtle, E. D.; Watson-Clark, R.; Tanzer, T. A.; Herman, T. K.; Chong, P. Y.; Waller, P. J.; Taurog, A. L.; Wagner, C. E. Langmuir 1996, 12, 61436150. (12) (13) Wagner, P.; Hegner, M.; Kernen, P.; Zaugg, F.; Semenza, G. Biophys. J. 1996, 70, 2052-2066. (14) Chechik, V. Annu. Rep. Prog. Chem., Sect. B 2008, 104, 331-348. 131 (15) Abiman, P.; Crossley, A.; Wildgoose, G. G.; Jones, J. H.; Compton, R. G. Langmuir 2007, 23, 7847-7852. (16) Frey, B. L.; Corn, R. M. Anal. Chem. 1996, 68, 3187-3193. (17) Arakawa, T.; Timasheff, S. N. Biochemistry 1982, 21, 6545-6552. (18) Chi, E. Y.; Krishnan, S.; Randolph, T. W.; Carpenter, J. F. Pharm. Res. 2003, 20, 1325-1336. (19) An analysis of the reaction of the adlayers in H2O resulted in no conversion over 24 h. (20) Porter, M. D. Anal. Chem. 1988, 60, 1143A-1155A. (21) Laibinis, P. E.; Nuzzo, R. G.; Whitesides, G. M. J. Phys. Chem. 1992, 96, 50975105. (22) Evans, S. D.; Goppert-Berarducci, K. E.; Urankar, E.; Gerenser, L. J.; Ulman, A. Langmuir 1991, 7, 2700-2709. (23) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559-3568. (24) Delamarche, E.; Sundarababu, G.; Biebuyck, H.; Michel, B.; Gerber, C.; Sigrist, H.; Wolf, H.; Ringsdorf, H.; Xanthopoulos, N.; Mathieu, H. J. Langmuir 1996, 12, 1997-2006. (25) Jiang, L.; Glidle, A.; Griffith, A.; McNeil, C. J.; Cooper, J. M. Bioelectrochem. Bioenerg. 1997, 42, 15-23. (26) Bain, C. D.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1989, 5, 723-727. (27) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321-335. (28) Petrovykh, D. Y.; Kimura-Suda, H.; Tarlov, M. J.; Whitman, L. J. Langmuir 2004, 20, 429-440. (29) Petrovykh, D. Y.; Kimura-Suda, H.; Whitman, L. J.; Tarlov, M. J. J. Am. Chem. Soc. 2003, 125, 5219-5226. (30) Dordi, B.; Schonherr, H.; Vancso, G. J. Langmuir 2003, 19, 5780-5786. (31) Dordi, B.; Pickering, J. P.; Schönherr, H.; Vancso, G. J. Eur. Polym. J. 2004, 40, 939-947. 132 (32) Ulman, A. An Introduction to Ultrathin Organic Films: From Langmuir-Blodgett to Self-Assembly. Academic Press: New York, 1991. (33) Bertilsson, L. L., B. Langmuir 1993, 9, 141-149. (34) Dameron, A. A.; Ciszek, J. W.; Tour, J. M.; Weiss, P. S. J. Phys. Chem. B 2004, 108, 16761-16767. (35) Zhang, M.; Vogel, H. J. J. Biol. Chem. 1993, 268, 22420-22428. (36) Lide, D. R., CRC Handbook of Chemistry and Physics 2004-2005: A ReadyReference Book of Chemical and Physical Data. CRC Press: New York, 2004. (37) Miyazawa, T.; Shimanouchi, T.; Mizushima, S. I. J. Chem. Phys. 1956, 24, 408418. (38) Miyazawa, T.; Shimanouchi, T.; Mizushima, S. I. J. Chem. Phys. 1958, 29, 611616. (39) Zeroka, D.; Jensen, J. O.; Samuels, A. C. J. Mol. Struct. 1999, 465, 119-139. (40) Wright, M. R., An Introduction to Chemical Kinetics. John Wiley & Sons, Ltd.: West Sussex, England, 2004. (41) Eyring, H. Chem. Rev. 1935, 17, 65-77. (42) Schönherr, H.; Feng, C.; Shovsky, A. Langmuir 2003, 19, 10843-10851. (43) Abiman, P.; Wildgoose, G. G.; Crossley, A.; Jones, J. H.; Compton, R. G. Chem. Eur. J. 2007, 13, 9663-9667. (44) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry Part A: Structure and Mechanisms. 4 ed.; Kluwer Academic/Plenum Publishers: New York, 2000. (45) Clugston, M.; Flemming, R. Advanced Chemistry. Oxford University Press: Oxford, 2000. (46) Houseton, P. L. Chemical Kinetics and Reaction Dynamics. The McGraw-Hill Companies, Inc.: New York, 2001. (47) Shovsky, A.; Schonherr, H. Langmuir 2005, 21, 4393-4399. (48) Jackson, R. A., Mechanism in Organic Reactions. The Royal Society of Chemistry: Cambridge, 2004. Schaleger, L. L.; Long, F. A. Adv. Phys. Org. Chem. 1963, 1, 1-33. (49) 133 (50) Kwon, Y.; Mrksich, M. J. Am. Chem. Soc. 2001, 124, 806-812. (51) Tournier, E. J. M.; Wallach, J. B.; Blond, P. Anal. Chim. Acta 1998, 361, 33-44. (52) Salem, L. J. Chem. Phys. 1962, 37, 2100-2113. (53) Tewari, Y. B.; Schantz, M. M; Pandey, P. C.; Rekharsky, M. V.; Goldberg, R. N. J. Phys. Chem. 1995, 99, 1594-1601. (54) Capon, B.; Ghosh, B. C. J. Chem. Soc. B 1966, 472-478. (55) Cline, G. W.; Hanna, S. B. J. Org. Chem. 1988, 53, 3583-3586. CHAPTER 5 LAYER-BY-LAYER MAGNETIC POLYELECTROLYTE MICROCAPSULES AS IMMUNOASSAY LABELS 5.1 Introduction Giant magnetoresistive (GMR) sensors represent an intriguing approach to the sensitive and rapid detection of biomarkers.1-4 In a sandwich immunoassay, a captured biomarker is tagged with a magnetic nanoparticle (MNP) label and a GMR sensor can be used to detect the magnetic signature of the bound label.5-6 This provides a quantifiable readout by correlating the response of the MNP label with the amount of captured biomarker.7-9 Until now, most of the work in our laboratory has used commercially available MNPs,10-12 which often have low inherent magnetizations and high levels of nonspecific adsorption (NSA) that limit the potential for low-level detection. Part of the NSA problem rests with the difficulty in producing highly magnetic MNPs that are stable in aqueous solutions (i.e., do not undergo sedimentation). The rate of settling of a particle in solution can be described by Stoke's law, which is given in Equation 5.1: 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 = 2𝑔𝑔(𝜌𝜌𝑝𝑝 − 𝜌𝜌𝑚𝑚 )𝑟𝑟 2 9𝜂𝜂 (5.1) 135 where the terminal sedimentation velocity of a particle (𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 ) is dependent on acceleration due to gravity (𝑔𝑔, m s−2), particle density (𝜌𝜌𝑝𝑝 , kg m−3), solution density (𝜌𝜌𝑚𝑚 , kg m−3), particle radius (𝑟𝑟, m), and dynamic viscosity of the solution (𝜂𝜂, kg m−1 s−1).13 Equation 5.1 indicates that an increase in 𝑟𝑟 will result in an increase in the sedimentation rate, whereas the sedimentation rate of a larger construct can be reduced by decreasing 𝜌𝜌𝑝𝑝 . The work described herein examines the possible use of a "hollow" microcapsule to decrease 𝜌𝜌𝑝𝑝 , and thus 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , with MNPs embedded in the skin of the microcapsule in large numbers to enhance magnetic signal. Polyelectrolyte-based microcapsules, developed in 1998,14-15 can potentially reduce or eliminate settling by closely matching the density of the capsule to that of solution in order to approach neutral buoyancy. Hollow capsules can be produced by the layer-bylayer (LbL) deposition of polyelectrolytes on nano- or microparticles followed by removal of the solid, underlying particle template. This procedure is outlined in Figure 5.1A.14-16 Here, the deposition of the cationic polyelectrolyte, polyallylamine (PAH), is accomplished by interaction of the amine groups of PAH with the anionic polystyrene (PS) bead functionalized with a proprietary coating of sulfate esters by the manufacturer.17 As per manufacturer information, we expect the sulfate ester coating will impart a negative charge on the PS bead and support the subsequent deposition of the cationic polyelectrolyte, PAH, by electrostatic attraction.17-18 This is followed by centrifugal washing steps and the deposition of the anionic polyelectrolyte, polystyrene sulfonate (PSS), due to interaction of the sulfonate groups of PSS with the underlying amine groups in the PAH layer. The LbL procedure has been further adapted to include the incorporation of MNPs within the polyelectrolyte layers to make magnetic microcapsules (MMCs) by electrostatic 136 Figure 5.1. Procedure for polyelectrolyte microcapsule synthesis. (A) Layer-bylayer deposition on polystyrene (PS) bead by the addition of a cationic polyelectrolyte, polyallylamine (PAH) or an anionic polyelectrolyte, polystyrene sulfonate (PSS). (B) MNP incorporation between the repeats of PAH and PSS deposition steps. (C) Template dissolution using tetrahydrofuran (THF). 137 interactions between cationic MNPs and anionic polymers, Figure 5.1B.19-21 This procedure can also be modified for the use of anionic MNPs by inclusion after deposition of the cationic polymer. The magnetic nature of these MMCs has been used to regulate release of encapsulated material by either capsule rupture19 or modulated permeability20 for applications such as controlled drug delivery. Finally, the template dissolution of the PS bead core by tetrahydrofuran (THF) produces a "hollow" construct, Figure 5.1C. The proposed use of MMCs as immunoassay labels is shown in Figure 5.2. In this case, a target antigen (Ag) is specifically captured by the underlying antibody (Ab) layer and tagged by an MMC label. The MMC label, which is functionalized with a layer of Abs specific to the target Ag, provides a means for Ag quantification in that the accumulation of MMC labels is directly dependent on the amount of bound Ag. Lastly, a GMR sensor detects changes in the magnetic field due to the bound magnetic labels. An exemplar plot of the GMR readout, Figure 5.2B, shows the decrease in relative signal (i.e., voltage which correlates with resistance via Ohm's law) as the sensor is passed over the immunoassay address. The shape of the signal (i.e., decrease in signal at the center of the address) is due to the opposing direction of the magnetic field produced by the MNPs in comparison to the applied magnetic field (Happ). By quantification of the relative signal (i.e., amplitude of the signal highlighted by the red arrow), a dose response plot, exemplified in Figure 5.2C, is then used to quantify the Ag concentration in the sample. In drawing on the above discussion, we hypothesized that the properties of MMC labels could be manipulated to reduce the complications due to sedimentation, and at the same time, result in a magnification of the GMR response. We are not aware of prior work along these lines. Herein, we report on the inclusion of in-house synthesized zinc- 138 Figure 5.2. Schematic for GMR immunoassay showing GMR readout and exemplar GMR dose response plot. The GMR-based sandwich immunoassay (A) is analyzed by a giant magnetoresistive (GMR) sensor (B). A dose response plot (C) can be constructed from the different addresses to give magnetic signal as a function of antigen (Ag) concentration. 139 ferrite nanoparticles in polyelectrolyte microcapsules for eventual use as immunoassay labels with a high magnetic signature. Before elaborating on the improvements in sedimentation and magnetization of MMCs, we present the results of a detailed characterization of the LbL synthesis by zeta (ζ) potential measurements, Raman spectroscopy, and atomic force microscopy. Then, we present the results of sedimentation rate calculations and simulated sedimentation curves which demonstrate the decrease in sedimentation of MMCs over the magnetic particles that we have used previously. Finally, we show that the magnetization per label of these new materials surpasses those of singular nanoparticles by a large margin. This work provides a basis for the eventual inclusion of MMCs in a GMR immunoassay. 5.2 Experimental 5.2.1 Reagents and Materials Iron acetylacetonate (Fe(acac)3, 99%), zinc acetylacetonate (Zn(acac)2, 95%), benzyl ether (99%), and sodium polystyrene sulfonate (PSS, MW: 70,000 g mol−1) were obtained from Acros Organics; polystyrene beads (0.36, 0.46, and 4.5 µm) from Polysciences, Inc.; polyethyleneimine (PEI, MW: 25,000 g mol−1), citric acid (CA, 99%), and poly(allylamine hydrochloride) (PAH, MW: 15,000 g mol−1) from Sigma Aldrich; sodium chloride (NaCl) from Fisher Scientific; oleic acid (90%) and tetrahydrofuran (THF) from Alfa Aesar; and ethanol (EtOH, 200 proof) from DECON lab. All chemicals were used as received. All aqueous solutions were prepared using water purified by passage through a Barnstead water polishing system to obtain ASTM type 1 water (DI H2O) at a resistivity of 18.2 MΩ. 140 5.2.2 Magnetic Nanoparticle (MNP) Synthesis The zinc ferrite MNPs (~26 nm edge length) were synthesized via thermal decomposition as described elsewhere.22 Briefly, 0.194 g Fe(acac)3 and 0.219 g Zn(acac)2 was added to 1.2 mL oleic acid and 10.0 mL benzyl ether and heated to 290 ˚C at a rate of 20˚C/min. After 30 min, the MNPs were cooled to room temperature, washed via centrifugation, and resuspended in 10 mL toluene. An additional ligand exchange step functionalized the as-synthesized hydrophobic zinc ferrite MNPs with a hydrophilic ligand for aqueous suspension. This required sonication with 5 mL of ligand solution (~6 mg/mL PEI in DMSO) for 1 h followed by agitation on a rocking table for 48 h. Finally, the MNPs were washed with 10 mL EtOH via centrifugation (10190 g) and resuspended in 10.0 mL water to ~5 mg/mL. 5.2.3 Magnetic Microcapsule (MMC) Synthesis The procedure for microcapsule (MC) preparation was adapted from a procedure described elsewhere to include the addition of MNPs as shown in Figure 5.1.14-15 Polystyrene beads (0.36, 0.46, and 4.5 µm) served as templates for capsule formation by the layer-by-layer (LbL) deposition of the positively and negatively charged polyelectrolytes, PSS and PAH, respectively. The polystyrene beads (200 μL of ~2.5% w/w) were first washed three times by: (1) centrifugation (1677 g); (2) removal of the supernatant; and (3) resuspension in 200 μL DI H2O. LbL deposition used 5.0 mg/mL PAH in 0.50 M NaCl for 20 min, 200 μL stock MNPs (5 mg/mL) in 0.50 M NaCl for 20 min, or 5.0 mg/mL PSS in 0.50 M NaCl for 15 min. After the formation of each layer (PAH, MNP, or PSS), the samples were rinsed by the centrifugation steps (3×) described above and the 141 next layer was applied. This procedure was repeated until reaching the desired number of layers. For m magnetic layers (PSS/MNP@PAH) followed by n layers of PAH and p layers of PSS, the construct will be described as (PSS/MNP@PAH)mPAHnPSSp. The samples were then incubated overnight in THF to allow dissolution of the polystyrene core and then rinsed one time with THF and three times with DI H2O. The incubation with THF to remove the PS core was based on previously established procedures that take advantage of the solubility of PS in THF.23-24 5.2.4 MC Characterization by Zeta (ζ) Potential, Raman Spectroscopy, and Atomic Force Microscopy (AFM) The ζ potential of the MMCs was measured with a Zetasizer Nano (Malvern Instruments) using a folded capillary cell (Malvern DTS1070). The MMCs were diluted in DI H2O (5 μL in 995 μL DI H2O to give ~6 × 106 capsules/mL for 0.36 μm, ~5 × 108 capsules/mL for 0.46 μm, and ~ 5 × 107 capsules/mL for 4.5 μm given the concentration as supplied by the manufacturer); the pH of these solutions was ~6 to 7. Raman spectra of capsules dried on a gold substrate were collected with a DXR Raman microscope (Thermo Scientific) using a 633 nm laser at 3.0 mW, a 25 µm entrance slit, two exposures of 5.00 s, and 512 dark scans of 5.00 s to account for detector noise. The thickness of the capsule layer was estimated by topography measurements using a Multimode NanoScope V AFM (Digital Instruments) after drying samples on atomically smooth 1×1 cm template-stripped gold (TSG) substrates. Images were obtained with a 200 µm Si3N4 cantilever (Nanoprobes). 142 5.2.5 MMC Characterization by Scanning Electron Microscopy (SEM), Optical Microscopy, and Vibrating Sample Magnetometry (VSM) SEM analysis was performed with a field emission SEM (Hitachi S-4800) using a working distance of 8.2 mm and accelerating voltage of 3.0 kV. Samples for SEM characterization were prepared by dropcasting ~20 µL of MMCs in DI H2O on clean 1×1 cm gold substrates. These samples were then mounted with carbon conductive tabs on SEM pin stubs. Optical images were obtained with an Olympus microscope (BX5WI) using a 50x objective. Magnetic properties were measured at room temperature using a vibrating sample magnetometer, VSM EZ7 (Microsense, Lowell, MA). VSM samples were prepared by dropcasting MNPs or MMCs on kapton tape. A blank signal of the kapton tape was subtracted for construction of hysteresis curves. 5.2.6 Calculating 𝜌𝜌𝑝𝑝 for Dynabeads, MCs, and MMCs For the 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 calculations, 𝜌𝜌𝑝𝑝 for the Dynabeads, MCs, and MMCs is based on the apparent density, 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 , which is given by Eq. 5.1: 𝑛𝑛 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 = 𝑉𝑉𝑓𝑓,𝑖𝑖 𝜌𝜌𝑖𝑖 (5.1) 𝑖𝑖=0 where the volume fraction, 𝑉𝑉𝑓𝑓,𝑖𝑖 , of component i is multiplied by the density, 𝜌𝜌𝑖𝑖 , of component i.1 In Eq. 5.1, the number of components (i) dictates the value of n and the sum of 𝑉𝑉𝑓𝑓,𝑖𝑖 is 1. The value of n depends on the composition of the labels; values are: n = 2 for Dynabeads (i.e., ferrite - a mixture of magnetite and maghemite - and polystyrene), n = 2 for MCs (i.e., polymer shell and solution within the shell), and n = 3 for MMCs (i.e., 143 polymer shell, MNP, and solution within the shell). The values of 𝑉𝑉𝑓𝑓,𝑖𝑖 are given by particle content, either reported by the manufacturer for Dynabeads or estimated by the shell thicknesses for the MC and MMCs. The value of 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 for Dynabeads is approximated by polystyrene (63% to give 𝑉𝑉𝑓𝑓,1 of 0.63) and ferrite (37% to give 𝑉𝑉𝑓𝑓,2 of 0.37).2 The value for 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 for the MC is given by the "hollow core" (𝑉𝑉𝑓𝑓,1 of 0.96) and "shell" (𝑉𝑉𝑓𝑓,2 of 0.04 where shell thicknesses were obtained via AFM). The values for 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 for the MMCs incorporate the MNPs (𝑉𝑉𝑓𝑓,3 of < 0.001 to 0.02 due to incorporation of 100 to 50,000 MNPs). The density of the zinc ferrite MNPs of 5400 kg m−3 is taken from literature.3 The density of ferrite is approximated as 4950 kg m−3 in order to give the "best case scenario" with regard to sedimentation for the Dynabeads, as the density of ferrite ranges from 4950 to 5600 kg m−3.4,5 The density of the MC shell was approximated as polystyrene (PS) at 1040 kg m−3 due to the possible presence of residual PS, which gives the "worst case scenario" with regard to sedimentation for the MMCs (𝜌𝜌𝑃𝑃𝑃𝑃𝑃𝑃 = 1020 kg m−3 and 𝜌𝜌𝑃𝑃𝑃𝑃𝑃𝑃 = 801 kg m−3).6,7 All of the volume fraction and density values (𝑉𝑉𝑓𝑓,𝑖𝑖 , 𝜌𝜌𝑖𝑖 , and 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 ), along with the 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 calculations, are summarized in Table 5.1. 5.2.7 Giant Magnetoresistance (GMR) Coupon Design, Sensor Design, and Magnetic Detection Station The GMR coupons consisted of 12 gold addresses equally interspersed between 13 nickel addresses. Each address is 200 × 200 µm with 500 µm spacing edge-to-edge between each address. The nickel addresses serve as references for normalization of the magnetic signal. The gold addresses (150 nm) were formed with a titanium adhesion layer (~10 nm) while the nickel addresses (10 nm) were coated with parylene (< 1 μm) to prevent 144 Table 5.1 Calculated sedimentation velocities, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , per Equation 5.1. For labels with multiple components, the particle density is approximated by volume fractions, 𝑉𝑉𝑓𝑓 , to obtain the apparent particle density, 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 . 𝜌𝜌1 𝜌𝜌2 𝑉𝑉𝑓𝑓,2 (kg/m3) (kg/m3) Dynabead 0.63* 1040* 0.37** 4950** MC 0.96‡ 1000‡ 0.04* 1040* ‡ MMC+100 MNPs 0.96 1000‡ 0.04* 1040* MMC+100 MNPs 0.96‡ 1000‡ 0.04* 1040* MMC+100 MNPs 0.96‡ 1000‡ 0.03* 1040* MMC+100 MNPs 0.96‡ 1000‡ 0.03* 1040* MMC+100 MNPs 0.95‡ 1000‡ 0.03* 1040* MNP * Polystyrene (PS), ** Ferrite, ‡ DI H2O, + MNP (zinc ferrite) label 𝑉𝑉𝑓𝑓,1 𝑉𝑉𝑓𝑓,3 < 0.001+ < 0.001+ < 0.001+ < 0.001+ < 0.001+ 1.00+ 𝜌𝜌3 (kg/m3) 5400 5400 5400 5400 5400 5400 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 (kg/m3) 2487 1001 1002 1003 1009 1017 1080 5400 145 oxidation. Details of the GMR sensor and setup are described elsewhere.10-12, 25 Briefly, the GMR sensor (NVE Corp.) is comprised of four GMRs arranged in a Wheatstone bridge configuration. Two GMRs function as sense resistors for coupon reading and two act as reference resistors. The GMR sensor is bonded to a printed circuit board and connected to a Keithley 220 current source; the voltage drop across the bridge is read at a Keithley 2182 nanovoltmeter. For coupon analysis, an external magnetic field of 150 Oe is applied by two electromagnetic coils. 5.3 Results and Discussion The following sections describe the characterization of magnetic polyelectrolyte microcapsules (MMCs) for the eventual application to the magnetic-based diagnostic arena using giant magnetoresistive (GMR) sensors. Section 5.3.1 presents characterizations of the surface charge, composition, and thickness of the MMCs by ζ potential, Raman spectroscopy, and atomic force microscopy (AFM). Section 5.3.2 confirms MNP incorporation in the MC structure by scanning electron microscopy (SEM) and dark-field optical microscopy. Section 5.3.3 presents data from a sedimentation model illustrating the decrease in sedimentation resulting from the hollow structure of the MMC. Section 5.3.4 describes magnetic properties of the MMCs by vibrating sample magnetometry (VSM). Section 5.3.5 evaluates the use of MMCs as assay labels by detection of dropcast MMCs by a GMR sensor. This chapter closes with a brief outlook at the potential use of MMCs in immunoassays. 146 5.3.1 Characterization of MC Formation by ζ Potential, Raman Spectroscopy, and AFM For the following section, the MCs are synthesized using a polystyrene (PS) bead template (0.36, 0.46, or 4.50 μm diameter) with deposition of a cationic polyelectrolyte, polyallylamine (PAH), and anionic polyelectrolyte, polystyrene sulfonate (PSS). The stepwise synthesis of the MCs was characterized via ζ potential, shown in Figure 5.3. The ζ potential is a measurement of the relative potential in the interfacial double layer at a slipping plane around the particle surface in comparison to the bulk solution potential.26 The magnitude of the ζ potential cannot be accurately quantitated to give surface charge density since the polyelectrolyte layers are not a finite thickness and the conformation of the polyelectrolyte layer varies.27-29 Thus, we use the ζ potential to characterize the layer deposition by the change in the sign of ζ potential following the alternate deposition of the cationic and anionic polyelectrolytes on the polystyrene bead template. First, analysis of the ζ potential of the as-received PS beads (4.5 μm diameter) gives a negative ζ potential value of −48.8 ± 0.6 mV (DI H2O, pH ~7.0) and fits with expectations.17 Second, the formation of the first PAH, which relies on electrostatic interaction of the positively charged protonated amine group with the negatively charged sulfate esters of the PS bead, changes the sign of ζ potential to a positive value of +41.3 ± 3.6 mV. Lastly, the subsequent deposition of the polyelectrolyte layer of PSS relies on electrostatic interaction of the negatively charged sulfonate group of PSS with the positively charged protonated amine group of PAH. Upon deposition of the anionic polyelectrolyte, PSS, the ζ potential reverts back to a negative value of −38.4 ± 2.7 mV, as expected due to the anionic charge of the polymer. As shown in Figure 5.3, the alternation 147 Figure 5.3. ζ potential of stock polystyrene (PS) particles before and after addition of each polyelectrolyte layer, polyallylamine (PAH), or polystyrene sulfonate (PSS). The 0th layer is the stock PS particle while the 1st layer is PAH, 2nd layer is PSS, 3rd layer is PAH, 4th layer is PSS, 5th layer is PAH, 6th layer is PSS, 7th layer is PAH, and 8th layer is PSS. The error is representative of measurements of several different samples (n = 3). 148 in sign of the ζ potential between positive and negative values is consistent with the deposition of either the cationic or anionic polymer layer. Moreover, the absolute values of the ζ potentials after the deposition of each type of layer (e.g., 1st, 3rd, 5th, and 7th layer of PAH) are nearly equal, pointing to a high level of deposition consistency. Raman spectroscopy was used to confirm the chemical composition of each polyelectrolyte layer (PAH versus PSS) on the polystyrene bead template (4.5 μm diameter) by the vibrational signature of the charged species, i.e., an amine or sulfonate group. An example of these spectra is shown in Figure 5.4 with band assignments shown in Table 5.2. First, spectra for the as-received PS bead (black) show several bands consistent with the expected composition: aromatic ν(CH) at 3055 cm−1; aliphatic ν(CH) at 2903 and 2855 cm−1; aromatic ν(C−C−H) at 1604 and 1584 cm−1; δ(C−H) at 1454, 1203, 1184, and 1158 cm−1; and aromatic ν(C−C) at 1002 cm−1.30-31 These bands are indicative of the aromatic ring and saturated alkane backbone of polystyrene. The sulfate ester coating of the PS bead likely gives rise to the sulfonate stretch, ν(SO3−), at 1032 cm−1.30-31 Second, the spectra of the PS bead after the deposition of the first PAH layer (PAH@PS in red) show the same bands as the PS bead (PS in black). The anticipated bands due to the primary amine of PAH (δ(N−H) at 1583 cm−1 and v(C−N) at 1450 cm−1) overlap with those of PS (ν(C−C−H) at 1584 cm−1 and δ(CH3) at 1454 cm−1), making it difficult to confirm the deposition of PAH on the PS bead.32 Lastly, the spectra of the PAH@PS beads after the deposition of PSS (PSS@PAH@PS in blue) show the same bands as the PS beads with an additional band at 1133 cm−1 which we assign to the sulfonate group, ν(SO2) of -SO3H. This indicates the successful application of PSS.30-31 149 Figure 5.4. Raman spectra of PS beads, PAH-coated PS beads (PAH@PS), and PSS/PAH-coated PS beads (PSS@(PAH@PS)), dried on a gold substrate. 150 Table 5.2. Raman band assignments for PAH and PSS deposited on PS beads. mode assignment description PS ν(CH), ar Aromatic CH stretch 3055 ν(CH), ali Aliphatic CH stretch 2903 ν(CH), ali Aliphatic CH stretch 2855 ν(CCH) CCH ring quadrant stretch 1604 ν(CCH) CCH ring quadrant stretch 1584 δ(N−H) N−H bend δ(CH2) Methylene bend 1454 ν(C−N) C−N stretch (CH) CH deformation δ(CH2) Methylene bend 1203 δ(CH2) Methylene bend 1184 δ(CH2) Methylene bend 1158 ν(SO2) SO2 stretch of -SO3H ν(SO3) SO3 stretch of -SO3− 1032 νs(C−C), ar Aromatic C−C stretch 1002 *overlap of bands makes differentiation difficult **overlap of bands makes differentiation difficult peak position (cm−1) PAH@PS PSS@(PAH@PS) 3054 3057 2909 2915 2853 1602 1601 1583* 1582 1583* 1450** 1453 1450** 1350 1201 1200 1183 1183 1155 1154 1133 1032 1031 1002 1001 151 Core dissolution of the PSS/PAH-coated PS spheres results in a hollow capsule which collapses upon drying. This structure is evident in the AFM image (6×6 μm) of the dried MC (PAHnPSSp where n = 4 and p = 4) shown in Figure 5.5A. Estimations of the MC shell thickness can be obtained from the topographic cross-section of the dried capsule based on the height of the capsule (where no folds are present) with respect to the underlying substrate. The location of the cross section analyzed is shown by the white line, with a location on the underlying smooth substrate marked by the red circle and the dried capsule layers marked by the blue circle, in Figure 5.5A. These locations correspond to those in the cross-sectional profile in Figure 5.5B, resulting in a thickness estimate of ~24 nm. This analysis was performed on MCs prepared using three different PS bead templates (diameters of 0.36, 0.46, and 4.5 μm) with either 3 or 9 layer depositions (PAHnPSSp where n = 2 or 4 and p = 1 or 4). The results are summarized in Figure 5.5C. As expected, the shell thickness for capsules prepared with 9 layers is larger than that for 3 layers. The shell thickness values can be used to determine an average layer thickness, calculated as follows: 3.1 nm (0.36 μm), 3.4 nm (0.46 μm), and 4.2 nm (4.50 μm) for the 3-layer capsules and 2.7 nm (0.36 μm), 2.1 nm (0.46 μm), and 2.1 nm (4.50 μm) for the 9-layer capsules. Based on the size of the molecular groups, the theoretical average value for a polyelectrolyte layer is ~0.7 nm, which is similar to the values reported here.33-34 Overall, the layer thicknesses are comparable for all MCs, which indicates a similar capsule structure (e.g., polyelectrolyte density and porosity). However, the MCs with 3 layers give larger layer thicknesses than those with 9 layers. This may be due to reorientation of the polyelectrolytes with additional layer deposition that results in a more compact polyelectrolyte shell. Second, the layer thickness increases with an 152 Figure 5.5. AFM analysis of microcapsule structure. (A) AFM image of microcapsule (9 layers) with markers representing the start (red) and end (blue) points along the cross-sectional line from which the shell thickness can be determined, (B) corresponding cross-section profile from line in (A) with lines representing the start (red) and end (blue) points shown in (A), and (C) MC shell thickness as a function of polystyrene bead template diameter and total number of deposited polyelectrolyte layers. The uncertainty in the shell thicknesses is attributed to a combination of variations in the polymeric deposition, underlying surface roughness, or differences in folding of the MC layers. 153 increase in bead diameter for the 3-layer MCs. This trend may originate from the increased curvature with a decrease in bead diameter. Similar to the Daoud-Cotton model, interpenetration of polymers on a large PS bead may increase due to a decrease of surface curvature which results in a larger layer thickness for larger PS beads.35 This trend does not hold, however, for the 9-layer MCs where instead the layer thickness decreases with an increase in bead diameter. Again, this may be due to a reorientation of the polymers after the accumulation of many layers. 5.3.2 Characterization of MMCs by SEM and Optical Microscopy Verification of the MMC hollow structure and inclusion of MNPs is based on the SEM and optical images in Figure 5.6. The SEM images show MCs before (Figure 5.6A) and after (Figure 5.6B) core removal and with MNPs (Figure 5.6C and 5.6D). Upon core removal, the structure of the microcapsule changes from a solid spherical shape to a collapsed hollow structure (PAHnPSSp where n = 4 and p = 4 on 4.5 μm diameter PS beads) after drying as previously shown in the AFM image in Figure 5.5A. The collapsed structure maintains its circular shape with a diameter (4.5 ± 0.4 µm, n = 11) similar to that of asreceived polystyrene beads (4.5 ± 0.3 µm, n = 30) used as the core template. The folding of the collapsed structure indicates the polymer layers are thin and provide little resistance to bending. Compared to the microcapsules without MNPs, the microcapsules with MNPs appear rough, which is due to the presence of MNPs in the microcapsule layers. Note that the layers deposited to encapsulate the MNPs appear to obscure some of the morphological features in the SEM images. The dark-field microscope images shown in Figure 5.6E and 5.6F give further 154 Figure 5.6. SEM images of (A) 0.46 µm diameter polystyrene beads used as template, (B) microcapsules (MCs) without magnetic nanoparticles (MNPs), and (C, D) magnetic microcapsules (MMCs) with MNPs. MMCs show rougher topography than MCs which is attributed to the embedded MNPs. Note the scale bars vary in size. Dark-field microscope images of (E) MCs without MNPs and (F) MMCs with MNPs. (G-J) Fluorescent microscope images of FITC-tagged MCs showing little aggregation in solution (DI H2O). 155 evidence for the presence of embedded MNPs. The dark-field attachment provides a means to visualize nanoparticles by their ability to scatter light.36 This is apparent when comparing the extent of scattered light from the MCs (without MNPs) in Figure 5.6E to that of the MMCs in Figure 5.6F. To assess MMC aggregation, the MMCs were also imaged as a suspension in DI H2O. To more clearly image the MMCs in solution, the outer PAH surface of the MMC was tagged with a fluorescent molecule, FITC, by simple adsorption. The representative fluorescent microscope images shown in Figure 5.6G-J give little to no evidence of MMC agglomeration or aggregation, represented by the single (or dual) MMCs suspended in DI H2O. These images verify that the MMCs exist as discrete labels in solution, and that the aggregation observed for the dried samples in the AFM, SEM, and microscope images is an artifact of the drying step needed for sample preparation. 5.3.3 Sedimentation Calculations for MMCs by the Stokes and Mason-Weaver Equations This section evaluates the sedimentation properties of the MMCs by an estimation of: (1) the settling rate, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , via Stokes equation (Equation 5.1), and (2) the concentration profiles via the Mason-Weaver equation (Equation 5.2) using MATLAB software (MATLAB R2012b, MathWorks Inc., Nattick, MA).37-38 The MMCs are expected to reduce sedimentation, and thus nonspecific adsorption, due to a density that is more closely matched to that of the solution. Thus far, GMR immunoassays in our laboratory have used a commercially available magnetic construct from Thermo Fisher Scientific.10-12 These magnetic constructs are cross-linked polystyrene microspheres with dispersed maghemite 156 (γ-Fe2O3) and magnetite (Fe3O4).39 The sedimentation properties for Dynabeads were also analyzed for comparison to the MMCs. The sedimentation of particles can be approximated by the Stokes equation (Equation 5.1), where the rate of settling of a particle in solution, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , is proportional to the density, 𝜌𝜌𝑝𝑝 , and radius, 𝑟𝑟, of the particle. The calculations for 𝜌𝜌𝑝𝑝 for each particle are shown in Table 5.1. Assuming the density and viscosity of the solution to be that of water at 25°C (𝜌𝜌𝑚𝑚 = 999.96 kg m−3 and 𝜂𝜂 = 0.00089 kg m−1 s−1), we can calculate 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 for the Dynabeads, MCs, and MMCs with different levels of MNP loading. These results are summarized in Table 5.3. From Table 5.3, we can see 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 for the MCs is small (𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 = 0.02) when compared to that of 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠,𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 . However, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 for MMCs increases in comparison to the MCs due to the incorporation of MNPs (𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 = 0.03, 0.04, 0.11, 0.21 by the incorporation of 100, 1,000, 5,000, and 10,000 MNPs in the MMCs, respectively). In fact, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 of the MMCs reaches that of Dynabeads with the incorporation of 50,000 MNPs (𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 = 0.99). At this point, the increase in density due to the large amount of MNPs in the MMCs coupled with the larger diameter of the MMCs vis-à-vis the Dynabeads, results in MMCs having little, if any, apparent sedimentation advantage over Dynabeads. On the other hand, due to the small size of the MNPs (0.03 μm diameter), 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 for the MNPs is insignificant in comparison to that of Dynabeads ( 𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 < 0.01) despite having the largest density. Although, with the MNPs, the gain in stability (i.e., lower 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 ) will likely be overshadowed by the lower magnetic signature. Interestingly, decreasing the MMC diameter to that of Dynabeads (~1.0 μm) results in 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 of the MMCs approaching the 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 of Dynabeads with the incorporation of ~600 MNPs rather than ~50,000 MNPs. Thus, there 157 Table 5.3. Calculated sedimentation velocities, 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , per Equation 5.1. For labels with multiple components, the particle density, 𝜌𝜌𝑝𝑝 , is approximated by volume fractions, 𝑉𝑉𝑓𝑓 , to obtain the apparent particle density, 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 , Table 5.1. label Dynabeads MC MMC + 100 MNPs MMC + 1,000 MNPs MMC + 5,000 MNPs MMC + 10,000 MNPs MMC + 50,000 MNPs MNPs *𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 = 𝜈𝜈 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 𝑠𝑠𝑠𝑠𝑠𝑠,𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝜌𝜌𝑝𝑝 or 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 (kg m−3) 2487 1001 1002 1003 1009 1017 1080 5400 d (μm) 1.05 4.5 4.5 4.5 4.5 4.5 4.5 0.03 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 (m s−1) 1.00 × 10−6 1.74 × 10−8 2.53 × 10−8 3.77 × 10−8 1.12 × 10−7 2.11 × 10−7 9.93 × 10−7 2.42 × 10−9 𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠 * 0.02 0.03 0.04 0.11 0.21 0.99 < 0.01 158 is a limit on the amount of MNP incorporation, which changes with diameter, in order to maintain the improvements in 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 for hollow capsules. Nevertheless, we can see that 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 can be reduced by the use of MMCs. The sedimentation of the particles can be further analyzed by the concentration profiles over suspension height and time by providing an overall picture for the sedimentation of the different particles when at equilibrium with diffusion. Using 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 , the net transport of particles can be analyzed by the Mason-Weaver equation shown in Equation 5.2: 𝜕𝜕𝜕𝜕 𝜕𝜕 2 𝐶𝐶 𝜕𝜕𝜕𝜕 = 𝐷𝐷 2 + 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 𝜕𝜕𝜕𝜕 𝜕𝜕 𝑧𝑧 𝜕𝜕𝜕𝜕 (5.2) where 𝐶𝐶 is the concentration of particles in the suspension (mol), 𝐷𝐷 is the diffusion coefficient (m2 s−1), 𝑡𝑡 is time (s), and +𝑧𝑧 is defined as the direction opposing the gravitational force.40 The Stokes-Einstein equation is used to approximate the diffusivity as: 𝐷𝐷 = 𝑘𝑘𝐵𝐵 𝑇𝑇 6𝜋𝜋𝜋𝜋𝜋𝜋 (5.3) where 𝑘𝑘𝐵𝐵 is the Boltzmann constant (1.38 × 10−23 J K−1) and 𝑇𝑇 is the temperature (K).13 The finalized form of the Mason-Weaver equation has been discussed in detail in another reference and can be numerically solved using MATLAB software to give net transport profiles for a given particle.37 The numerical solution of the Mason-Weaver equation requires an input for the height of the suspension (i.e., total suspension height, z) to determine concentration profiles as a function of height. In addition, a window location (i.e., a window in the z direction where an average particle concentration can be calculated) is needed for analysis of the 159 concentration profiles as a function of time. Unless stated otherwise, we set z to 2 mm and the window location as 5 μm. These parameters are based on the current setup of our immunoassays where accumulation of labels occurs at the surface (which may range from 1 to 5 μm, depending on the label diameter) of a 2 mm high droplet. The net transport profiles from the MATLAB model, shown as plots of concentration as a function of suspension height in Figure 5.7A (t < 100 h), demonstrate the differences in the equilibrium between sedimentation and diffusion for the various labels. Note that the concentration profiles versus height for the MCs and MNPs were obtained at t = 5000 h (due to the lower sedimentation rates) while the other labels were obtained at t = 100 h. From Figure 5.7A, we can estimate the boundary layer, taken to be the point at which the particle concentration decreases to less than 1% of the initial particle concentration, at the equilibrium of sedimentation and diffusion. The boundary layer thicknesses for each label are as follows: 6 μm (Dynabead), 1 μm (MMC + 50,000 MNPs), 7 μm (MMC + 10,000 MNPs), 12 μm (MMC + 5,000 MNPs), 32 μm (MMC + 1,000 MNPs), and 46 μm (MMC + 100 MNPs). The MCs and MNPs at equilibrium had boundary layer thicknesses greater than ~1.5 mm. From this analysis, it is evident that the boundary layer thickness differs for each label, even with surprisingly small differences in density (e.g., 32 μm for MMC + 1,000 MNPs with 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 of 1003 kg m−3 and 46 μm for MMC + 100 MNPs with 𝜌𝜌𝑝𝑝,𝑎𝑎𝑎𝑎𝑎𝑎 of 1002 kg m−3). Further, the boundary layer thickness trends with label density and size, with the largest, densest labels (MMC + 50,000 MNPs) having the smallest boundary layer thickness of 1 μm. The differences in boundary layer thickness can be examined by solving for the 160 Figure 5.7. Particle concentration profiles as a function of (A) solution height (where x refers to +z direction), and (B) time (t = 100 h or 5000 h). In (A), the data for MMCs+10,000 MNPs (grey) overlaps with Dynabeads (red). In (B), the data for MMCs+50,000 MNPs (orange) overlaps with Dynabeads (red). 161 equilibrium expression of the Mason-Weaver equation (Equation 5.2) (i.e., 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 = 0) upon proper integration and substitution for 𝜈𝜈𝑠𝑠𝑠𝑠𝑠𝑠 (Equation 5.1) and 𝐷𝐷 (Equation 5.3): 4𝜋𝜋𝜋𝜋(𝜌𝜌𝑝𝑝 − 𝜌𝜌𝑚𝑚 )𝑟𝑟 3 𝐶𝐶2 𝑙𝑙𝑙𝑙 = (𝑧𝑧2 − 𝑧𝑧1 ) 𝐶𝐶1 3𝑘𝑘𝐵𝐵 𝑇𝑇 (5.4) where 𝐶𝐶1 and 𝐶𝐶2 are the equilibrium concentrations at positions 𝑧𝑧1 and 𝑧𝑧2 , respectively.37, 41 In Equation 5.4, an increase in 𝜌𝜌𝑝𝑝 would result in a decrease in solution height (𝑧𝑧2 − 𝑧𝑧1 ) 𝐶𝐶 𝐶𝐶 with a given concentration ratio, 𝐶𝐶2. Thus, when we set the concentration ratio, 𝐶𝐶2, to that 1 1 𝐶𝐶 obtained by the boundary layer (i.e., 𝐶𝐶2 = 0.01𝐶𝐶1 or 𝐶𝐶2 = 0.01), the boundary layer thickness 1 (𝑧𝑧2 − 𝑧𝑧1 ) should decrease as 𝜌𝜌𝑝𝑝 is increased, as shown in Figure 5.7A. In a particle suspension that is sedimentation-free (i.e., ideal particle stability), the boundary layer thickness will approach that of the entire solution height. In other words, the particles will be truly suspended at equilibrium of diffusion and sedimentation. This is nearly the case with the MCs and MNPs which both have boundary layers greater than 1.5 mm in a 2 mm solution. The MMCs, and to an even greater extent the Dynabeads, with boundary layer thicknesses that range from ~6 to 50 μm, deviate from an "ideal" particle suspension. In Equation 5.4, an increase in 𝜌𝜌𝑝𝑝 would also result in an increase in 𝐶𝐶2 𝐶𝐶1 , given a solution height (𝑧𝑧2 − 𝑧𝑧1 ). This is exemplified when setting the solution height to that of a sedimentation layer (i.e., 𝑧𝑧2 − 𝑧𝑧1 = 5 μm) where an increase in 𝜌𝜌𝑝𝑝 results in an increase in 𝐶𝐶2 𝐶𝐶1 𝐶𝐶 𝐶𝐶 . The 𝐶𝐶2 values at equilibrium (i.e., > 100 h), 𝐶𝐶2 , from Figure 5.7A are as follows: 405 1 1 0 (Dynabead, MMC + 50,000 MNPs, and MMC + 10,000 MNPs), 403 (MMC + 5,000 MNPs), 332 (MMC + 1,000 MNPs), and 276 (MMC + 100 MNPs). Due to the small 162 𝐶𝐶 suspension height, the 2 values of the less dense labels were much lower than those for 𝐶𝐶 1 0 the Dynabeads and MMCs at 9 (MCs) and 1 (MNPs). Thus, this analysis also exhibits a trend with particle density with the larger, denser labels (Dynabeads, MMC + 50,000 𝐶𝐶 MNPs, and MMC + 10,000 MNPs) giving the highest value of 𝐶𝐶2 . In order to compare settling rates, the 𝐶𝐶2 𝐶𝐶1 1 0 𝐶𝐶 values are normalized with 𝐶𝐶2 and 1 0 plotted as a function of time, Figure 5.7B. These concentration profiles provide a comparison of the sedimentation rates of the labels by (1) the rise of the sedimentation 𝐶𝐶 curve (i.e., slope), and (2) the time needed to reach the 2 . First, the rise of the 𝐶𝐶 1 0 sedimentation curve is steepest for Dynabeads and MMCs + 50,000 MNPs (i.e., undergo fastest sedimentation) and decreases as the number of MNPs in the MMCs is decreased. 𝐶𝐶 Second, the time at which the particles reach 𝐶𝐶2 are as follows: ~36 min (Dynabead and 1 0 MMCs + 50,000 MNPs), ~3 h (MMC + 10,000 MNPs), ~6 h (MMC + 5,000 MNPs), ~18 h (MMC + 1,000 MNPs), ~29 h (MMC + 100 MNPs), ~650 h (MNPs), and ~3,600 h (MCs). Clearly, the time needed to reach the saturation concentration at the bottom of the suspension trends with particle density and size, with the larger, denser labels reaching 𝐶𝐶 𝐶𝐶2 the fastest. 1 0 In comparison to Dynabeads, the sedimentation curves of the MMCs demonstrate that incorporation of the MNPs in the MC structure causes increases in sedimentation. However, the increase in sedimentation of the MMCs is accompanied by an enhancement in magnetization which also increases upon incorporation of MNPs. Further, in application of the MMCs as magnetic labels in an immunoassay, the incubation period may be well 163 under 16 h. Thus, with tuning of the MMC structure, we anticipate an ideal balance between the (1) sedimentation for the incubation period, and (2) magnetic enhancement can be achieved. 5.3.4 Characterization of MMCs by VSM To test the use of the MMCs as magnetic labels, MMCs, Dynabeads, and MNPs were analyzed via VSM to obtain remnant magnetization (Mr), coercivity (Hc), saturation magnetization (Ms), and magnetization values at the field used in our immunoassay GMR readout. The hysteresis curves of the MMCs, Dynabeads, and in-house MNPs used to construct the MMC are shown in Figure 5.8. Note that for comparison as assay labels, magnetization values (emu) are displayed with respect to label concentration (emu/label). Comparisons of the synthesized MMCs to commercially available MNPs, Dynabeads, and the in-house MNPs are shown in Table 5.4. As seen in Figure 5.8, the hysteresis curves have a ferromagnetic character, with a nonzero remnant magnetization (Mr) (i.e., the magnetization when the magnetic field is removed) and a nonzero coercivity (Hc) (i.e., the magnetic field that must be applied to eliminate any remnant magnetization of the MNPs). Here, upon incorporation of the MNPs in the MC to produce MMCs, Hc increases from 64 to 116 Oe. The increase in Hc observed here is likely a result of the decreased interactions between particles in the MC layers due to physical separation of the individual labels once immobilized within the MC layers.22, 42-43 This follows previous results shown for these MNPs by Park et al. where an increase in Hc was observed with the addition of silica shells, which was attributed to a decrease in magnetic dipole interactions due to the physical separation of the magnetic cores by the 164 Figure 5.8. Hysteresis curves of as-synthesized PEI-coated MNPs and MMCs with embedded PEI-coated MNPs. Note the normalization of magnetization to label concentration results in a y-axis that cannot fully represent both the MNPs and MMCs. Thus the hysteresis curve for the MNPs is shown in the inset with a magnetization that is three orders of magnitude lower than the original plot. 165 Table 5.4. Magnetization parameters for MMCs, in-house PEIcoated MNPs, and Dynabeads. label MMC MNP Dynabead Ms (emu/label) 2.49 × 10−12 1.93 × 10−15 1.75 × 10−11 Hc (Oe) 116 64 2 M @ H = 150 Oe (emu/label) 1.86 × 10−13 7.17 × 10−16 1.09 × 10−13 166 silica shell. With this analysis, the MMCs give an Ms of 2.49 × 10−12 emu/label due to the MNPs present within the MMCs layers, calculated at ~50-100 MNPs/MMC. The Ms value for the MNPs of 1.93 × 10−15 emu/label is four orders of magnitude lower than Dynabeads and three orders of magnitude lower than the MMCs. This decrease for the MNPs suggests that without enhancement such as the increase in quantity in the MMCs, the MNPs will produce a much lower magnetic signal per binding event in an immunoassay platform. For immunoassay readout, the magnetic field is typically 150 Oe to take advantage of the magnetoresistive nature of the sensor. At this field, the magnetization of the MMCs follows the same trend as with Ms values with an increase three orders of magnitude above that of the MNPs. More importantly, a comparison of the MMCs to the Dynabeads magnetic labels shows a magnetic signature with similar strengths per label. 5.3.5 MMCs as Immunoassay Labels Though the MMCs gave slightly higher magnetization values than Dynabeads in the VSM analysis, readout via the GMR sensor will allow for a utility assessment of MMCs as assay labels. For this experiment, MCs and MMCs were dropcast on the gold addresses of GMR coupons. Shown in Figure 5.9A is a depiction of the GMR coupon for two nickel addresses (Ni1 and Ni2) on either side of a gold address. The GMR readout in voltage is directly related to the changing resistance of the GMR sensor via Ohms law due to the magnetic field of the nickel address or MNPs. The voltage will decrease as the GMR sensor is scanned over a magnetic address (nickel addresses or gold addresses with MNPs) due to the magnetic field produced by the magnetic address in the opposing direction of the 167 (A) (B) 10 μm Figure 5.9. Schematic and exemplar signal of GMR readout. (A) Schematic of GMR coupon with nickel addresses (Ni1 and Ni2) for reference and gold address (Au) for assay architecture and GMR sensor with reference resistors (RR1 and RR2) and sense resistor (RS); (B) GMR trace from dropcast MMCs on gold addresses. 168 applied magnetic field. The magnetic signature for the gold address (MAu) can be normalized to the two neighboring nickel reference addresses (MNi1 and MNi2) using Equation 5.5 to give a GMR signal %: GMR signal % = MAu M +M Ni1 2 Ni2 (5.5) The GMR trace in Figure 5.9B, with an optical image of the gold address with MMCs, shows the magnetic signal (decrease in voltage) for the dropcast MMCs. This GMR trace can be quantified using Equation 5.5 to give 0.044% GMR signal/label, in comparison to the commercially available Dynabeads which give 0.0062% GMR signal/label. This is ~7× increase in GMR signal for the MMCs versus Dynabeads. The GMR signals can further be used to estimate LODs for each of the labels. Given instrument noise is ~ 0.010 ± 0.003 mV, we can estimate an LOD by the signal from the noise plus three times the standard deviation to 50 and 350 labels for the MMC and Dynabeads, respectively.44 For the MMC labels, this LOD corresponds to ~100 yM, or for a typical immunoglobulin protein with a molecular weight of 150,000 kDa, ~10−17 g or 10 ag. 5.4 Conclusions The success of a magnetic-based assay relies heavily on the magnetic label, both in the magnetic signal produced by a single label and the physical characteristics (e.g., size and stability) that maximize labeling and reduce settling of the particles, lowering nonspecific adsorption of the label on the substrate. The incorporation of many MNPs in a larger hollow construct such as a microcapsule allows for enhancement of magnetization per label while preserving the stability with regards to sedimentation of a small nanoparticle. Here, we report on the synthesis and characterization of magnetic 169 microcapsules (MMCs) for use as an immunoassay label in a giant magnetoresistive (GMR) system. With evidence for greater magnetization over the leading competitor, Dynabeads, we have shown the high utility of MMCs as immunoassay labels in a GMR system. Ultimately, this construct overcomes some of the pitfalls (i.e., stability with regards to sedimentation through a hollow capsule and low magnetization through the incorporation of numerous nanoparticles in a single label) we have encountered with commercially available magnetic nanoparticles that were developed for, and used in, technologies other than immunoassays. With a model to theoretically assess the sedimentation properties of the MMC structures, we anticipate MMCs can serve as a method to enhance the utility of immunoassays with GMR readout by tuning the increase in sedimentation with the increase the magnetization of the label (i.e., by incorporation of more MNPs within the layers). 5.5 References (1) Gaster, R., S.; Hall, D. A.; Wang, S. X. Magneto-Nanosensor Diagnostic Chips In Point-of-Care Diagnostics on a Chip, Issadore, D.; Westervelt, R. M., Eds. Springer: New York, 2013; pp 153-177. (2) Gu, H.; Zhang, X.; Wei, H.; Huang, Y.; Wei, S.; Guo, Z. Chem. Soc. Rev. 2013, 42, 5907-5943. (3) Rocha-Santos, T. A. P. Trends Anal. Chem. 2014, 62, 28-36. (4) Swierczewska, M.; Liu, G.; Lee, S.; Chen, X. Chem. Soc. Rev. 2012, 41, 2641-2655. (5) Daughton, J. M.; Bade, P. A.; Jenson, M. L.; Rahmati, M. M. M. IEEE Trans. Magn. 1992, 28, 2488-2493. (6) Daughton, J. M. J. Magn. Magn. Mater. 1999, 192, 334-342. (7) Weiss, R.; Mattheis, R.; Reiss, G. Meas. Sci. Technol. 2013, 24, 1-17. 170 (8) Hirota, E.; Sakakima, H.; Inomata, K. Giant Magneto-Resistance Devices. Springer: New York, 2002. (9) Barnaś, J.; Dugaev, V. K. Giant Magnetoresistance and Applications. In Magnetism of Surfaces, Interfaces, and Nanoscale Materials, Camley, R.; Celinski, R.; Stamps, R. L., Eds. Elsevier: Waltham, MA, 2016; pp 371-420. (10) Nordling, J.; Millen, R. L.; Bullen, H. A.; Porter, M. D.; Tondra, M.; Granger, M. C. Anal. Chem. 2008, 80, 7930-7939. (11) Millen, R. L.; Kawaguchi, T.; Granger, M. C.; Porter, M. D.; Tondra, M. Anal. Chem. 2005, 77, 6581-6587. (12) Millen, R. L.; Nordling, J.; Bullen, H. A.; Porter, M. D.; Tondra, M.; Granger, M. C. Anal. Chem. 2008, 80, 7940-7946. (13) Bird, R. B.; Stewart, W. E.; Lightfoot, E. N. Transport Phenomena. John Wiley & Sons, Inc.: New York, 2007. (14) Sukhorukov, G. B.; Donath, E.; Davis, S.; Lichtenfeld, H.; Caruso, F.; Popov, V. I.; Möhwald, H. Polym. Adv. Technol. 1998, 9, 759-767. (15) Sukhorukov, G. B.; Donath, E.; Lichtenfeld, H.; Knippel, E.; Knippel, M.; Budde, A.; Möhwald, H. Colloids Suf. 1998, 137, 253-266. (16) Decher, G. Science 1997, 277, 1232-1237. (17) Polysciences, Inc. . (2016). Polybead® microspheres. http://www.polysciences.com/default/catalog-products/microspheresparticles/polymer-microspheres/polybead-sup-r-sup-microspheres/polybead-sup-rsup-non-functionalized-microspheres. (18) Poly(allylamine hydrochloride); MSDS No. 283223[Online];. Sigma-Aldrich: February 11, 2016. (19) Hu, S.-H.; Tsai, C.-H.; Liao, C.-F.; Liu, D.- M.; Chen, S.-Y. Langmuir 2008, 24, 11811-11818. (20) Lu, Z.; Prouty, M. D.; Guo, Z.; Golub, V. O.; Kumar, C. S. S. R. Lvov, Y. M. Langmuir 2005, 21, 2042-2050. (21) Caruso, F.; Spasova, M.; Susha, A.; Giersig, M.; Caruso, R. A. Chem. Mater. 2001, 13, 109-116. (22) Park, J.; Porter, M. D.; Granger, M. C. Langmuir 2015, 31, 3537-3545. 171 (23) Yang, H.; Jiang, W.; Lu, Y. Mater. Lett. 2007, 61, 2789-2793. (24) García, M. T.; Gracia, I.; Duque, G.; de Lucas, A.; Rodríguez, J. F. Waste Management 2009, 29, 1814-1818. (25) Rife, J. C.; Miller, M. M.; Sheehan, P. E.; Tamanaha, C. R.; Tondra, M.; Whitman, L. J. Sens. Actuators, A 2003, 107, 209-218. (26) Hunter, R. J. Zeta Potential in Colloid Science: Principles and Applications. Academic Press, Inc.: San Diego, CA, 1981. (27) Donath, E.; Budde, A.; Knippel, E.; Bäumler, H. Langmuir 1996, 12, 4832-4839. (28) Donath, E.; Walther, D.; Shilov, V. N.; Knippel, E.; Budde, A.; Lowack, K.; Helm, C. A.; Möhwald, H. Langmuir 1997, 13, 5294-5305. (29) Caruso, F.; Lichtenfeld, H.; Donath, E.; Möhwald, H. Macromolecules 1999, 32, 2317-2328. (30) Edwards, H. G. M.; Brown, D. R.; Dale, J. A.; Plant, S. Vib. Spectrosc. 2000, 24, 213-224. (31) Edwards, H. G. M.; Brown, D. R.; Dale, J. A.; Plant, S. J. Molec. Struct. 2001, 595, 111-125. (32) Zucolotto, V.; Ferreira, M.; Cordeiro, M. R.; Constantino, C. J. L.; Balogh, D. T.; Zanatta, A. R.; Moreira, W. C.; Oliveira, O. N. Jr. J. Phys. Chem. B. 2003, 107, 3733-3737. (33) Gillespie, R. J.; Robinson, E. A. Canadian J. Chem. 1963, 41, 2074-2085. (34) Ege, S. Organic Chemistry: Structure and Reactivity. 5th ed.; Houghton Mifflin Company: Boston/New York, 2004. (35) Hariharan, R.; Biver, C.; Mays, J.; Russel, W. B. Macromolecules 1998, 31, 75067513. (36) Dreaden, E. C.; Alkilany, A. M.; Huang, X.; Murphy, C. J.; El-Sayed, M. A. Chem. Soc. Rev. 2012, 41, 2740-2779. (37) Crawford, A. C. Surface-enhanced Raman Scattering for the Reliable and Reproducible Detection of Disease Antigens in Biologically Relevant Media. University of Utah: Salt Lake City, UT, 2016. (38) Cho, E. C.; Zhang, Q.; Xia, Y. Nature Nanotechnology 2011, 6, 385-391. 172 (39) (2016). Streptavidin-coupled Dynabeads®. https://www.thermofisher.com/us/en/home/brands/productbrand/dynal/streptavidin-coupled-dynabeads.html. (40) Mason, M.; Weaver, W. Phys. Rev. 1924, 23, 412-426. (41) Hiemenz, P. C.; Rajagopalan, R. , Principles of Colloid and Surface Chemistry. 3rd ed.; Marcel Dekker, Inc.: New York, 1997. (42) Che, X.-D.; Bertram, H. N. J. Magn. Magn. Mater. 1992, 116, 121-127. (43) Morales, M.; Munoz-Aguado, M.; Garcia-Palacios, J.; Lazaro, F.; Serna, C. J. Magn. Magn. Mater. 1998, 183, 232-240. (44) Young, C. C.; Blackley, B. W.; Porter, M. D.; Granger, M. C. Anal. Chem. 2016, 88, 2015-2020. CHAPTER 6 CONCLUSION 6.1 Future Outlook Recent advancements in technology and instrumentation have driven the development of novel molecular diagnostic tests for the detection of disease markers or bioterrorism agents to ever increasing levels of performance. Still, translating this body of work from the research laboratory to a field-deployable point-of-need (PON) test remains a significant challenge with regards to time, cost, and portability. The development of a PON test becomes increasingly important in the detection of bioterrorism agents, for example, where the direct utility of the test is dependent on sensitive and rapid detection. For these situations, a PON test must be rapid to provide prompt results, low in cost to facilitate widespread use in a variety of non-laboratory settings, and easy to use to limit operator error. Retaining high degrees of specificity and sensitivity, though, has proven difficult when developing or adapting a test to meet these objectives. With the serious potential threat of biowarfare in today's world, a PON test would be an important tool to effectively limit the damage from a bioterrorism attack. This body of work has focused on addressing key aspects in the demand for PON tests, especially for the detection of bioterrorism agents in the fight against biowarfare, and includes (1) the development of an 174 immunoassay for the potential bioterrorism agent, botulinum neurotoxin (BoNT), (2) an investigation of the efficiency of covalent protein immobilization via a commonly used linking chemistry - N-hydroxysuccinimide (NHS) esters, and (3) the synthesis and characterization of magnetic microcapsules (MMCs) for eventual use as an immunoassay label. The development of a PON test for BoNTs, specifically, is critical due to the required timely detection in bioterrorism situations. The current testing method, the mouse bioassay, falls short as a PON test due to the required inoculation and care of live mice for up to 7 days. Thus, in Chapter 2, the development of a surface-enhanced Raman scattering (SERS)-based immunoassay for BoNT is presented. The detection limits of two BoNT serotypes, BoNT-A and BoNT-B, in the low- to sub-pM (pg/mL) range in buffer and human serum demonstrates the utility of these SERS-based immunoassays to effectively detect BoNTs at levels relevant to lethal toxicity. These results reflect detection of BoNTs in sample matrices that encompass the testing of a suspect powder sample that can be dispersed in a chosen solution for analysis (i.e., buffer) and the diagnosis of patients during a bioterrorism attack (i.e., human serum). In addition, with recent advancements in the development of portable Raman systems, the SERS-based immunoassay is expected to translate to a PON test for BoNT. To further develop these immunoassays as PON tests, we must also begin to investigate the underpinnings that may limit their sensitivity, specificity, robustness, and reproducibility. As such, Chapters 3 and 4 present results of an examination of the application of NHS chemistry that is commonly used for the immobilization of proteins in immunoassays. This chemistry is used to form amide linkages (i.e., aminolysis) between 175 the exposed amine groups of proteins and an underlying NHS-terminated monolayer. However, it is known that the competition of hydrolysis may inactivate the NHSterminated monolayer. Thus, in Chapter 3, the hydrolysis and aminolysis kinetic rate constants (𝑘𝑘ℎ and 𝑘𝑘𝑎𝑎 ) of an NHS-terminated monolayer, formed from the spontaneous adsorption of the linker molecule, dithiobis (succinimidyl propionate) (DSP), are determined via external reflection infrared spectroscopy (IR-ERS) by monitoring the decreasing intensity of the IR band from the NHS group. This chapter also includes a thorough characterization of the DSP-based monolayer using IR-ERS, X-ray photoelectron spectroscopy (XPS), and electrochemistry. From these studies, it is concluded that proteins are likely not covalently linked, as previously assumed, under the conditions typically used in an immunoassay. Instead, it is likely that the proteins are physically adsorbed to the hydrolyzed DSP-based monolayer, which may result in desorption in subsequent liquid application steps. Though this finding does not invalidate the detection strategies that have been developed using NHS linking chemistry, it does generate a potential challenge that with improvement, may lead to more sensitive, reproducible immunoassays. To this end, Chapter 4 furthers the investigation of NHS-based linking chemistry for proteins by studying two potential avenues to improve its effectiveness: (1) methylene chain length to investigate monolayer structure, and (2) solution composition to determine conditions at which hydrolysis may be suppressed. First, the hydrolysis and aminolysis of four NHS-containing disulfides with various methylene chain lengths (n) in borate buffer (pH 8.5) are investigated. Use of dithiobis succinimidyl undecanoate (DSU, where n = 10) results in a reduction in rate of hydrolysis by an order or magnitude when compared to that of DSP, where n = 2. Characterization of these monolayers via IR-ERS, XPS, 176 electrochemistry, and contact angles with determinations of the thermodynamic activation parameters obtained from kinetic studies at three different temperatures indicates that the decrease in the rate of hydrolysis for the DSU-based monolayer is due to increased steric hindrance, interchain interactions, and hydrophobicity. In addition, the hydrolysis of DSPbased monolayers in various solutions (i.e., borate buffer, phosphate buffer, and sodium hydroxide) was investigated. These experiments point to the potential use of sodium hydroxide or water to reduce hydrolysis, with little to no hydrolysis observed over 2 h. Overall, these results translate to nearly a 30× increase in amidization with DSUbased monolayers over DSP-based monolayers. Still, the absolute extent of aminolysis is low (i.e., 0.08%) when compared to hydrolysis, which points to the potentially more effective method of increasing amidization by use of different immobilization reaction conditions to hinder hydrolysis and promote aminolysis. Thus, with careful consideration of the conditions used for the protein immobilization, these studies will hopefully provide a guide to aid the development of a more robust capture antibody layer for future immunoassays which may improve sensitivity, specificity, and reproducibility. Lastly, this dissertation concludes with the synthesis of MMCs for use as a magnetic label in immunoassays utilizing giant magnetoresistance (GMR) readout. In addition to SERS, we anticipate that GMR-based immunoassays can be translated to a PON test through the rapid readout of GMR sensors and handheld instrumentation. One of the limitations with a GMR-based immunoassay lies in the challenge of finding a highly magnetic label that is stable in immunoassay conditions (i.e., undergoes little sedimentation) to facilitate high sensitivity while reducing sedimentation-based nonspecific binding. The use of MMCs as an immunoassay label aims to achieve high 177 magnetization through the incorporation of multiple magnetic nanoparticles (MNPs) within the capsule layers while maintaining neutral buoyancy through the "hollow" nature of the MMC construct (i.e., a density that is closely matched to that of the solution). MMC composition and structure is first characterized by zeta (ζ) potential, atomic force microscopy (AFM), and scanning electron microscopy (SEM). Then, the sedimentation properties are investigated with a sedimentation model using Stoke's law and the MasonWeaver equation. This model demonstrates the potential reduction in sedimentation of MMCs when compared to the commercially available magnetic particle, Dynabeads. Further, the magnetization of the MMCs, obtained via vibrating sample magnetometer (VSM) and a GMR sensor, establishes their possible use as an immunoassay label for sensitive, low-limit detection. In summary, this dissertation has contributed several valuable advancements towards the development of PON tests. With applications in biowarfare, the sensitivity of SERS-based detection was proven through the low-level detection of BoNTs. These SERSbased immunoassays are expected to translate to a PON test (i.e., rapid, field-deployable, and easy to use), therefore transforming the way we mitigate the effects of bioterrorism. In addition, the work examining binding chemistries and magnetic label fabrication is expected to further the in-depth examinations needed to improve our current methods for diagnostics and aid their translation to PON testing. APPENDIX SUPPORTING INFORMATION FOR CHAPTER 4 Experimental XPS spectra (Figure A.1 and Table A.1) were collected with a Kratos Axis Ultra DLD X-ray photoelectron spectrometer. The monochromatic Al X-ray source has an incidence angle of 60°, 15 kV anode potential, and 12 mA emission current. Other system parameters include a 700 by 300 μm hybrid slot size and a system pressure of 10−9 to 10−10 torr. High resolution scans were collected using a 40 eV pass energy, 0.1 eV step size, and 300 ms dwell time for Au(4f), C(1s), O(1s) regions or 1.2 s for N(1s) and S(2p) regions. Bands were analyzed with deconvoluted Gaussian-Lorentzian profiles and a linear or Shirley background subtraction.1 All binding energies were calibrated to the Au(4f7/5) band at 84.0 eV. Additional constraints, full widths at half-maximum of 0.9 to 1.3 and S(2p1/2):S(2p3/2) band areas of 2:1, were applied to the S(2p) bands in accordance with the spin-orbit splitting and a single S(2p) couplet.2 Contact angle measurements (Figure A.2) were made using a Dataphysics OCA15EC with deionized water at room temperature. Advancing (θa) and receding (θr) contact angles were made by increasing or decreasing the volume of the water droplet, respectively (n = 5). Experimental details for all other figures (Figures A.3-A.5) and tables (Tables A.2-A.3) are described in the text for chapter 4. 179 Figure A.1. XPS spectra of as-formed (a) DSP-, (b) DSH-, (c) DSO-, and (d) DSU-based adlayers on gold. All band intensities (counts per second - CPS) have been normalized to the Au (4f7/2) band. The residuals (not shown) from the deconvolution analysis for O(1s), S(2p), N(1s), and C(1s) were at most 1.2, 1.2, 0.9, and 2.0%, respectively. 180 Table A.1. XPS band assignments and positions for as-prepared DSP-, DSH-, DSO-, and DSU-based adlayers on gold. band position (eV)a DSP DSH DSO DSU O(1s) NHS ester oxygen 534.5 534.8 534.9 535.0 O(1s) carbonyl oxygen 532.1 532.4 532.4 532.5 C(1s) carbonyl carbon 288.7 289.0 289.1 289.2 C(1s) methylene carbon next to carbonyl carbon 285.4 285.4 285.4 285.5 C(1s) methylene carbonb 284.4 284.3 284.3 284.6 S(2p1/2) gold-bound thiolate 163.2 163.2 163.3 163.4 S(2p3/2) gold-bound thiolate 161.9 161.9 162.0 162.1 N(1s) succinimidyl nitrogen 401.7 401.9 402.0 402.2 a ± 0.1 eV in the S(2p) region and ± 0.5 eV in the O(1s), C(1s), and N(1s) regions b Assigned to methylene groups (both in alkyl chains and the NHS group) core level assignment 181 Figure A.2. Contact angle measurements of DSP- (n = 2), DSH- (n = 5), DSO- (n = 7), and DSU-based (n = 10) adlayers on gold. Advancing (θa) and receding (θr) contact angles were proved with deionized water and used to calculate the hysteresis (H) by H = θa-θr. 182 Figure A.3. IR-ERS spectra of DSP- (black), DSH- (red), DSO(blue), and DSU-based (green) adlayers on gold shown in the C−H region from 3000 cm−1 to 2800 cm−1 before and after reaction (hydrolysis or aminolysis). 183 Figure A.4. IR-ERS spectra of DSP- (black) and DSU-based (green) monolayers after immersion in 50 MM BB, pH 8.50, for 5 minute intervals. 184 Figure A.5. Plot of extent of reaction (x) versus time with fit to 5-parameter sigmoid function. The 2nd derivative, x", (solid red) gives the inflection point (dotted red) when set to 0 while the 3rd derivative, x"', (solid blue) gives the transition points (dotted blue) when set to 0. This serves as a guide to determine the three reaction regions: initial, bulk transformation, and final. Table A.2. IR spectral peak positions and band assignments for reaction products of DSP-, DSH-, DSO-, and DSU-based adlayers on gold after 1 h immersion in hydrolysis (50 mM BB, pH 8.50) and aminolysis (500 mM EA in NaOH, pH 8.50) solutions. 185 186 Table A.3. IR-ERS band positions of C-H stretches in high energy region (3000 to 2800 cm-1) for DSP-, DSH-, DSO-, and DSU-based adlayers before (as-is) and after hydrolysis (hydro-) or aminolysis (amino-) by immersion in borate buffer (50 mM, pH 8.50) or ethylamine (500 mM, pH 8.5), respectively. Adlayer DSP DSH DSO DSU as-is hydroaminoas-is hydroaminoas-is hydroaminoas-is hydroamino- band positions (cm−1) ν(CH2) νa(CH2) νs(CH2) 2962 ± 1 2928 ± 1 2858 ± 1 2959 ± 3 2926 ± 2 2857 ± 2 2964 ± 3 2928 ± 1 2857 ± 1 2960 ± 1 2928 ± 1 2857 ± 0 2965 ± 1 2927 ± 0 2857 ± 1 2965 ± 1 2926 ± 2 2855 ± 1 2954 ± 1 2926 ± 1 2855 ± 2 2959 ± 6 2927 ± 0 2857 ± 1 2971 ± 4 2926 ± 1 2856 ± 1 2961 ± 2 2925 ± 3 2855 ± 2 2923 ± 1 2858 ± 2 2965 ± 2 2924 ± 6 2855 ± 2 187 Student t Values for Activation Parameters The t value for the Student's t test is calculated, as shown in Equation A.1: 𝑡𝑡 = 𝑥𝑥1 − 𝑥𝑥2 (A.1) 𝜎𝜎 2 𝜎𝜎 2 1 + 2 𝑁𝑁1 𝑁𝑁2 𝑥𝑥1 and 𝑥𝑥2 are the means of populations 1 and 2, respectively; 𝜎𝜎1 and 𝜎𝜎2 are the where standard deviations of populations 1 and 2, respectively; and 𝑁𝑁1 and 𝑁𝑁2 are the number of measurements in populations 1 and 2, respectively.3 The t test analysis, which determines the difference (if any) between two values to a given significance level, is summarized in Table A.4. References (1) Shirley, D. A. Phys. Rev. B 1972, 5, 4709-4714. (2) Moulder, J. F., Handbook of X-ray Photoelectron Spectroscopy. Physical Electronics, Inc.: Eden Prairie, MN, USA, 1992. (3) Barlow, R. J., Statistics: A Guide to the Use of Statistical Methods in the Physical Sciences. John Wiley & Sons: New York, 1989. Table A.4. Student t values for thermodynamic activation parameters for the hydrolysis of DSP- and DSU-based monolayers. The calculated t values are compared to a table of critical values (e.g., for n = 16, t = 0.865, 1.337, 1.746, 2.120, 2.583, and 2.921 for confidence levels of 60%, 80%, 90%, 95%, 98%, and 99%, respectively) to determine to what significance level (i.e., a confidence level of 99% would give a significance level of 1%) the two values are different. 𝑨𝑨 (M−1 s−1) 2.9 ± 0.3 × 107 DSP3.0 ± 1.5 × 107 DSU-* t value -* Significance level *Not significantly different 𝑬𝑬𝑨𝑨 (kJ mol−1) 24 ± 4 34 ± 6 4.16 1% 𝚫𝚫𝑺𝑺‡ (J mol−1 K−1) −111 ± 13 −111 ± 18 -* -* 𝚫𝚫𝑯𝑯‡ (kJ mol−1) 21 ± 4 31 ± 6 4.16 1% 𝚫𝚫𝑮𝑮‡ @T = 298 K (kJ mol−1) 54 ± 12 64 ± 16 1.5 20%