ABSTRACT Title of dissertation: MOLECULARLY IMPRINTED POLYMERS FOR THE SELECTIVE RECOGNITION OF PROTEINS Daniel S. Janiak, Doctor of Philosophy, 2009 Dissertation directed by: Professor Peter Ko nas Fischell Department of Bioengineering Molecular imprinting is a technique used to synthesize polymers that display selective recognition for a given template molecule of interest. In this study, the role of hydrogel electrostatic charge density on the recog- nition properties of protein-imprinted hydrogels was explored. Using 3- methacrylamidopropyl trimethylammonium chloride (MAPTAC) as a pos- itively charged monomer and 2-acrylamido-2-methylpropane sulfonic acid (AMPS) as a negatively charged monomer, a number of acrylamide-based polyelectrolyte hydrogels with varying positive and negative charge den- sities were prepared. The imprinted hydrogels were synthesized in the presence of the target molecule bovine hemoglobin (Bhb). The ability of the hydrogels to selectively recognize Bhb was examined using a compet- itive template molecule, cytochrome c. The Bhb imprinted gels exhibited template recognition properties that were dependent on both the monomer charge density and on whether the chosen monomer carried a positive or negative charge. In addition to polyelectrolye hydrogels, polyampholyte hydrogels con- taining both positively and negatively charged monomers were also synthe- sized. The simultaneous presence of two oppositely charged monomers in the pre-polymerization mixture resulted in imprinted hydrogels with cavi- ties that contain highly speci c functional group orientation. The polyam- pholyte hydrogels exhibited decreased swelling when compared to their polyelectrolyte counterparts, due to the shielding of repulsive interactions between oppositely charge monomers. This decreased swelling resulted in greater template recognition, but lower selectivity, when compared to their polyelectrolyte counterparts. In addition, we found that common agents used in template extrac- tion may be responsible for the speci c and selective binding properties exhibited by molecularly imprinted polymers in many published studies, and the e ect of variations of the template extraction protocol on the MIP recognition properties were also studied in depth. MOLECULARLY IMPRINTED POLYMERS FOR THE SELECTIVE RECOGNITION OF PROTEINS by Daniel S. Janiak Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park in partial ful llment of the requirements for the degree of Doctor of Philosophy 2009 Advisory Committee: Professor Peter Ko nas, Chair/Advisor Professor Mohamad Al-Sheikhly Professor William E. Bentley Professor Robert M. Briber Professor Lourdes G. Salamanca-Riba c Copyright by Daniel S. Janiak 2009 Dedication To my parents. I love you and owe you everything. ii Acknowledgments First, I would like to thank my advisor, Dr. Peter Ko nas, for provid- ing me with unconditional support, encouragement, and advice. I would also like to thank my committee, Dr. Mohamad Al-Sheikhly, Dr. William Bentley, Dr. Robert Briber, and Dr. Lourdes Salamanca-Riba for their insightful comments and suggestions. I owe a great deal of thanks to Omar Ayyub, an undergraduate research assistant in the Fischell Department of Bioengineering, who made an enormous contribution to this work. Without the love and support of my family, this work would not have been possible. I would also like to thank all the members of my lab, especially Linden Bolisay, Von Cresce, Ayan Ghosh, and Ta-I Yang. I am indebted to you all. iii Table of Contents List of Tables vi List of Figures vii 1 Introduction 1 1.1 Signi cance . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1.1 Background . . . . . . . . . . . . . . . . . . . . . . 3 1.1.2 The Epitope Approach . . . . . . . . . . . . . . . . 4 1.1.3 Surface Imprinting . . . . . . . . . . . . . . . . . . 5 1.1.4 Recognition Mechanisms . . . . . . . . . . . . . . . 9 1.1.5 Microenviroment . . . . . . . . . . . . . . . . . . . 11 1.1.6 Composition . . . . . . . . . . . . . . . . . . . . . . 12 1.1.7 Template Removal . . . . . . . . . . . . . . . . . . 13 1.1.8 Electrostatic Charge Density . . . . . . . . . . . . . 14 2 Protein Imprinted Polyelectrolyte Hydrogels 17 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . 19 2.2.1 Materials . . . . . . . . . . . . . . . . . . . . . . . 19 2.2.2 Synthesis . . . . . . . . . . . . . . . . . . . . . . . 19 2.2.3 Bhb Template Extraction . . . . . . . . . . . . . . 20 2.2.4 Template Rebinding . . . . . . . . . . . . . . . . . 23 2.2.5 UV-Vis Analysis . . . . . . . . . . . . . . . . . . . 23 2.2.6 Hydrogel Swelling . . . . . . . . . . . . . . . . . . . 24 2.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . 26 2.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3 Protein Imprinted Polyampholyte Hydrogels 50 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 50 3.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . 51 3.2.1 Synthesis . . . . . . . . . . . . . . . . . . . . . . . 51 3.2.2 Bhb Template Extraction . . . . . . . . . . . . . . 52 3.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . 54 3.3.1 Positively Charged Polyampholytes . . . . . . . . . 58 3.3.2 Negatively Charged Polyampholytes . . . . . . . . . 58 3.3.3 Neutral Polyampholytes . . . . . . . . . . . . . . . 61 3.3.4 Selectivity . . . . . . . . . . . . . . . . . . . . . . . 62 3.3.5 Swelling . . . . . . . . . . . . . . . . . . . . . . . . 64 3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4 Future Work 67 4.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 iv Bibliography 69 v List of Tables 2.1 Synthesis parameters for neutral, positively (MAPTAC) and negatively charged (AMPS) hydrogels. All values are in mg. 21 2.2 Summary of re-binding results for uncharged Bhb-imprinted and non-imprinted hydrogels. . . . . . . . . . . . . . . . . 31 2.3 Results of Bhb re-binding experiments performed on nega- tively charged (AMPS containing) Bhb-imprinted and non- imprinted hydrogels of varying charge density. Data repre- sents the amount of Bhb bound within each hydrogel. All values are in mg. . . . . . . . . . . . . . . . . . . . . . . . 32 2.4 Results of Bhb re-binding experiments performed on pos- itively charged (MAPTAC containing) Bhb-imprinted and non-imprinted hydrogels of varying charge density. Data represents the amount of Bhb bound within each hydrogel. All values are in milligrams. . . . . . . . . . . . . . . . . . 35 2.5 Results of selectivity experiments performed using cytochrome C template on Bhb-imprinted and non-imprinted hydrogels of varying charge density. This data represents the amount of cytochrome C extracted from each gel in milligrams. . . 37 2.6 Results of Bhb re-binding experiments performed on nega- tively charged (AMPS containing) Bhb-imprinted and non- imprinted hydrogels of varying charge density washed using the modi ed wash protocol. Data represents the amount of Bhb bound within each hydrogel. All values are in mg. . . 39 2.7 Results of Bhb re-binding experiments performed on neg- atively charged (MAPTAC containing) Bhb-imprinted and non-imprinted hydrogels of varying charge density washed using the modi ed wash protocol. Data represents the amount of Bhb bound within each hydrogel. All values are in mg. . 45 2.8 Selectivity factor ( ) for gels containing MAPTAC monomer washed under the modi ed protocol. . . . . . . . . . . . . 47 3.1 Synthesis parameters for neutral, positively (MAPTAC) and negatively charged (AMPS) hydrogels. All values are in mg. 52 3.2 Selectivity factor ( ) for polyampholyte gels washed under the modi ed protocol. . . . . . . . . . . . . . . . . . . . . 64 vi List of Figures 1.1 Schematic of the molecular imprinting process. . . . . . . . 2 1.2 Schematic of the epitope approach utilized for molecular imprinting of proteins and peptides. (a) Target molecule (b) Epitope (Imprinted) portion of target molecule (c) Im- printed polymer before template removal (d) Imprinted poly- mer after template removal, with cavities that display shape which is complementary to the imprinted portion (epitope) of the target molecule. (e) MIP subsequent to template re- binding. The MIP selectively recognizes the imprtined (ep- tiope) portion of the target molecule, thus, the entire target molecules is successfully bound . . . . . . . . . . . . . . . 6 1.3 Schematic of the hierarchical approach to molecular imprint- ing. (a) The target peptide or protein, bound to a sacri cial substrate. (b) Cross-linking and functional monomers. (c) Functional monomers associated with the target peptide or protein (d) MIP subsequent to removal of the sacri cial sub- strate and removal of the template molecules. . . . . . . . 8 2.1 UV-Vis scan of Bhb in water at various concentrations. The intensity of the peaks decreases as the concentration of Bhb is lowered. . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.2 UV-Vis scan of Bhb in SDS-HOAc at various concentrations. The intensity of the peaks decreases as the concentration of Bhb is lowered. . . . . . . . . . . . . . . . . . . . . . . . . 25 2.3 Calibration curve for Bhb in water. . . . . . . . . . . . . . 25 2.4 Wash pro le for an uncharged (0% ionic groups) Bhb-imprinted hydrogel. Washes (1-5) were performed using deionized wa- ter, while washes (6-10) were performed using SDS-HOAc solution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.5 Bhb re-binding experiment performed on an uncharged (0% ionic groups) Bhb-imprinted polymer hydrogel. Washes (1- 5) were performed using deionized water, while washes (6- 10) were performed using SDS-HOAc solution. . . . . . . . 29 vii 2.6 Binding experiment performed on a uncharged (0% ionic groups) non-imprinted hydrogel. Washes (1-5) were per- formed using deionized water, while washes (6-10) were per- formed using SDS-HOAc solution. . . . . . . . . . . . . . . 30 2.7 Swelling data and imprinting factor (IF) for gels with vari- ous amounts of negatively charged AMPS monomer. The columns represent the swelling data, while the solid and dashed lines represent the IF of gels washed under the orig- inal and modi ed protocol, respectively. . . . . . . . . . . . 42 2.8 Swelling data and imprinting factor (IF) for gels with vari- ous amounts of positively charged MAPTAC monomer. The columns represent the swelling data, while the solid and dashed lines represent the IF of gels washed under the orig- inal and modi ed protocol, respectively. The peak in the dashed line at 0.25% MAPTAC represents the optimum monomer concentration for binding Bhb. . . . . . . . . . . 46 3.1 Binding experiment performed on a net neutral (50% AMPS - 50% MAPTAC), non-imprinted polyampholyte hydrogel washed with a 10% HOAc solution. Washes (1-5) were per- formed using deionized water. . . . . . . . . . . . . . . . . 55 3.2 Binding experiment performed on a net neutral (50% AMPS - 50% MAPTAC), non-imprinted hydrogel washed with a 10% SDS solution. Washes (1-5) were performed using deion- ized water. . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 3.3 Binding experiment performed on a net neutral (50% AMPS - 50% MAPTAC), non-imprinted hydrogel washed with a solution containing 10% SDS and 10% HOAc. Washes (1-5) were performed using deionized water. . . . . . . . . . . . 57 3.4 Bhb template rebinding data for positively charged, Bhb imprinted polyampholyte hydrogels containing 25% AMPS and 75% MAPTAC . . . . . . . . . . . . . . . . . . . . . . 59 3.5 Bhb template rebinding data for positively charged, non- imprinted polyampholyte hydrogels containing 25% AMPS and 75% MAPTAC. . . . . . . . . . . . . . . . . . . . . . 59 3.6 Bhb template rebinding data for negatively charged, Bhb imprinted polyampholyte hydrogels containing 25% MAP- TAC and 75% AMPS. . . . . . . . . . . . . . . . . . . . . 60 viii 3.7 Bhb template rebinding data for negatively charged, non- imprinted polyampholyte hydrogels containing 25% MAP- TAC and 75% AMPS. . . . . . . . . . . . . . . . . . . . . 61 3.8 Bhb template rebinding data for net neutral, Bhb imprinted polyampholyte hydrogels containing 50% MAPTAC and 50% AMPS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 3.9 Bhb template rebinding data for net neutral, non-imprinted polyampholyte hydrogels containing 50% MAPTAC and 50% AMPS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 3.10 Swelling ratio (SR) for polyampholyte hydrogels. . . . . . . 65 ix Chapter 1 Introduction 1.1 Signi cance Molecularly imprinted polymers (MIPs) [71, 67, 76, 64, 50] are syn- thetic materials produced by the cross-linking of functional monomers or polymers in the presence of a template molecule. The template is sub- sequently removed, leaving cavities possessing size, shape, and functional group orientation which are complementary to the target molecule. The size and shape of the cavity allow the target molecule or similar molecules to occupy the cavity space, while the functional group orientation within the cavity will preferentially bind in speci c locations to only the target molecule and not to similar molecules. Figure 1.1 provides a schematic of the molecular imprinting process. While MIPs have been prepared for a large number of target molecules and applications [27, 62, 11, 61] over the years, the majority of the template molecules studied have been character- ized by their low molecular weight and insolubility in aqueous systems. Much of the pioneering work in the molecular imprinting of pep- tides and proteins was conducted by Mosbach [3, 38, 63, 4, 2] using amino acid derivatives as template molecules to create stationary phases for chro- matographic applications. The MIPs synthesized in these early studies 1 Figure 1.1: Schematic of the molecular imprinting process. were tested for their ability to selectively recognize their respective target molecules in organic solvents. While these experiments laid the ground- work for the synthesis of arti cial receptors capable of recognizing peptides and proteins, most molecules of biological importance are water soluble, and many natural recognition events such as antigen-antibody binding oc- cur in aqueous media. Therefore, there exists a strong need to create ar- ti cial receptors which are capable of recognizing peptides and proteins in aqueous media in order to create materials and devices capable of mimick- ing natural processes. For instance, MIPs which can selectively recognize speci c proteins or peptide sequences in aqueous media have the potential to be used as substrates in medical diagnostic applications. In addition, aqueous MIPs could be used as the solid phase for chromatographic analy- sis and puri cation of peptides and proteins. Finally, the synthesis of MIPs which exhibit high a nity, selectivity, and binding capacity may provide 2 researchers with a low-cost, easily obtainable method for studying the fun- damental interactions which occur during biological recognition processes. The goal of this research is to understand the role of electrostatic charge density on the recognition and selectivity properties of protein-imprinted hydrogels. The results of this project will aid in the development of ra- tionally designed MIPs with higher speci city and selectivity than those currently available. 1.1.1 Background There are a number of di erent strategies for creating polymeric re- ceptors targeting peptides and proteins. Functional groups forming strong template interactions with the target molecules are commonly used. Exam- ples of this include electrostatic and metal-chelating groups [47, 26]. In ad- dition, shape interactions between the template molecule and the synthetic receptor can be exploited. Weak interactions such as hydrophobic and hy- drogen bonding have also been employed in numerous cases [56, 10, 51]. Some notable recent accomplishments are the detection of peptides using a molecularly imprinted piezoelectric sensor [44], the synthesis of MIPs in the presence of a cloned bacterial protein which are capable of selectively recognizing the authentic protein when incubated in cell extract [79], the development of MIPs with enzyme-like properties [17], and the synthesis of MIPs in the presence of proteins within chromatography columns provid- 3 ing a monolithic bed for the selective recognition of the imprinted protein [60]. The following sections outline a number of the approaches used to create MIPs capable of recognizing biological macromolecules. 1.1.2 The Epitope Approach The process of molecular imprinting of macromolecules such as pep- tides and proteins is complicated by a number of factors. Well de ned recognition sites are more easily formed by the imprinting of small molecules with rigid structures, therefore imprinting small sequences of amino acids (dipeptides, tripeptides, etc.) should be fairly straightforward. Proteins, however, are inherently capable of assuming a large number of confor- mations depending on a number of factors, including but not limited to, solution temperature, pH and ionic strength. Steric factors also make molecular recognition of proteins di cult, as it is extremely di cult for their large structures to move freely through highly cross-linked polymer networks. One proposed method of reducing the complications associated with the imprinting of proteins and peptides has been termed the epitope approach [57, 58, 52]to molecular imprinting. Traditionally, an epitope refers to the small active site located within the larger protein structure on an antigen, which combines with the antigen-binding site on an antibody or lymphocyte receptor [1]. Figure 1.2 shows a schematic of the epitope approach to molecular imprinting. In this technique, a small sequence of 4 amino acids from the larger protein target molecule is used to create the imprint. When a protein containing this speci c amino acid sequence is allowed to incubate in the presence of the MIP, the entire protein can thus be recognized and bound. The epitope approach has been successfully used to bind oxytocin by imprinting the Tyr-Pro-Leu-Gly amino acid sequence [57]. In this study, synthesis of the MIP was performed in an organic en- vironment, but subsequent rebinding experiments were performed using chromatographic methods in both aqueous-rich and aqueous-poor mobile phases. In the aqueous-poor mobile phase, hydrogen bonds and ionic in- teractions are the dominating factor in creating selective recognition sites. In the aqueous rich phase, ionic and hydrophobic interactions provide the dominant binding interaction. 1.1.3 Surface Imprinting Con ning the recognition sites of MIPs to surfaces is another tech- nique which has been developed to circumvent the steric problems associ- ated with the imprinting of biological macromolecules in aqueous systems. In a classic example of surface imprinting [65], protein imprinted surfaces were created by rst adsorbing a layer of protein onto a mica substrate, followed by the deposition of disaccharide molecules onto the protein layer. Upon dehydration, hydrogen bonds were formed between hydroxyl groups on the sugars and functional groups on the protein molecules. Radio- 5 Figure 1.2: Schematic of the epitope approach utilized for molecular im- printing of proteins and peptides. (a) Target molecule (b) Epitope (Im- printed) portion of target molecule (c) Imprinted polymer before template removal (d) Imprinted polymer after template removal, with cavities that display shape which is complementary to the imprinted portion (epitope) of the target molecule. (e) MIP subsequent to template re-binding. The MIP selectively recognizes the imprtined (eptiope) portion of the target molecule, thus, the entire target molecules is successfully bound 6 frequency (RF) glow discharge plasma, a common plasma deposition tech- nique used to create controlled thin lms of uoropolymers, was then used to deposit hexa uoropropylene onto the protein-dissacharide complexes. Upon removal of the mica substrate and target protein, the hydroxyl groups of the disaccharide molecules which were spatially complementary to the templated proteins were available as binding sites. In another example of MIPs in which the recognition sites have been con ned to the surface, Sell- ergren used hierarchical imprinting in an attempt to create well structured MIPs with homogeneous morphology. The template molecule is attached to the surface of a porous solid. Polymerization occurs within the pores of a silica bead and subsequent to polymerization, the silica is etched away leaving a porous structure with sites complementary to the structure of the template molecule. The mobile phase in these experiments consisted of a combination of water, acetontrile, and acetic acid. Solid phase synthesis products were also used directly as a template [72]. In one case, the surface of microtiter plates was coated with 3-aminophenylboronic acid and poly- merized in the presence of various protein templates. All plates showed se- lectivity for their respective templates [10]. Molecularly imprinted polymer layers were formed around silica beads in the presence of glucose oxidase under conditions which support electrostatic interaction between monomer precursors and the template molecules [48]. 7 Figure 1.3: Schematic of the hierarchical approach to molecular imprint- ing. (a) The target peptide or protein, bound to a sacri cial substrate. (b) Cross-linking and functional monomers. (c) Functional monomers associ- ated with the target peptide or protein (d) MIP subsequent to removal of the sacri cial substrate and removal of the template molecules. 8 1.1.4 Recognition Mechanisms In addition to overcoming the steric and conformational problems associated with molecular imprinting in aqueous media, the exact mech- anism by which recognition occurs has also proved di cult to determine. The addition of functional monomers capable of participating in electro- static interactions with the templated protein was shown, in early stud- ies, to weaken the adsorption characteristics of the MIP, suggesting that electrostatic interactions did not play a large role in the recognition of peptides and proteins within this particular system [42]. On the other hand, molecularly imprinted polymers synthesized for the selective recog- nition of N-terminal histidine peptides in aqueous solutions most certainly exploit template-receptor complexation through electrostatic interaction [26]. There is strong evidence to suggest that the cooperative interaction between hydrogen bond formation and hydrophobic interaction is the most dominant form of template-receptor complexation in aqueous MIPs. Hy- drogen bonds are highly directional, and many arti cial receptors rely on the high directionality of hydrogen bonds and shape speci city of target molecules to create receptors capable of strong interaction and molecular recognition [65, 5]. The role of hydrophobic and hydrogen bonding e ects in the recognition of peptides has been studied by chromatographic anal- ysis of MIPs imprinted with various peptide sequences in aqueous-organic mixtures [51]. It was observed that the separation factors of these poly- 9 mers were signi cantly lower in acetonitrile-water mixtures than in pure acetonitrile. In addition, it was shown that water signi cantly lowers the e ectiveness of hydrogen bonds between the template and the receptor, as the increased water content weakens the hydrogen bonding interaction between the template and the receptor [57, 58, 39]. Hydrogen bonds are a major factor in the formation of strong template-receptor interactions in non-polar media [51]. Other studies on MIP hydrogels suggest that a combination of hydrogen bonding and electrostatic e ects allow for the selective recognition of target molecules in aqueous media [14]. It has been proposed that binding in aqueous media can be enhanced by exploit- ing cooperative interaction e ects and selecting proper hydrophobic mi- croenvirmonents to create better receptors [5]. In one of the most striking examples of molecular recognition by the cooperative action of hydropho- bic and electrostatic interactions, -cyclodrextin, a cyclic oligosaccharide was acrylated and co-polymerized with an electrostatic functional poly- mer, 2-acryoylamido-2,2-dimethylpropane sulfonic acid, in the presence of D-phenylalanine as the template molecule. -cyclodrextin has a hydropho- bic core which can bind aromatics groups on the side chains of peptides and protein molecules. While -cyclodrextin has an inherent selectivity for L-phenylalanine, this natural selectivity is reversed upon imprinting with D-phenylalanine. This MIP was tested in mixtures of acetonitrile and water at various ratios. As is expected, the hydrophobic e ect is the dominant factor in imprinting in aqueous-rich mixtures while electrostatic 10 e ects dominate template-receptor complexation in aqueous-poor media [55, 56]. 1.1.5 Microenviroment In addition to the number of interactions which are possible between the template and the MIP receptor, the microenviroment surrounding the binding site can have a large role in determining how e ective the MIP will be in recognizing its target molecule. For this reason, it is extremely di - cult to make a quantitative comparison between MIPs that have been syn- thesized and tested under a number of di erent conditions. Ionic strength e ects were studied in a system in which ionic interactions were the dom- inant force associated with template-receptor-recognition. A number of bu er systems were used to observe the e ect of ionic strength and bu er composition on the binding capacity of the MIPs. All bu ers examined re- sulted in a decrease in binding capacity of the MIP hydrogels. Interestingly, the absence of a bu er, with a 60 mM KCl solution showed the largest in- hibition of MIP binding capacity. It was therefore concluded that the ionic strength of the binding solution, not the bu er composition itself, plays a large role in determining the e ectiveness of the MIP [26]. The inhibi- tion of MIP binding capacity in these experiments is most likely the result of a change in polymer conformation and/or variations in the microenvi- roment surrounding the template binding sites caused by changes in salt 11 concentration. For example charge screening e ects may e ectively block the ionic interaction between the template at high salt concentration, as the salt dissociates into highly mobile anions and cations which can shield sites on both the target and the template. The e ect of bu er composition and ionic strength has not been studied in other systems; however, it is highly possible that these two variables will have a large e ect on MIP systems, even those which do not use ionic interactions as their primary recognition mechanism. This is because proteins can adopt a wide number of conformations depending on their environment. 1.1.6 Composition The structure and composition of MIPs used for aqueous imprinting can also have a large e ect on its ability to recognize its target molecule. Re-binding experiments conducted with MIP hydrogels imprinted in the presence of bovine serum albumin (BSA) were shown to have a strong de- pendence on the initial BSA concentration used during synthesis. This is most likely related to the fact that high template concentration dur- ing synthesis will result in a larger number of available binding sites upon subsequent imprinting experiments. Studies of lysozyme imprinted sil- ica beads revealed that the amount of lysozyme adsorbed onto the beads depended on the composition of functional monomers used during prepa- ration of the particles. This dependence suggests that there exists some 12 optimum distance between charged groups on the synthetic receptor which will correspond to a maximum in the complexation between receptor and template. The zeta-potential of the lysozyme molecules and the imprinted silica beads was also studied, and results revealed a coincidence between the zeta-potential of the template and that of the imprinted polymer beads at the value where the maximum amount of speci cally adsorbed lysozyme was observed. This yields the possibility that zeta-potential matching may play an important role in the design of synthetic receptors [33]. Cross- linking concentration is also a critical factor in creating synthetic receptors with high a nity for their target molecules [26, 73]. 1.1.7 Template Removal Although commonly overlooked, the removal of target molecules from MIPs subsequent to re-binding experiments is a critical factor in the im- printing capability of the MIP. While very little work has focused solely on the study of template removal strategies, the methods which have been studied have revealed a number of important details. Template re- moval from polyacrylamide hydrogels synthesized in the presence of bovine hemoglobin was achieved with an aqueous solution consisting of varying ratios of sodium dodecyl sulfate (SDS) and acetic acid (HOAc) [30]. A qualitative model for the interaction between the SDS-HOAc solution and target protein within the cavity is presented and an optimum SDS/HOAc 13 ratio of 10% (weight/volume) SDS and 10% (volume/volume) HOAc is discovered. Interestingly, it was found that while increasing the amount of SDS and HOAC in the wash solution improves template removal, the re-binding e ect is clearly compromised. In another MIP system, a solu- tion consisting of acetic acid and Tween 20 was used prior to conducting template re-binding experiments [10]. It is quite possible that the combi- nation of an acidic solution with a detergent is the most e ective method of template removal for MIPs imprinted with peptide or proteins. Another obvious choice for template removal in these systems would be the use of proteolytic enzymes (proteases) which are capable of cleaving peptide bonds and subsequently destroying the structure of the imprinted protein or peptide [46]. The peptide fragments could then be removed from the cavity by simply rinsing with bu er solution. The protease trypsin, for example, was investigated as a template removal agent. However, after washing of an MIP with a trypsin solution, the re-binding e ciency suf- fered as cleaved protein fragments blocked binding sites within the imprint cavity [30]. 1.1.8 Electrostatic Charge Density Relatively few studies have been conducted thus far to explore the role of electrostatic interactions in molecularly imprinted polymers. Hjerten et al. [42] have explored the role of weakly ionizable acidic groups on the 14 recognition properties of imprinted polymers with human hemoglobin as the template, and found that the adsorption characteristics of the MIPs are weakened when charged groups were introduced into the polymer ma- trix. Zheng et al. [34] synthesized amphoteric, acrylamide based imprinted polymers with bovine serum albumin (BSA) and chicken egg lysozyme as templates. Using high-pressure liquid chromatography (HPLC), they con- cluded that the cooperative e ects of multiple oppositely charged func- tional moieties enhanced the binding properties of the MIPs. However, they did not include data that compares the recognition properties of the charged polymers to an uncharged imprinted counterpart. Kameoka et al. [33] synthesized imprinted polymer particles composed of acrylamide and either acrylic acid or N,N-dimethylaminopropylacrylamide as co-monomers for lysozyme recognition. They found that peak lysozyme recognition occurred at speci c molar amounts of charged (acrylic acid) functional groups. The overall goal of this research is to understand the role that charge density plays in in uencing the recognition properties of protein-imprinted hydrogels. Protein-imprinted hydrogels were synthesized at various charge densities, and the recognition properties of these hydrogels were measured using batch template rebinding experiments. The hydrogels were tested for their selectivity towards the template using a competitive protein, Cy- tochrome c. The rst goal of this dissertation was to study the e ects of charge density on the recognition properties of polyelectrolyte protein- 15 imprinted hydrogels. The hydrogel charge density was varied by intro- ducing varying amounts of positively or negatively charged monomers into neutral hydrogels. The second goal was to study the e ect of e ects of charge density on the recognition properties of hydrogels containing a com- bination of positively and negatively charged monomers (polyampholyte hydrogels). The third goal was to determine the e ect that the template removal process had on the recognition properties of the protein-imprinted hydrogels. 16 Chapter 2 Protein Imprinted Polyelectrolyte Hydrogels 2.1 Introduction The goal of this research is to study how charge density variations within protein imprinted polymeric hydrogels e ect their ability to rec- ognize bovine hemoglobin (Bhb) template molecules. To accomplish this goal, the charge density of Bhb imprinted hydrogels is systematically var- ied by incorporating small amounts of negatively and positively charged monomers to study its e ect on template recognition, selectivity and speci- city. The cationic monomer used in the synthesis of the MIP gels in this study was 3-methacrylamidopropyl trimethylammonium chloride (MAP- TAC), while the anionic monomer was 2-acrylamido-2-methylpropane sul- fonic acid (AMPS). The template, Bhb, is an oxygen transport protein with a molecular weight of 64,500 g/mol and an isoelectric point (pI) of approximately 6.8. The absorption properties of Bhb-imprinted and non- imprinted hydrogels were measured using batch re-binding experiments and selectivity experiments were performed using bovine cytochrome C as a competitive template. Cytochrome C was used to determine whether or not the gels synthesized in this study are capable of selectively recog- nizing Bhb on the basis of both molecular weight and isoelectric point. 17 Cytochrome C has a lower molecular weight (12,384 g/mol) and higher isoelectric point (pI = 10-10.5) than Bhb. As the template re-binding experiments were being conducted, it be- came evident that the wash used to extract protein from the imprinted hydrogels, which consisted of sodium dodecyl sulfate (SDS) and acetic acid (HOAc), was partly responsible for the speci c recognition properties dis- played by the hydrogels. Recently, Fu [19] noticed that a similar situation existed in chitosan-based MIPs. They found that both Bhb-imprinted and non-imprinted beads based on chitosan and acrylamide showed excessively high-a nity for the Bhb template after washing with a solution containing SDS and HOAc. In other words, there was no di erence in the binding properties of Bhb-imprinted gels compared to non-imprinted gels. They also conceded that in their previously published papers [20, 21] the ex- cessively high template a nity of their MIPS and uncharacteristically low a nity of the non-imprinted polymers was most likely the result of im- proper template extraction experiments [23, 24, 77]. In these experiments, the MIPs were washed with SDS-HOAc solutions, but the non-imprinted (control) polymers were not. This inadequate control resulted in unusually high selectivity and speci city. We suspect that a number of published studies related to protein-imprinted polymers su er from the same short- falls as those outlined by Fu?s publications. Therefore, the second goal of this work was to determine how im- printed and non-imprinted gels containing varying amounts of charged 18 monomer behave under two distinct wash protocols. The rst protocol involves washing the gels with an aqueous solution containing sodium do- decyl sulfate (SDS) and acetic acid (HOAc). As it is believed that SDS entrapped within the gels is the main impetus behind the high-a nity binding reported in previous studies, the second protocol that we devel- oped involves an additional step to ensure the removal of any SDS that remains bound in the gel after washing. 2.2 Materials and Methods 2.2.1 Materials Acrylamide, ammonium persulfate, bovine hemoglobin, bovine cy- tochrome C, N,N-methylenebisacrylamide, N,N,N,N-tetramethylethyldiamine, 3-methacrylamidopropyl trimethylammonium chloride, 2-acrylamido-2-methylpropane sulfonic acid, sodium dodecyl sulfate, and acetic acid were purchased from Sigma-Aldrich (St. Louis, MO) and used as received. 2.2.2 Synthesis Stock solutions of the AMPS and MAPTAC monomers were pre- pared by dissolving a speci ed amount of monomer into deionized wa- ter and titrating with 1 M NaOH or 1 M HCl to pH 7. In a typical imprinted polymer synthesis, 54 mg acrylamide (monomer), 6 mg N,N- methylenebisacrylamide (BAAm, cross-linker), 10 l of 5% (v/v) N,N,N,N- 19 tetramethylethylenediamine (TEMED, catalyst) and 12 mg Bhb template were dissolved in 1 ml of deionized water in a microcentrifuge tube. Nitro- gen was bubbled through the solution for 5 minutes to purge any oxygen that is capable of inhibiting the formation of free radicals. Subsequent to nitrogen bubbling, 10 l of 10% (w/v) ammonium persulfate (APS, initia- tor) was added to the solution. Free radical cross-linking co-polymerization occurred overnight, producing gels that were then removed from the micro- centrifuge tubes and granulated by passing through a 75 m sieve (Fisher Scienti c, Pittsburgh, PA) prior to washing. Non-imprinted, neutral hy- drogels were prepared in the same manner, in the absence of the template molecule Bhb. All polymer gels in this work were prepared as described above, with a small percentage (0.25-1.0%) of the uncharged acrylamide monomer being replaced with an anionic (AMPS) or cationic (MAPTAC) monomer to allow for systematic variation of polymer charge density. The synthesis parameters for all hydrogels prepared in this study are given in Table 2.1. 2.2.3 Bhb Template Extraction Template extraction occurred in two distinct steps. First, a series of ve washes was performed using deionized water. Subsequent to these wa- ter washes, a series of ve washes was performed using a solution contain- ing 10% (v/v) acetic acid and 10% SDS (w/v), referred to as SDS-HOAc 20 Acrylamide AMPS MAPTAC BAAm Bhb Neutral Imprinted 54.0 0.0 0.0 6.0 12.0 Non-Imprinted 54.0 0.0 0.0 6.0 0.0 0.25% AMPS Imprinted 53.87 0.14 0.0 6.0 12.0 Non-Imprinted 53.87 0.14 0.0 6.0 0.0 0.5% AMPS Imprinted 53.73 0.27 0.0 6.0 12.0 Non-Imprinted 53.73 0.27 0.0 6.0 0.0 1.0% AMPS Imprinted 53.46 0.54 0.0 6.0 12.0 Non-Imprinted 53.46 0.54 0.0 6.0 0.0 0.25% MAPTAC Imprinted 53.87 0.0 0.14 6.0 12.0 Non-Imprinted 53.87 0.0 0.14 6.0 0.0 0.5% MAPTAC Imprinted 53.73 0.0 0.27 6.0 12.0 Non-Imprinted 53.73 0.0 0.27 6.0 0.0 Table 2.1: Synthesis parameters for neutral, positively (MAPTAC) and negatively charged (AMPS) hydrogels. All values are in mg. 21 hereafter. Following synthesis and granulation, the imprinted and non-imprinted hydrogels were suspended in 2 ml of deionized water in a 15 ml polycarbon- ate centrifuge tube and centrifuged (Sorvall, Waltham, MA) for 5 minutes at 3000 rpm. This was repeated 5 times. Subsequent to these ve water washes, the gel was washed 5 times in the same manner using the SDS- HOAc solution. The supernatants from the water and SDS-HOAc washes were collected for analysis. Following these washes, all gels were equili- brated by washing repeatedly in an excess of deionized water. A total of 5 water washes and 5 SDS-HOAc were found to be su cient, as the amount of Bhb detected in the wash supernatants was found to be below the detectable limits of the UV-Vis spectrophotometer. In addition to the original protocol described above, a modi ed pro- tocol was developed to study the binding properties of Bhb imprinted and non-imprinted hydrogels under varying extraction conditions. The modi- ed protocol is nearly identical to the original protocol, the only di erence being the addition of a triplicate wash with 2 ml of 3 M NaCl prior to equilibration with deionized water. The goal of these NaCl washes was to remove any SDS remaining within the gel subsequent to the SDS-HOAc washes. 22 2.2.4 Template Rebinding In a typical template rebinding experiment, washed and granulated Bhb-imprinted and non-imprinted hydrogels were loaded with 2 ml of a deionized water solution containing 3 mg/ml of Bhb. The gels were placed on a Labquake mixer (Barnstead International, Dubuque, IA) and allowed to associate with the template for 10 minutes. Following template as- sociation, gels were removed from the mixer and subjected to the same extraction experiments described in the Bhb template extraction section. 2.2.5 UV-Vis Analysis Analysis of the supernatants collected from the washing and re- binding experiments were performed using UV-Vis measurements on a Perkin-Elmer Lambda 25 spectrophotometer. A calibration curve was pre- pared by performing absorbance measurements on deionized water solu- tions containing known amounts of Bhb. A similar curve was prepared for solutions of Bhb in SDS-HOAc. Spectral scans (190-900 nm) performed on these solutions revealed peaks at 404 nm for Bhb in water (Figure 2.1) and 395 for Bhb in the SDS-HOAc solution (Figure 2.2). Similar scans per- formed on solutions of cytochrome C in deionized water and SDS-HOAc revealed peaks at 407 nm and 397 nm respectively. Calibration curves that correlate the concentration of Bhb and cytochrome C to the measured value of UV-Vis absorbance were generated using the peak absorbance values. 23 An example of a calibration curve derived from spectral scans performed on a solution of Bhb in water is shown in Figure 2.3. The supernatant from each step of the extraction experiment was diluted appropriately to ensure that it fell within the range of the calibration curve. Figure 2.1: UV-Vis scan of Bhb in water at various concentrations. The intensity of the peaks decreases as the concentration of Bhb is lowered. 2.2.6 Hydrogel Swelling Swelling experiments were designed to determine the swollen state of the hydrogels in conditions that mimic those present during template re-binding. All swelling experiments were performed using non-imprinted hydrogels. Hydrogels of varying charge density were synthesized, granu- lated, and weighed to determine the initial mass of the gels, Mi. The gels were then washed using the original protocol described above. Subsequent 24 Figure 2.2: UV-Vis scan of Bhb in SDS-HOAc at various concentrations. The intensity of the peaks decreases as the concentration of Bhb is lowered. Figure 2.3: Calibration curve for Bhb in water. 25 to washing, the gels were then placed in 15 ml centrifuge tubes along with 2 ml of deionized water. The gels were swollen in these tubes for 10 min- utes and then centrifuged for 5 minutes at 3000 rpm. The deionized water supernatant was removed and the gels were weighed again to determine the nal weight, Mf. The swelling ratio (SR) was determined by dividing the nal mass by the initial mass as shown in Equation 2.1. SR = Mf=Mi (2.1) 2.3 Results and Discussion Characterization of molecularly imprinted polymers is usually per- formed through comparison of the absorption/binding characteristics of an imprinted polymer relative to the absorption/binding characteristics of the non-imprinted counterpart. In our studies, we perform a similar com- parison, however, we also compared non-imprinted and imprinted charged polymers relative to their uncharged counterparts to determine the role of charge density in the recognition properties of MIPs. Charge density refers to the ratio of the mass of charged monomers to the total mass of monomers within each gel. The template extraction experiment is separated into two steps, which allows us to determine the amount of Bhb bound with high and low-a nity in the hydrogels. We believe that is extremely critical to determine the 26 amount of template bound with low and high-a nity within the gels as low- a nity binding sites will be of little interest in commercial applications. Figure 2.4 shows a typical Bhb wash pro le for an acrylamide (uncharged) hydrogel imprinted with 12 mg of Bhb. Figure 2.4: Wash pro le for an uncharged (0% ionic groups) Bhb-imprinted hydrogel. Washes (1-5) were performed using deionized water, while washes (6-10) were performed using SDS-HOAc solution. Washes 1-5 represent the amount of Bhb bound (2.6 mg) in low- a nity sites, as this fraction of Bhb was removed using only deionized water, requiring no harsh solvents or high ionic strength solutions. Washes 6-10 represent the amount of Bhb bound (4.1 mg) in high-a nity sties, as this fraction of protein is only removed during washes performed using the harsh SDS-HOAc solution. The Bhb fraction removed during these washes contributes to the formation of high-a nity imprinted sites that 27 can selectively recognize Bhb in subsequent template binding experiments. It is interesting to note that a substantial fraction, (45%) 5.4 mg of the imprinted protein remains within the hydrogel, even after extraction cy- cles using the harsh removal solvent were performed. Similar results were noted in previous studies involving free-radical cross-linking copolymeriza- tion of neutral, anionic, and cationic monomers in the presence of protein molecules [53]. It is likely that functional groups on the protein are sus- ceptible to attack from free radicals, introducing the possibility of covalent bond formation between functional monomers within the gel and the tem- plate Bhb molecules. Breaking of these bonds would be extremely di cult, even under the harsh extraction conditions, and thus we expect a fraction of the template molecule to remain embedded within the hydrogel. Figures 2.5 and 2.6 are representative of template-binding experi- ments performed on uncharged Bhb-imprinted and non-imprinted poly- mer hydrogels, respectively. The imprinted polymer hydrogel (Figure 2.5) clearly demonstrates that the molecular imprinting process has a marked e ect on the absorption properties of the gel. In wash 1, a signi cant amount of protein (1.5 mg) is detected in the supernatant. It is important to make the distinction that the Bhb detected in wash 1 during rebinding experiments is not necessarily bound with low- a nity; rather, it represents the unbound fraction of Bhb left in solution upon termination of the 10-minute association step. In other words, a fraction of Bhb does not enter the gel during the 10-minute association 28 Figure 2.5: Bhb re-binding experiment performed on an uncharged (0% ionic groups) Bhb-imprinted polymer hydrogel. Washes (1-5) were per- formed using deionized water, while washes (6-10) were performed using SDS-HOAc solution. 29 step. Therefore, it will be present in the supernatant of wash 1. Figure 2.6: Binding experiment performed on a uncharged (0% ionic groups) non-imprinted hydrogel. Washes (1-5) were performed using deion- ized water, while washes (6-10) were performed using SDS-HOAc solution. Performing a summation of the Bhb extracted in washes 2-5 reveals that only a small fraction of Bhb (0.57 mg) was detected in the wash super- natant, indicating that a large fraction of Bhb remains within the imprinted hydrogels after washing them with water. A large fraction (2.1 mg) of Bhb was detected in wash cycles 6-10, indicating that most of the Bhb within this hydrogel was bound in high-a nity sites, as it was only removed us- ing the harsh SDS-HOAc wash. The non-imprinted hydrogel (Figure 2.6) shows the opposite e ect, as most of the Bhb was either unbound (2.3 mg) or extracted during washes 2-5 (0.88 mg), indicative of low-a nity inter- action between the polymer hydrogel and the Bhb molecules. Results for 30 Unbound Low-A nity(H2O) High-A nity (SDS) Uncharged Imprinted 1.8 0.15 0.57 0.01 2.1 0.31 Non-Imprinted 2.3 0.20 0.88 0.05 0.47 0.003 Table 2.2: Summary of re-binding results for uncharged Bhb-imprinted and non-imprinted hydrogels. binding experiments on uncharged gels are summarized in Table 2.2. An increase in the negative charge density of the hydrogels was achieved by copolymerizing a speci ed amount of the negatively charged AMPS monomer with uncharged acrylamide. We expected that increasing the charge density would encourage the formation of high-a nity binding sites, by allowing for the association of positive charges on the protein surface with negative charges on the AMPS functional groups. Table 2.3 shows the results of rebinding experiments performed on Bhb-imprinted hydrogels containing varying amounts of AMPS monomer. Within the gels containing 0.25% AMPS functional monomer, ap- proximately 52% (3.1 mg) of the Bhb bound within the imprinted hydrogels resided within high-a nity binding sites. Comparatively, approximately 35% (2.1 mg) was bound in high-a nity binding sites in the uncharged acrylamide gels. Thus, as the charge density of the imprinted hydrogels was increased from 0% to 0.25%, the fraction of Bhb bound in high-a nity 31 Unbound Low-A nity(H2O) High-A nity (SDS) 0.25% AMPS Imprinted 0.90 0.58 0.82 0.34 3.1 0.25 Non-Imprinted 1.6 0.25 1.0 0.06 2.2 0.51 0.50%AMPS Imprinted 0.51 0.10 0.24 0.04 2.8 0.05 Non-Imprinted 0.80 0.19 0.57 0.26 2.9 0.10 1.0%AMPS Imprinted 0.021 0.01 0.065 0.05 4.6 0.50 Non-Imprinted 1.6 0.25 1.0 0.06 2.2 0.51 Table 2.3: Results of Bhb re-binding experiments performed on negatively charged (AMPS containing) Bhb-imprinted and non-imprinted hydrogels of varying charge density. Data represents the amount of Bhb bound within each hydrogel. All values are in mg. 32 sites increased. Table 2.3 shows the data from rebinding experiments performed with Bhb on a non-imprinted 0.25% AMPS hydrogel. In these non-imprinted hydrogels, 37% (2.2 mg) of the protein bound was bound in high-a nity sites within the non-imprinted hydrogel. In addition, 27% (1.6 mg) of the Bhb remained unbound after the 10 minute association step, and 17% (1.0 mg) was bound in low-a nity sites. Both these values represent slight increases from the imprinted gel, suggesting a transition to low-a nity binding as the charge density of the gels increases. The data from re-binding experiments conducted on hydrogels con- taining 0.5% AMPS functional groups are shown in Table 2.3. In the imprinted hydrogels, approximately 4% (0.24 mg) is bound in low-a nity sites, while a much larger fraction 47% (2.8 mg) is bound in high-a nity sites within the hydrogel. Thus, this imprinted hydrogel appears to dis- play optimum performance. However, comparison of this data with data from non-imprinted hydrogels containing 0.5% AMPS functional monomers (Table 2.3) shows that at this charge density both Bhb-imprinted and non- imprinted hydrogels behave nearly identically. In other words, the im- printing e ect found in Bhb imprinted hydrogels containing 0% and 0.25% AMPS functional monomer was eliminated as the amount of AMPS groups within imprinted hydrogels was increased to 0.5% of the total monomer concentration. Bhb-imprinted and non-imprinted hydrogels containing 1% AMPS functional monomer were synthesized to con rm the disappearance 33 of the imprinting e ect at higher AMPS charge densities. Table 2.3 clearly shows that there was very little low-a nity binding in either hydrogel, as is indicated by the extremely small fraction of protein removed dur- ing the deionized water extraction cycles. Both the Bhb-imprinted and non-imprinted hydrogels displayed a large fraction of high-a nity bind- ing, as 77% (4.6 mg) of the total Bhb bound within the imprinted and non-imprinted resided in high-a nity binding sites. In addition to analyzing the e ects of the anionic AMPS functional group on the absorption properties of Bhb-imprinted polymers, we also looked at the e ects of a cationic MAPTAC group on the MIPs recogni- tion properties. All experiments conducted using the MAPTAC functional monomer were performed in an identical manner to the experiments using AMPS. Table 2.4 shows the results of re-binding experiments for Bhb- imprinted and non-imprinted polymer hydrogels containing 0.25% MAP- TAC functional monomers. The Bhb-imprinted hydrogel containing 0.25% MAPTAC functional groups is clearly superior in its Bhb recognition properties when compared with its non-imprinted counterpart. While the unbound Bhb fraction is similar in both gels, the non-imprinted gel had 15% (0.88 mg) of the total Bhb bound in low-a nity sites and 22% (1.3 mg) bound in high-a nity sites. The imprinted gel has only 7% (0.41 mg) bound in low-a nity sites while 35% (2.1 mg) is bound in high-a nity sites. Table 2.4 also shows the data for Bhb imprinted and non-imprinted 34 Unbound Low-A nity(H2O) High-A nity (SDS) 0.25% MAPTAC Imprinted 2.2 0.02 0.41 0.11 2.1 0.50 Non-Imprinted 2.3 0.10 0.88 0.03 1.3 0.15 0.50%MAPTAC Imprinted 0.022 0.08 0.046 0.01 2.5 0.32 Non-Imprinted 0.017 0.01 0.067 0.07 3.0 0.10 Table 2.4: Results of Bhb re-binding experiments performed on positively charged (MAPTAC containing) Bhb-imprinted and non-imprinted hydro- gels of varying charge density. Data represents the amount of Bhb bound within each hydrogel. All values are in milligrams. 35 hydrogels containing 0.5% MAPTAC functional groups. The data shows that the results of these binding experiments for both imprinted and non- imprinted hydrogels are nearly identical. These results suggests that unlike the anionic AMPS hydrogels, in which the imprinting e ect was not entirely eliminated until a charge density of 1% was reached, the imprinting e ect in hydrogels containing the cationic MAPTAC monomer vanished at a charge density of 0.5% MAPTAC. To test the selectivity of the Bhb-imprinted hydrogels in this study, template-rebinding experiments were performed by loading Bhb imprinted gels with cytochrome C. It is expected that Bhb imprinted gels, which exhibit an imprinting e ect will have an a nity for cytochrome C that is equal to that of their non-imprinted counterparts. In other words, nei- ther imprinted nor non-imprinted gels should have a preference for the cytochrome C template. Selectivity experiments were performed on un- charged gels and gels containing 0.25% AMPS and 0.25% MAPTAC, as these were the only samples that exhibited high-a nity for Bhb. The se- lectivity factor, , which is de ned as the ratio of bound Bhb to bound cytochrome C bound for these gels is reported in Table 2.5. The data in Table 2.5 indicates that uncharged gels have approxi- mately twice the a nity for Bhb than cytochrome C. Unlike the uncharged hydrogels, the negative (AMPS containing) hydrogels both exhibited only slight selectivity for the Bhb template. Cytochrome C has a net positive charge at pH 7, due to its relatively high isoelectric point of approximately 36 Gel ( ) Uncharged 2.1 0.25% AMPS 1.1 0.25% MAPTAC 0.62 Table 2.5: Results of selectivity experiments performed using cytochrome C template on Bhb-imprinted and non-imprinted hydrogels of varying charge density. This data represents the amount of cytochrome C extracted from each gel in milligrams. 10.0 - 10.5. This net positive charge on the cytochrome C molecules should attract the negatively charged sulfonic acid groups of the imprinted hydro- gel containing AMPS, resulting in higher a nity for cytochrome C. The selectivity experiments con rmed that the Bhb-imprinted gels absorb both Bhb and cytochrome C with similar a nity. Gels containing 0.25% MAP- TAC monomer have a slight positive charge and therefore, they should repel the positively charged cytochrome C template. Instead, we nd that Bhb- imprinted 0.25% MAPTAC gels displayed a large amount of high-a nity absorption for cytochorme C. It is interesting to note that the selectivity factor ( ) of these gels is lower than unity. In other words, Bhb-imprinted gels containing 0.25% MAPTAC actually had a higher preference for cy- tochrome C than for the original Bhb template. Possible reasons for this unexpected behavior are discussed in the following discussion. 37 In addition to the results obtained above, a separate set of exper- iments was performed using a new, modi ed wash protocol as follows. Granulated hydrogels were suspended in 2 ml of deionized water in a 15 ml polycarbonate centrifuge tube and centrifuged for 5 minutes at 3000 rpm. Aliquots were taken from the wash and analyzed using UV-Vis spec- troscopy to determine the amount of Bhb extracted from the gels. This was repeated 5 times. Subsequent to these ve water washes, the gel was washed ve times in the same manner using the SDS-HOAc solution with all supernatant fractions being collected for analysis. Following the SDS- HOAc wash series, all gels were washed in triplicate using 2 ml of a 3M NaCl solution. The purpose of this additional NaCl is to ensure the removal of any excess SDS molecules that may be bound in the gel subsequent to the SDS-HOAc washes. The hydrogels were then washed excessively with deionized water to remove any residual SDS, HOAc, or NaCl molecules that may have been absorbed. The data from a series of gels washed using this modi ed protocol are shown in Table 2.6. Comparison between the data obtained from the modi ed wash pro- tocol and that of the original protocol reveals only slight di erences in the amount of unbound and low-a nity bound Bhb. Therefore, SDS does not play a signi cant role in the low-a nity binding sites present in the gel. There was, however, a clear di erentiation between the high-a nity inter- actions in Bhb-imprinted and non-imprinted hydrogels under the original and modi ed wash conditions. Compared to the original conditions, the 38 Unbound Low-A nity(H2O) High-A nity (SDS) 0.25% AMPS Imprinted 1.02 0.60 0.33 0.014 1.46 0.18 Non-Imprinted 1.08 0.28 0.48 0.25 2.43 0.38 0.50% AMPS Imprinted 0.53 0.010 0.20 0.021 4.26 0.028 Non-Imprinted 1.25 0.17 0.47 0.049 4.03 0.42 1.0% AMPS Imprinted 0.06 0.03 0.19 0.04 4.72 0.22 Non-Imprinted 0.042 0.06 0.11 0.014 4.99 0.16 Table 2.6: Results of Bhb re-binding experiments performed on negatively charged (AMPS containing) Bhb-imprinted and non-imprinted hydrogels of varying charge density washed using the modi ed wash protocol. Data represents the amount of Bhb bound within each hydrogel. All values are in mg. 39 amount of high-a nity bound Bhb in the imprinted gel containing 0.25% AMPS monomer decreased by approximately 50% after the modi ed wash. The non-imprinted gel containing 0.25% AMPS monomer showed slightly increased high-a nity binding, although this is not believed to be signif- icant as it is within the experimental margin of error. The decrease in high-a nity binding of gels containing 0.25% AMPS can be explained in the following manner. Bhb imprinted gels washed using the original pro- tocol and modi ed protocol always contain a fraction of Bhb protein that cannot be extracted. During the SDS-HOAc wash series, SDS molecules are bound to the permanently entrapped proteins, and a fraction of this SDS remains bound within the Bhb imprinted hydrogel even after the deionized water equilibration steps are completed. When the rebinding experiments are performed, Bhb interacts with the SDS bound in the imprinted gel, resulting in high-a nity binding. The non-imprinted gels consistently dis- played similar high-a nity binding data, regardless of whether the modi- ed or original binding protocol was used. The non-imprinted gel does not initially contain Bhb, and therefore, no substrate exists within these gels to promote high-a nity binding of SDS. The absence of SDS within the non-imprinted gel resulted in binding a nity that was independent of the wash protocol used. As the amount of negatively charged AMPS groups is increased, there is a dramatic decrease in the high-a nity binding properties of the Bhb imprinted and non-imprinted hydrogel which can be explained in terms of 40 hydrogel swelling. Figure 2.7 shows the swelling data for non-imprinted hydrogels containing varying amounts of negatively charged AMPS func- tional groups. As the amount of negatively charged AMPS groups increases from 0% to 0.25%, the gel swells 14% more, resulting in similar template re- binding behavior between neutral and negatively charged gels. The hydro- gels containing 0.5% AMPS and those containing 1.0% AMPS swelled 36% and 95% more than neutral hydrogels, respectively. Bhb-imprinted and non-imprinted gels containing 0.5% AMPS and 1.0% AMPS, no detectable di erence was detected between the high-a nity binding properties of the gels. Therefore, any recognition properties imparted to the gel through the molecular imprinting process ceases to exist as the charge density increases beyond 0.5% AMPS. Gels containing 0.5% AMPS and 1.0% AMPS swelled signi cantly, enabling them to absorb a considerable amount of SDS molecules. These SDS molecules remained within the gel after equilibration with deionized water and NaCl. The high SDS content in these gels resulted in higher apparent speci c binding, increased binding capacity, and diminished im- printing factor. In fact, physically entrapped SDS molecules can also cause an imprinting factor to decrease below unity, as is indicated by the results of selectivity experiments performed on 0.25% MAPTAC gels (Table 2.5). In contrast to the gels containing negatively charged AMPS monomers, positively charged gels containing MAPTAC monomers displayed stark dif- 41 Figure 2.7: Swelling data and imprinting factor (IF) for gels with various amounts of negatively charged AMPS monomer. The columns represent the swelling data, while the solid and dashed lines represent the IF of gels washed under the original and modi ed protocol, respectively. 42 ferences across the original and modi ed wash protocols. Table 2.7 con- tains Bhb template re-binding data for gels containing varying amounts of positively charged MAPTAC monomer. The data reveals that gels containing 0.25% MAPTAC washed with the modi ed protocol exhibited a pronounced improvement in high-a nity Bhb binding over their non- imprinted counterparts. After being washed using the modi ed protocol, imprinted gels containing 0.25% MAPTAC exhibited a decrease in the un- bound Bhb fraction and a large increase in the amount of Bhb bound in high-a nity sites. The non-imprinted gels containing 0.25% MAPTAC dis- played a signi cant decrease in the amount of Bhb bound in high-a nity sites and an increase in the unbound fraction. The low-a nity binding pro les for gels containing 0.25% MAPTAC remained relatively constant across both wash protocols. Upon increasing the charge density to 0.5% MAPTAC, Bhb-imprinted gels washed under the modi ed protocol showed a slight increase in the amount of Bhb bound in high-a nity sites, while non-imprinted gels showed a slight decrease in high-a nity binding accom- panied by an increase in the unbound fraction of Bhb. This resulted in the emergence of an imprinting e ect in 0.5% MAPTAC gels which was not originally present in 0.5% MAPTAC gels washed under the original proto- col. Due to the emergence of this e ect, gels containing 1.0% MAPTAC were synthesized in order to study whether this e ect persisted at higher charge densities. Bhb-imprinted and non-imprinted gels containing 1.0% MAPTAC showed similar high-a nity binding data, indicating a dimin- 43 ished recognition e ect above 0.5% MAPTAC. To better understand the binding properties of the gels containing various amounts of positively charged MAPTAC monomer, swelling exper- iments were performed on the non-imprinted gels. Figure 2.8 displays the results of these experiments. The swelling data reveals an abrupt increase in swelling upon the addition of 0.25% MAPTAC to neutral gels with the charged gel swelling approximately 400% more than the neutral gels. In addition to swelling data, Figure 2.8 also shows the variation of imprinting factor (IF) as a function of increasing charge density. There is a marked di erence between the behavior of the MAPTAC gels washed using the two di erent protocols. When MAPTAC gels were washed using the original protocol, the imprinting factor (IF) decreased as the charge density increased from 0% to 1.0%. This decrease in IF also corresponded to an increase in swelling as the charge density increased. The decrease in IF as charge density and swelling ratio increase can be understood in terms of imprinted cavity deformation. As the charge den- sity increases, the shape of the imprinted cavities is distorted as a result of increased swelling. Highly swollen gels result in severe binding cavity dis- tortion that leads to indiscriminant binding of Bhb, regardless of whether the gel has been imprinted in or not. In gels washed using the modi ed protocol the situation becomes more complicated. As the swelling and charge density increase, MAPTAC containing gels washed with the modi- ed protocol exhibit a pronounced increase in imprinting factor at 0.25% 44 Unbound Low-A nity(H2O) High-A nity (SDS) 0.25% MAPTAC Imprinted 0.13 0.064 0.28 0.042 4.6 0.12 Non-Imprinted 4.7 0.28 0.92 0.60 0.32 0.02 0.50% MAPTAC Imprinted 0.39 0.014 0.37 0.014 3.77 0.01 Non-Imprinted 1.2 1.1 0.64 0.39 2.3 0.22 1.0% MAPTAC Imprinted 0.021 0.0 0.060 0.004 3.6 0.37 Non-Imprinted 0.021 0.0 0.69 0.13 3.6 0.49 Table 2.7: Results of Bhb re-binding experiments performed on negatively charged (MAPTAC containing) Bhb-imprinted and non-imprinted hydro- gels of varying charge density washed using the modi ed wash protocol. Data represents the amount of Bhb bound within each hydrogel. All values are in mg. 45 Figure 2.8: Swelling data and imprinting factor (IF) for gels with various amounts of positively charged MAPTAC monomer. The columns represent the swelling data, while the solid and dashed lines represent the IF of gels washed under the original and modi ed protocol, respectively. The peak in the dashed line at 0.25% MAPTAC represents the optimum monomer concentration for binding Bhb. 46 Gel ( ) Uncharged 2.1 0.25% MAPTAC 6.6 0.5% MAPTAC 4.6 Table 2.8: Selectivity factor ( ) for gels containing MAPTAC monomer washed under the modi ed protocol. MAPTAC. As the charge density continues increase, the imprinting factor decreases until it reaches a terminal value of 1 at a charge density of 1.0% MAPTAC. There is a large increase in imprinting factor for gels containing 0.25% MAPTAC washed under the modi ed protocol. The data reveals that gels containing 0.25% MAPTAC may reside at an optimum concen- tration of positively charged monomer to bind the target protein Bhb. In an attempt to verify the presence of this high imprinting factor and en- sure that this data point was not an anomaly, ve additional re-binding experiments were performed on gels washed under the modi ed protocol. Gels containing 0.25% MAPTAC consistently exhibited this high imprint- ing factor. The selectivity of gels containing 0.25% MAPTAC and 0.5% MAPTAC washed under the modi ed protocols is shown in Table 2.8. The selectivity of neutral gels is included for comparison purpose. As expected, the gel containing 0.25% MAPTAC exhibits the highest selectiv- ity, providing further con rmation that the optimum charge density exists 47 at this point. 2.4 Summary Protein imprinted hydrogels are highly complex systems, and there- fore, there are numerous explanations for the results outlined above. The binding experiments performed in this work were conducted at pH 6.8, which is nearly equal to the isoelectric point of the Bhb template. Under these conditions the Bhb template will display both positive and negative charges on its surface while maintaining a net charge of zero under the template binding conditions studied. O sets (positive or negative) in the charge density of the imprinted hydrogels should provide increased recog- nition, as the strength of the electrostatic between functional groups on the Bhb template and the complementary groups on the imprinted gels is increased. However, our experiments showed decreasing recognition with increasing charge density in case of negatively charged AMPS contain- ing gels, while gels containing positively charged MAPTAC moieties dis- played maximum selectivity and recognition properties at a charge density of 0.25%. Our results indicate that there is a complex interplay between hydrogel swelling, electrostatic interaction between the hydrogels and the template, and cavity shape recognition mechanisms that occur within im- printed polymers. The data also reveals that it may be possible to locate optimum imprinting conditions, although the correlation between hydrogel 48 charge density and template chemistry has yet to be determined. The ef- fect of template extraction chemistry on the measured binding properties of MIPs is particularly interesting, as a signi cant number of published studies [23, 24, 77] reveal that the use of SDS-HOAc for template extrac- tion is widespread. The mechanism presented within this paper reveals that the high template binding a nity measured in many studies may be the result of the e ect of the wash chemistry on the gels, and not neces- sarily the result of any recognition properties imparted to the gel through the imprinting process. 49 Chapter 3 Protein Imprinted Polyampholyte Hydrogels 3.1 Introduction Polyampohlytes are polymers that contain both positive and nega- tive charges along their backbone. The presence of both positively and negatively charged monomers with a protein imprinted hydrogel should enable increased template molecule recognition for two reasons. First, the simultaneous presence of two oppositely charged monomers in the pre- polymerization mixture should result in imprinted hydrogels with cavi- ties that contain highly speci c functional group orientation. The Bhb protein contains a distribution of positively and negatively charged func- tional groups on its surface, and therefore an imprinted hydrogel contain- ing both positively and negatively charged monomers should result in a more accurate complementary structure. Secondly, the polyampholyte hy- drogels should exhibit decreased swelling when compared to their poly- electrolyte counterparts. Repulsive interaction between oppositely charge monomers are shielded within polyampholyte hydrogels, resulting in de- creased swelling and a lower degree of cavity deformation. 50 3.2 Materials and Methods 3.2.1 Synthesis The synthesis of protein imprinted polyampholyte hydrogels was con- ducted in a similar manner to the synthesis of polyelectrolyte hydrogels as discussed in Chapter 1. The total charge density of all polyampholyte hydrogels was xed at 0.25%. A typical synthesis for a net neutral polyam- pholyte hydrogel containing 50% AMPS monomer and 50% MAPTAC monomers was performed as follows. Stock solutions of the AMPS and MAPTAC monomers were prepared by dissolving a speci ed amount of monomer into deionized water and titrating with 1 M NaOH or 1 M HCl to pH 7. 53.86 mg acrylamide (monomer), 0.07 mg AMPS (monomer), 0.07 mg (MAPTAC) monomer, 6 mg N,N-methylenebisacrylamide (cross- linker), 10 l of 5% (v/v) N,N,N,N-tetramethylethylenediamine (TEMED, catalyst) and 12 mg Bhb template were dissolved in 1 ml of deionized wa- ter in a microcentrifuge tube. Nitrogen was bubbled through the solution for 5 minutes to purge any oxygen that is capable of inhibiting the forma- tion of free radicals. Subsequent to nitrogen bubbling, 10 l of 10% (w/v) ammonium persulfate (APS, initiator) was added to the solution. Free radical cross-linking co-polymerization occurred overnight, producing gels that were then removed from the microcentrifuge tubes and granulated by passing through a 75 m sieve (Fisher Scienti c, Pittsburgh, PA) prior to washing. Non-imprinted, neutral hydrogels were prepared in the same 51 AMPS/MAPTAC Acrylamide AMPS MAPTAC BAAm Bhb 50%/50% Imprinted 53.86 0.07 0.07 6.0 12.0 Non-Imprinted 53.86 0.07 0.07 6.0 0.0 25%/75% Imprinted 53.86 0.04 0.1 6.0 12.0 Non-Imprinted 53.86 0.04 0.1 6.0 0.0 75%/25% Imprinted 53.86 0.1 0.04 6.0 12.0 Non-Imprinted 53.86 0.1 0.04 6.0 0.0 Table 3.1: Synthesis parameters for neutral, positively (MAPTAC) and negatively charged (AMPS) hydrogels. All values are in mg. manner, in the absence of the template molecule Bhb. Net neutral hy- drogels were prepared using the procedure described above and adjusting the amount of positively charged AMPS monomer and negatively charged MAPTAC monomer accordingly. The synthesis parameters for all hydro- gels prepared in this study are given in Table 3.1. 3.2.2 Bhb Template Extraction Template extraction experiments were identical to the experiments performed on protein imprinted polyelectrolyte hydrogels discussed in Chap- 52 ter 1 and were performed in the following manner. First, a series of ve washes was performed using deionized water. Subsequent to these water washes, a series of ve washes was performed using a solution contain- ing 10% (v/v) acetic acid and 10% SDS (w/v), referred to as SDS-HOAc hereafter. Finally, the gels were washed with 2 ml of 3 M NaCl prior to equilibration with deionized water. The goal of these NaCl washes was to remove any SDS remaining within the gel subsequent to the SDS-HOAc washes. The e ects of the template removal protocol on the binding proper- ties of protein imprinted and non-imprinted polyampholyte hydrogels were studied in order to determine whether or not these gels behave in a similarly to their polyelectrolyte counterparts. Speci cally, a study was performed to determine the e ects of SDS on the binding a nity of the polampholyte gels. The study was conducted in the following manner. A number of non-imprinted gels, each containing 50% AMPS and 50% were synthesized according to the experimental procedure outlined in the section above. The resultant gels were washed with either a 10% SDS solution, a 10% HOAc solution, or the SDS-HOAc solution normally used for template extraction. Subsequent to washing the gels, a typical re-binding experiment was con- ducted where the gels were exposed to a 2 ml solution containing 6 mg Bhb and allowed to associate for 10 minutes. The gels were then washed ve times with water and aliquots were taken from each wash and analyzed using UV-Vis at 404 nm to determine the concentration of Bhb within each 53 wash. Using the three components of the template removal solution sepa- rately, the e ect that each of the components has on the binding properties of the gels can be determined. A dependence (increase or decrease) of low- a nity binding on the composition of the template removal solution would be indicative of a binding a nity that is subject to modulation through simple variation of the template removal protocol. 3.3 Results and Discussion Figure 3.1 shows the results of a template re-bind experiment per- formed on a non-imprinted, net neutral polyampholyte hydrogel containing 50% AMPS and 50% MAPTAC that has been washed solely with HOAc. The gure clearly shows that after a large fraction (4.6 mg) of the Bhb template is removed after ve washes with deionized water have been com- pleted. These results indicate that HOAc itself has little e ect on the template recognition properties of the polyampholyte hydrogels. The results of a Bhb re-bind experiment performed on a non-imprinted, net neutral polyampholyte hydrogel containing 50% AMPS and 50% MAP- TAC that was washed solely with SDS is shown in Figure 3.2. The gel washed only with SDS exhibits a much lower unbound Bhb fraction (1.6 mg) compared in comparison to the unbound fraction (2.8 mg) of the sim- ilar gel that was washed only with HOAc. The decrease in the unbound fraction of Bhb in supernatant of the gel washed with SDS indicates that, 54 Figure 3.1: Binding experiment performed on a net neutral (50% AMPS - 50% MAPTAC), non-imprinted polyampholyte hydrogel washed with a 10% HOAc solution. Washes (1-5) were performed using deionized water. after template association occurs, a large fraction of the 6 mg Bhb available resides within the gel. Therefore, similar to their polyelectrolyte counter- parts, SDS molecules di use into the polyampholyte hydrogel matrix dur- ing the template extraction step and remain trapped within the matrix, acting as a high-a nity sink for the Bhb template. Figure 3.3 contains the results of a Bhb template rebinding experiment performed on a non- imprinted polyampholyte hydrogel containing 50% AMPS and 50% MAP- TAC that was washed with the SDS-HOAc solution. Approximately 1.4 mg Bhb was unbound following the template association. Therefore, the polyampholyte gels washed with SDS and SDS-HOAc behaved in a similar fashion to the polyelectrolyte gels of Chapter 1, absorbing a considerable 55 fraction of Bhb with relatively high-a nity when compared with the gel washed with HOAc only. These results indicate that SDS alone, not HOAc or the SDS-HOAc combination, is responsible for the high-a nity binding exhibited by both protein imprinted and non-imprinted polyampholyte hy- drogels. In an e ort to remove any excess SDS and prevent non-speci c binding caused by SDS entrapped within the polymer matrix, all protein imprinted and non-imprinted hydrogels were washed using the modi ed protocol discussed in Chapter 1. Speci cally, an additional wash with 3 M NaCl was added subsequent to the SDS-HOAc wash. Figure 3.2: Binding experiment performed on a net neutral (50% AMPS - 50% MAPTAC), non-imprinted hydrogel washed with a 10% SDS solution. Washes (1-5) were performed using deionized water. 56 Figure 3.3: Binding experiment performed on a net neutral (50% AMPS - 50% MAPTAC), non-imprinted hydrogel washed with a solution containing 10% SDS and 10% HOAc. Washes (1-5) were performed using deionized water. 57 3.3.1 Positively Charged Polyampholytes Figure 3.4 shows the results of a Bhb template re-binding experiment performed on an imprinted polyampholyte hydrogel containing 25 % AMPS and 75% MAPTAC. The gel shows excellent recognition properties, as approximately 4.6 mg (77%) of the Bhb template resided in high-a nity binding sites subsequent to the template association step. In comparison, an insigni cant amount (1%) of the Bhb template remained in low-a nity binding sites. The results indicate that the inclusion of both positive and negatively charged monomers in the hydrogel matrix results in an increase of the recognition properites of the MIP hydrogel. Figure 3.5 shows the Bhb template re-binding for an identical, non-imprinted polyampholyte hydrogel. Notice that the re-binding pro le is nearly the opposite of the re-binding pro le for imprinted polyampholyte hydrogel. 3.3.2 Negatively Charged Polyampholytes The negatively charged polyampholytes, those containing 25% MAP- TAC and 75% AMPS displayed similar behavior to the their positively charged counterparts. Figure 3.6 shows data from Bhb template re-binding studies performed on negatively charged, imprinted polyampholyte hydro- gels. The imprinted hydrogels displayed template recognition properties that were nearly identical to those displayed by the positively charged polyampholyte hydrogels. Approximately 4.8 mg (80%) of the Bhb tem- 58 Figure 3.4: Bhb template rebinding data for positively charged, Bhb im- printed polyampholyte hydrogels containing 25% AMPS and 75% MAP- TAC Figure 3.5: Bhb template rebinding data for positively charged, non- imprinted polyampholyte hydrogels containing 25% AMPS and 75% MAP- TAC. 59 plate was bound in high-a nity binding sites subsequent to the template association step. In addition, a small amount (1%) of the Bhb template was present in low-a nity binding sites, indicating a strong preference for the Bhb template by the imprinted gels. The non-imprinted gels (Figure 3.7) showed slight di erences from the behavior of the positively charged polyampholyte hydrogels. The non-imprinted, negatively charged polyam- pholyte gels exhibited slightly higher high-a nity binding (0.54 mg) than the non-imprinted positively charged gels (0.31 mg), although this di er- ence does not appear to be signi cant. Figure 3.6: Bhb template rebinding data for negatively charged, Bhb imprinted polyampholyte hydrogels containing 25% MAPTAC and 75% AMPS. 60 Figure 3.7: Bhb template rebinding data for negatively charged, non- imprinted polyampholyte hydrogels containing 25% MAPTAC and 75% AMPS. 3.3.3 Neutral Polyampholytes Net neutral polyampholytes, those containing equal amounts of pos- itive and negative charge, should undergo only a minimal amount of cav- ity deformation. The equal number of positive and negatively charged monomers should e ectively screen one another and therefore limit swelling. The results of Bhb template rebinding experiments performed on net neu- tral, Bhb imprinted polyampholyte hydrogels containing 50% AMPS and 50% MAPTAC are show in Figure 3.8. The gels exhibit excellent tem- plate recognition properties, with 3.6 mg Bhb bound in high-a nity sites and only 0.4 mg Bhb bound in low a nity sites. The binding experi- ment performed on non-imprinted gels, show in Figure 3.9, revealed that 61 non-imprinted net neutral hydrogels exhibited template a nity that was nearly opposite of their imprinted counterparts. Nearly 5.2 mg of Bhb was either unbound or bound in low a nity sites subsequent to the template association step, while only 0.2 mg was bound in high a nity sites. Figure 3.8: Bhb template rebinding data for net neutral, Bhb imprinted polyampholyte hydrogels containing 50% MAPTAC and 50% AMPS. 3.3.4 Selectivity The selectivity of the polyampholyte hydrogels was measured against a competitive template, cytochrome c. The selectivity is de ned as the ratio of bound Bhb to bound Cytochrome in the imprinted hydrogels. Cy- tochrome c has a lower molecular weight and higher isoelectric point than the Bhb template, therefore, it can be used to assess the recognition prop- erties of the hydrogels in relation to template size and isoelectric point. 62 Figure 3.9: Bhb template rebinding data for net neutral, non-imprinted polyampholyte hydrogels containing 50% MAPTAC and 50% AMPS. Selectivity experiments were conducted by loading Bhb imprinted and non-imprinted hydrogels with cytochrome c. Ideally, the Bhb imprinted polyampholyte hydrogels should exhibit low a nity for cytochrome c and higher a nity for the Bhb template. As Table 3.2 shows, the selectivity exhibited by the polyampholyte gels for Bhb was moderate. The positively charged polyampholyte gels exhibited slightly higher selectivity than the negatively charged gels, al- though the di erence was minor. Overall, the polyampholyte hydrogels showed a lower selectivity than gels containing only positive or negatively charged monomers. 63 Gel ( ) 25% MAPTAC - 75% AMPS 1.3 0.1 25% AMPS - 75% MAPTAC 1.2 0.1 50% AMPS - 50% MAPTAC 0.8 0.1 Table 3.2: Selectivity factor ( ) for polyampholyte gels washed under the modi ed protocol. 3.3.5 Swelling To gain a better understanding of the recognition properties of the polyampholyte hydrogels, swelling experiments were performed on non- imprinted polyampholyte hydrogels. The swelling data (Figure 3.10) re- veals that polyampholyte hydrogels containing a net negative charge showed increased swelling compared to those gels containing a net positive charge. In comparison with many of the polyelectrolye hydrogels, the swelling ratio of all polyampholyte hydrogels is considerably lower. In fact, only the poly- electrolyte hydrogels containing 0.25% AMPS, and the uncharged polyelec- trolyte hydrogels (100% Acrylamide) exhibited lower swelling ratios. The lower swelling ratios exhibited by the polyampholyte hydrogels should re- sult in higher a nity for the Bhb template, due to the fact that lower swelling ratios generally correspond to decreased cavity deformation. The polyampholyte hydrogels did, in general, exhibit a higher binding a nity 64 for the Bhb template as well as an increased imprinting factor. Figure 3.10: Swelling ratio (SR) for polyampholyte hydrogels. 3.4 Summary The results of the experiments performed on polyampholyte hydro- gels indicate that swelling can be controlled to eliminate excess cavity de- formation through the simultaneuous inclusion of positively and negatively charged co-monomers within the hydrogel. While this decreased swelling results in an increased imprinting factor, the selectivity of the hydrogels is signi canlty decreased when compared with gels containing only posi- tively or negatively charged monomers. The high speci city arises from the decreased swelling of the polyampholyte hydrogels compared to their polyelectrolyte counterparts. Decreased swelling results in decreased cavity deformation, and therefore the cavity retains its size, shape, and functional 65 group orientation upon Bhb template re-binding, resulting in higher tem- plate a nity and increased imprinting factors. The low selectivity is a result of the presence of the AMPS monomer. The positive charge of the competitive cytochrome c template at neutral pH is attracted to the nega- tively charged AMPS monomer. The polyampholyte hydrogels all contain AMPS, and therefore, the cytochrome c is attracted to negatively charged regions of the polyampholyte hydrogels, resulting in decreased selectivity. 66 Chapter 4 Future Work 4.1 Overview Despite the recent advancements made in the synthesis and charac- terization of MIPs which can selectively recognize proteins and peptides in aqueous media, there is still an enormous amount of work that needs to be conducted in the eld. Hydrogen bonds, hydrophobic and electrostatic in- teractions are the dominant mechanisms through which template-receptor complexation occurs in aqueous systems. However, the exact recognition mechanism, and the way in which the aforementioned forces cooperate to recognize target molecules in many MIP systems is largely unknown. In addition, many fundamental properties of aqueous MIPs, such as the de- pendence of MIP a nity on cross-linker concentration, functional monomer concentration, solvent, bu er composition, ionic strength, pH, and tem- perature have yet to be systematically studied. Finally, the recognition binding of larger macromolecular complexes, such as viruses [7, 17] and cells, has gone largely unstudied. These factors combined with the increas- ing interest in creating materials which are capable of mimicking biological processes, ensure that the study of MIPs capable of the selective recogni- tion of peptides, proteins, and other biological macromolecules in aqueous 67 media will receive considerable attention in years to come. The recognition properties of molecularly imprinted hydrogels should be studied using a mixture of proteins in solution elucidate a greater un- derstanding of the speci city and selectivity. This is currently a very dif- cult task, as the current method for measuring the amount of protein bound within the hydrogels involves concentration measurements using a UV/Vis spectrophotometer. The presence of two competitive proteins in solution simultaneously makes them nearly indistinguishable, and there- fore the binding properties would be extremely di cult to elucidate. This di culty may be overcome by carefully selecting proteins that have UV- Vis peaks at easily distinguishable wavelengths, or by tagging the proteins with uorescent dyes such as rhodamine. Protein imprinted hydrogels should be synthesized with monomers other than those used in this study. Acrylic acid, for example, could be used in place of the AMPS monomer, to determine how the binding properties of the hydrogels change when a weak acid is used in place of a strong one. Similarly, 2-dimethylaminoethyl methacrylate (DMAEMA) could be used in place of MAPTAC. In addition, derivatives of acrylamide such as an N-(tert-butyl)acrylamide, which contain hydrophobic moieties, could be added to the hydrogels to determine their e ect on imprinting. There exists a large library of acrylamide co-monomers available which are susceptible to free radical polymerization, and therefore, numerous combinations exist to optimize selectivity and speci city within hydrogels. 68 Bibliography [1] Alberts, B.; Bray, D.; Lewis, J.; Ra , M.; Roberts, K.; Watson, J. D., Molecular Biology of the Cell. 3rd ed.; Garland Publishing, Inc.: New York, 1994. [2] Andersson, L. I.; Miyabayashi, A.; O?Shannessy, D. J.; Mosbach, K. Journal of Chromatography 1990, 516, 323-331. [3] Andersson, L. I.; Mosbach, K. Journal of Chromatography 1990, 516, 313-322. [4] Andersson, L. I.; O?Shannessy, D. J.; Mosbach, K. Journal of Chro- matography 1990, 513, 167-179. [5] Ariga, K.; Kunitake, T. Accounts of Chemical Research 1998, 31, 371-378. [6] Bacskay, I.; Takatsy, A.; Vegvari, A.; Elfwing, A.; Balllagi-Pordany, A.; Kilr, F.; Hjertn, S. Electrophoresis 2006, 27, (23), 4682-4687. [7] Bolisay, L. D., Culver, J.N, Ko nas, P. Biomaterials 2006, 27, (22), 4165-4168. [8] Bolisay, L. D.; Culver, J. N.; Ko nas, P. Biomacromolecules 2007, 8, (12), 3893-3899. [9] Bossi, A.; Bonini, F.; Turner, A. P. F.; Piletsky, S. A. Biosensors and Bioelectronics 2007, 22, (6), 1131-1137. [10] Bossi, A.; Piletsky, S. A.; Piletska, E. V.; Righetti, P. G.; Turner, A. P. F. Analytical Chemistry 2001, 73, (21), 5281-5286. [11] Byrne, M. E.; Park, K.; Peppas, N. A. Advanced Drug Delivery Re- views 2002, 54, 149-161. [12] Camilli, A.; Bassler, B. L. Science 2006, 311, (5764), 1113-1116. [13] Cheng, Z.; Zhang, L.; Li, Y. Chemistry - A European Journal 2004, 10, 3555-3561. 69 [14] Demirel, G.; Ozcetin, G.; Turan, E.; Caykara, T. Macromolecular Bioscience 2005, 5, 1032-1037. [15] Ceylan, D.; Can, V.; Okay, O. Journal of Macromolecular Science, Part A: Pure and Applied Chemistry 2006, 43, 1635-1649. [16] Dickert, F.; Hayden, O.; Lieberzeit, P.; Pal nger, C.; Pickert, D.; Wol , U.; Scholl, G. Sensors and Actuators B-Chemical 2003, 95, (1-3), 20-24. [17] Dickert, F. L., Hayden, O., Bindeus, R., Mann, K.J., Blaas, D., Waig- mann, E. Analytical and Bioanalytical Chemistry 2004, 378, (8), 1929- 1934. [18] Fernandes, R.; Tsao, C.; Hashimoto, Y.; Wang, L.; Wood, T. K.; Payne, G. F.; Bentley, W. E. Metabolic Engineering 2007, 9, (2), 228-239. [19] Fu, G.-Q.; Yu, H.; Zhu, J. Biomaterials 2008, 29, 2138-2142. [20] Fu, G. Q.; Li, H. Y.; Yu, H. F.; Liu, L.; Yuan, Z.; He, B. L. Reactive and Functional Polymers 2006, 66, 239-246. [21] Fu, G. Q.; Zhao, J. C.; Yu, H.; Liu, L.; He, B. L. Reactive and Func- tional Polymers 2007, 67, 442-450. [22] Greene, N. T.; Shimizu, K. D. Journal of the American Chemical Society 2005, 127, 5695-5700. [23] Guo, T. Y.; Xia, Y. Q.; Hao, G. J.; Song, M. D.; Zhang, B. H. Bio- materials 2004, 25, 5905-5912. [24] Guo, T. Y.; Xia, Y. Q.; Wang, J.; Song, M. D.; Zhang, B. H. Bioma- terials 2005, 26, 5737-5745. [25] Hansen, D. E. Biomaterials 2007, 28, (29), 4178-4191. [26] Hart, B. R.; Shea, K. J. Macromolecules 2002, 35, 6192-6201. [27] Haupt, K.; Mosbach, K. Chemical Reviews 2000, 100, 2495-2504. [28] Haupt, K. Reactive and Functional Polymers 1999, 41, (1-3), 125-131. 70 [29] Haupt, K.; Mayes, A. G.; Mosbach, K. Analytical Chemistry 1998, 70, (18), 3936-3939. [30] Hawkins, D. M.; Stevenson, D.; Reddy, S. M. Analytica Chimica Acta 2005, 542, 61-65. [31] Hayden, O.; Bindeus, R.; Haderspock, C.; Mann, K.; Wirl, B.; Dick- ert, F. L. Sensors and Actuators B-Chemical 2003, 91, 316-319. [32] Henry, O. Y. F.; Cullen, D. C.; Piletsky, S. A. Analytical and Bioan- alytical Chemistry 2005, 382, 947-956 [33] Hirayama, K.; Sakai, Y.; Kameoka, K. Journal of Applied Polymer Science 2001, 81, 3378-3387. [34] Huang, J.-T.; Zhang, J.; Zhang, J.-Q.; Zheng, S.-H. Journal of Applied Polymer Science 2005, 95, 358-361. [35] Ishizu, K.; Shiratori, S.; Hosokawa, T. Angewandte Chemie Interna- tional Edition 2001, 40, (15), 2821-2823. [36] Janiak, D. S.; Ko nas, P. Analytical and Bioanalytical Chemistry 2007, 389, (2), 399-404. [37] Kang, Y.; Walish, J. J.; Gorishnyy, T.; Thomas, E. L. Nature Mate- rials 2007. [38] Kempe, M.; Mosbach, K. Journal of Chromatography A 1995, 691, 317-323. [39] Klein, J. U.; Whitcombe, M. J.; Mullholland, F.; Vulfson, E. N. Ange- wandte Chemie International Edition 1999, 38, (13/14), 2057-2060. [40] Kriz, D.; Ramstrm, O.; Svensson, A.; Mosbach, K. Analytical Chem- istry 1995, 67, (13), 2142-2144. [41] Lemieux, R. U. Accounts of Chemical Research 1966, 29, 373-380. [42] Liao, J. L.; Wang, Y.; Hjertn, S. Chromatographia 1996, 42, (5/6), 259-262. [43] Lin, T.-Y.; Hu, C.-H.; Chou, T.-C. Biosensors and Bioelectronics 2004, 20, 75-81. 71 [44] Lin, C.-Y.; Tai, D.-F.; Wu, T.-Z. Chemistry - A European Journal 2003, 9, 5107-5110. [45] March, J. C.; Rao, G.; Bentley, W. E. Applied Microbiology and Biotechnology 2003, 62, (4), 303-315. [46] Mathews, C. K.; Holde, K. E. V., Biochemistry. 2nd ed.; The Ben- jamin/Cummings Publishing Company, Inc.: Menlo Park, 1995; p 1159. [47] Matsui, J.; Nicholls, I. A.; Takeuchi, T.; Mosbach, K.; Karube, I. Analytica Chimica Acta 1996, 335, 71-77. [48] Minoura, N.; Burow, M. Biochemical and Biophysical Research Com- munications 1996, 227, 419-422. [49] Miyata, T.; Jige, M.; Nakaminami, T.; Uragami, T. Proceedings of the National Academy of Sciences 2006, 103, (5), 1190-1193. [50] Mosbach, K.; Ramstrm, O. Bio-Technology 1996, 14, 163-170. [51] Nicholls, I. A.; Ramstrm, O.; Mosbach, K. Journal of Chromatography A 1995, 691, 349-353. [52] Nishino, H.; Huang, C.-S.; Shea, K. J. Angewandte Chemie Interna- tional Edition 2006, 45, 2392-2396. [53] Ou, S. H., Wu, M.C., Chou, T.C., Liu, C.C. Analytica Chimica Acta 2004, 504, 163-166. [54] Pang, X.; Cheng, G.; Lu, S.; Tang, E. Analytical and Bioanalytical Chemistry 2006, 384, 225-230. [55] Piletsky, S. A.; Andersson, H. S.; Nicholls, I. A. Journal of Molecular Recognition 1998, 11, 94-97. [56] Piletsky, S. A.; Andersson, H. S.; Nicholls, I. A. Macromolecules 1999, 32, 633-636. [57] Rachkov, A.; Minoura, N. Journal of Chromatography A 2000, 889, 111-118. 72 [58] Rachkov, A.; Minoura, N. Biochimica et Biophysica Acta 2001, 1544, 255-266. [59] Ramstrm, O.; Andersson, L. I.; Mosbach, K. Journal of Organic Chemistry 1993, 58, 7562-7564. [60] Rezeli, M.; Kilr, F.; Hjertn, S. Journal of Chromatography A 2006, 1109, 100-102. [61] Sellergren, B. Trends in Analytical Chemistry 1997, 16, (6), 310-320. [62] Sellergren, B. Angewandte Chemie International Edition 2000, 39, (6), 1031-1037. [63] Sellergren, B.; Lepisto, M.; Mosbach, K. Journal of the American Chemical Society 1988, 110, 5853-5860. [64] Shea, K. J. Trends in Polymer Science 1994, 2, 166. [65] Shi, H.; Tsai, W.-B.; Garrison, M. D.; Ferrari, S.; Ratner, B. D. Nature 1999, 398, 593-597. [66] Sperandio, V.; Torres, A. G.; Giron, J. A.; Kaper, J. B. Journal of Bacteriology 2001, 183, (17), 5187-5197. [67] Steinke, J.; Sherrington, D. C.; Dunkin, I. R. Advances in Polymer Science 1995, 123, 81-125. [68] Tai, D.-F.; Lin, C.-Y.; Wu, T.-Z.; Chen, L.-K. Analytical Chemistry 2005, 77, 5140-5143. [69] Takatsy, A.; Sedzik, J.; Kilr, F.; Hjertn, S. Journal of Separation Science 2006, 29, (18), 2810-2815. [70] Takatsy, A.; Vegvar, A.; Hjertn, S.; Kilr, F. Electrophoresis 2007, 28, (14), 2345-2350. [71] Takeuchi, T.; Haginaka, J. Journal of Chromatography B 1999, 728, (1), 1-20. [72] Titirici, M. M.; Sellergren, B. Analytical and Bioanalytical Chemistry 2004, 378, 1913-1923. 73 [73] Vaidya, A. A.; Lele, B. S.; Kulkarni, M. G.; Mashelkar, R. A. Journal of Applied Polymer Scienced 2001, 81, 1075-1083. [74] Vendeville, A.; Winzer, K.; Heurlier, K.; Tang, C. M.; Hardier, K. R. Nature Reviews Microbiology 2005, 3, (5), 383-396. [75] Wang, L.; Hashimoto, Y.; Tsao, C.; Valdes, J. J.; Bentley, W. E. Journal of Bacteriology 2005, 187, (6), 2066-2076. [76] Wul , G. Angewandte Chemie International Edition 1995, 34, (17), 1812-1832. [77] Xia, Y. Q.; Guo, T. Y.; Hao, G. J.; Song, M. D.; Zhang, B. H.; Zhang, B. L. Biomacromolecules 2005, 6, 2601-2606. [78] Yang, L.; Portugal, F.; Bentley, W. E. Biotechnology Progress 2006, 22, (2), 387-394. [79] Zhao, Z.; Wang, C.; Guo, M.; Shi, L.; Fan, Y.; Long, Y.; Mi, H. FEBS Letters 2006, 580, 2750-2754. [80] Zhou, H. J.; Zhang, Z. J.; He, D. Y.; Hu, Y. F.; Huan, Y.; Chen, D. L. Analytica Chimica Acta 2004, 523, (2), 237-242. 74