Alle rechten voorbehouden. Niets uit deze uitgave mag worden vermenigvuldigd en/of openbaar gemaakt worden door middel van druk, fotokopie, microfilm, elektronisch of op welke andere wijze ook zonder voorafgaandelijke schriftelijke toestemming van de uitgever.

ISBN [90-8649-058-1] D/ [2006/10.705/55]

Departement Chemie Biochemie, mol. en struct.biologie, Celestijnenlaan 200G B-3001 Leuven

Constructing Novel Interface Layers towards the Realization of High Performance Biosensors

Au

Promotoren: Prof. Dr. Y. Engelborghs Prof. Dr. G. Borghs

Proefschrift voorgedragen tot het behalen van de graad van Doctor in de Wetenschappen Door:

Zhou Cheng 2006

All rights reserved. No part of the publication may be reproduced in any form by print, photoprint, microfilm, electronic or any other means without written permission from the publisher.

Constructing Novel Interface Layers towards the Realization of High Performance Biosensors

© 2006 Faculteit Wetenschappen, Geel Huis, Kasteelpark Arenberg 11, 3001 Heverlee (Leuven)

KATHOLIEKE UNIVERSITEIT LEUVEN FACULTEIT WETENSCHAPPEN

Cheng Zhou In samenwerking met Interuniversitair Micro-Elektronica Centrum Kapeldreef 75 B-3001 Leuven

KATHOLIEKE UNIVERSITEIT LEUVEN FACULTEIT WETENSCHAPPEN Biochemie, mol. en struct.biologie, Celestijnenlaan 200G B-3001 Leuven

Constructing Novel Interface Layers towards the Realization of High Performance Biosensors

Promotoren: Prof. Dr. Y. Engelborghs Prof. Dr. G. Borghs

Proefschrift voorgedragen tot het behalen van de graad van Doctor in de Wetenschappen Door: Cheng Zhou

In samenwerking met Interuniversitair Micro-Elektronica Centrum Kapeldreef 75 B-3001 Leuven 30, October 2006

ii

iii

Acknowledgement

I want to give my grateful thanks to Prof. Gustaaf Borghs and Prof. Yves Engelborghs. They introduced me the great opportunity of conducting my PhD in the brilliant field of biosensors. As my promoters, they had provided me endless support and assistance in not only the scientific research but also the personal life. Without their help, I could not have been able to get used to the life in Belgium so fast and smoothly. Without their support, I could not have been able to develop my full potential for my PhD study. Without their guidance, I could not have been able to make those achievements during my PhD research. I want to thank Prof. Wim Dehaen and Prof. Mario Smet. They had kindly provided me the access to their lab facilities, which enabled me to conduct my research in a time- and cost-efficient way. I would like to also thank them for their invaluable discussion and assistance in preparing the new materials used in my research. I want to thank my wonderful colleagues in IMEC. They made me feel just like in family. I want to thank Dr. Andrew Campitelli, Dr. Jean-Michel Friedt, Dr. Angelina Angelova, Dr. Wim Laureyn, Dr. Filip Frederix, Dr. Kristien Bonroy, Dr. Carmen Bartic, Dr. Laurent Francis, Randy De Palma, and Gunter Reekmans. As experienced colleagues, they provided lots of help in getting me familiar with the lab and the new instruments. They also provided invaluable guidance, assistance, and discussion in my PhD search. I want to thank Koen De Keersmaecker. He taught me a lot of useful knowledge of microelectronics. He helped significantly in the nanofabrication part of my PhD work. He provided lots of scientific input and language input for my publications. I want to thank Dries Braeken, Kurt Winter, and Bart van Meerbergen, their strong support in the biology field enabled me to combine the surface chemistry work closely with the real biology application. They also learned me plenty of knowledge of biology, which are great helpful in my PhD study. I also want to thank other colleagues in my groups for their great corroboration work. Finally, those IMEC colleagues, who provided both lab and technical supports for the PhD work, are gratefully thanked.

iv

I would like to thank my parents. They brought me to this world and cultivated me to a man willing and able to take responsibility and commitment. I want to give my grateful thanks to my wife. In the past 5 years, she always stood together with me to face all kinds of challenges. She shared my laugh and my tears. She encouraged me keeping on making efforts toward success. She took care of my life. She also assisted in invaluable discussion for my research. She is not only a good wife in my life but also a good partner in my career. I will also thank my little angel, Chunyi. Her birth is the greatest gift for the celebration of my PhD success. I would like to thank KULeuven (Chem. depart.) and IMEC for providing me the opportunity for PhD study. IMEC is gratefully thanked for the generous financial support in both research and personal life. I want to give my thanks to all those people who had helped me during the past 5 years. Thank you all!

Cheng Zhou

v

List of publications

Patents: “Self-assembled grafted polymeric monolayer for use in biosensor technology”, by Cheng zhou, Gustaaf Borghs, Wim Laureyn, US patent filing in March, 2005.

Journal papers 1) Cheng zhou, Jean-Michel Friedt, Angelina Angelova, Kang-Hoon Choi, Wim Laureyn, Filip Frederix, Laurent Francis, Andrew Campitelli, Yves Engelborghs, and Gustaaf Borghs, “Human immunoglobulin adsorption investigated by means of quartz crystal microbalance dissipation, atomic force microscopy, surface acoustic wave, and surface plasmon resonance techniques”, Langmuir, 20(14): 5870-5878, 2004 2) Laurent Francis, Jean-Michel Friedt, Randy De Palma, Cheng zhou, Carmen Bartic, Patrick Bertrand, Andrew Campitelli, “Techniques to evaluate the mass sensitivity of Love mode surface acoustic wave biosensors”, IEEE International Ultrasonics, Ferroelectrics, and Frequency Control. 50th Anniversary Joint Conference, 241-249, 2004 3) Cheng zhou, Vadim K. Khlestkin, Dries Braeken, Koen De Keersmaecker, Wim Laureyn, Yves Engelborghs, and Gustaaf Borghs, “Solvent controlling organization of polymeric self-assembled monolayers on gold: an approach for construction of protein-resistant surfaces”, Langmuir, 21(13): 59885996, 2005 4) Dries Braeken, Cheng Zhou, Roeland Huys, Carmen Bartic, Koen De Keersmaecker, Kurt Winters, Geert Callewaert and Gustaaf Borghs, “LGlutamate Detection Using A Poly-L-Lysine coated ENFET”, Proc. SPIE, 5839:443-452, 2005 5) Laurent Francis, Jean-Michel Friedt, Cheng Zhou, Patrick Bertrand, “In Situ Evaluation of Density, Viscosity and Thickness of Adsorbed Soft Layers by

vi

Combined Surface Acoustic Wave and Surface Plasmon Resonance”, Anal. Chem., 78(12): 4200-4209, 2006 6) Cheng Zhou, Gunter Reekmans, Koen De Keersmaecker, Filip Frederix, Thierry Conard, Carmen Bartic, Yves Engelborghs, and Gustaaf Borghs, “Poly(ethylene glycol)-grafted polymeric monolayers on gold: Interfacial Architecture and Resistance to Protein Adsorption”, Colloids and Surfaces: B-biointerface, first revision 7) Cheng zhou, Gunter Reekmans, Yves Engelborghs, and Gustaaf Borghs, “SPR immunosensor based on novel polymeric monolayers: Minimized nonspecific adsorption and Enhanced sensitivity” submitted to Analytical Chemistry

vii

Short CV Cheng Zhou, born in 20 Aug, 1975, P.R. China 1994 to 1998: Bachelor of Engineering, Department of Polymer Materials and Engineering, East China University of Science and Technology, P.R. China 1999 to 2001: Master of Chemistry, Department of Polymer Science, Zhejiang University, P.R. China 2001 to 2006: PhD of Biochemistry KULeuven and IMEC (Interuniversity Microelectronic Centre), Belgium

viii

ix

Abstract

Biosensors have experienced a phenomenal growth in the past two decades. Specific features, such as high sensitivity and specificity, fastness of response, easiness of use, and ability of miniaturization, have made biosensors suitable for a variety of applications including biomedical science, healthcare, agriculture, industrial process control, environmental monitoring, defense and security. Biosensors have to meet the requirements of sensitivity, stability, and reliability, before they can be exploited in real applications. The performance, i.e. sensitivity, stability, and reliability, of a biosensor depends on the properties of its components: transducer, interface layer, and biological sensing elements. Among them, the properties of the interface layer play a crucial role. The aim of the current study was to develop high quality interface layers for constructing high performance biosensors. In the first part of this study, a fundamental phenomenon, i.e. protein adsorption at solid-liquid interface, which has great importance and relevance in biosensors, was studied by a novel methodology. This methodology, termed as combined Surface Plasmon Resonance (SPR) and Quartz Crystal Microbalance with Dissipation (QCM-D), combined SPR and Surface Acoustic Wave (SAW), and combined QCM-D/AFM, enable us to obtain both quantitative and qualitative details of the protein adsorption process. The insight of the mechanism of protein adsorption at solid-liquid interface provided invaluable information for the optimal design of biosensors. The significant nonspecific protein adsorption has been a long term problem that demolishes the performance of biosensors. Conventional interface layers, which are used to prevent protein adsorption, suffer either from limited stability or from complexity in the preparation. In the second part of this study, an easy and fast approach, which base on a self-assembled polymeric monolayer, was developed to construct robust protein-resistant interface layers. The resulted polymeric film demonstrated excellent properties in preventing nonspecific protein adsorption. The third part of this study was focused on the realization of a high performance immunosensor using a novel functional Poly(ethylene glycol) (PEG) grafted polymeric monolayer. Compared to SPR immunosensor basing on the conventional interface layers (mixed self-assembled monolayers and

x

commercial CM5 chip), The SPR immunosensor basing on the novel functional PEG grafted polymeric monolayer showed improved sensitivity. The performance of the SPR immunosensor can be further optimized through suitable antibody modification and oriented antibody immobilization strategies. The fourth part of this study coped with an efficient approach to achieve heating controllable immobilization of DNA probes, which is the crucial step in the realization of microelectronic DNA chip. The silane technique was combined with specific surface Diels-Alder reactions to accomplish addressable immobilization of various types of DNA probes on the desirable sensor areas. The organization of neurons cells on the transistor has to be well controlled in order to realize the specific function of hybrid neuron transistor. The last part of this study was aiming to deal with this challenge. A patterned PEG gel, which provides both physiochemical features and topographic features, was explored to guide the growth of neuron network. Preliminary studies suggest that this novel pattering method is very promising in controlling the organization of neuronal networks and has potential for the realization of the hybrid neurontransistors. In conclusion, the multidisciplinary research has realized several accomplishments with regards to constructing high quality interface layers for various biosensor applications.

xi

List of abbreviations Ab Antibody AC Alternating Current ACS Antigen-Combining Site Ag Antigen AFM Atomic Force Microscopy AlGaN Aluminum Gallium Nitride APTES 3-(AminoPropyl)TriEthoxySilane BMB 1,4-Bis-MaleimidoButane BSA Bovine Serum Albumin CA Contact Angle CDRs Complementarity Determining Regions CDT Carbohydrate-Deficient Transferrin CV Cyclic Voltammetry Da Dalton DIC Differential Interference Contrast DNA DeoxyriboNucleic acid DMF N,N-DiMethylFormamide DMSO DiMethyl SulphOxide EDC N-ethyl-N’-(3-dimethylaminopropyl)carbodiimide EDTA Ethylene-Diamine-Tetra-Acetate EG/EO Ethylene oxide EG3OH 2-(2-(2-(11-mercaptoundecyloxy)ethoxy)ethoxy)-ethanol Fab Fragment antigen binding FB Fibrinogen GA-FT-IR Grazing-Angle Fourier Transform Infrared Spectroscopy HBS Hepes Buffered Saline HT Human Transferring HSA Human Serum Albumin IDE InterDigitated Electrode IDTs Interdigital Transducers IgG Immunoglobin G IMEC Interuniversity Microelectronic Center IR Infrared Spectroscopy ISFETs Ion Sensitive Field-Effect Transistors

xii

LBs LOD mAbs 16-MHA

Lipid Bilayers Limit of Detection monoclonal Antibodies 16-Mercapto-1-Hexadecanoic Acid

MOS MPTMS 11-MUOH Mw NHS NMR ODT OEG pAbs PBS PCR PEG PEG-DA PSA QCM QCM-D RMS RNA RU SA SAW SAMs SBMs SEM SMCC SPR THF TLC TPB TR UV-vis XPS

Metal Oxide Silicon (3-MercaptoPropyl)TriMethoxySilane 11-Mercapto-1-Undecanol Molecular Weight N-HydroxySuccinimide Nuclear Magnetic Resonance 1-OctaDecaneThiol OligoEthylene Oxide polyclonal Antibodies Phosphate Buffered Saline Polymerase Chain Reaction PolyEthylene Glycol Poly (Ethylene Glycol) DiAcrylate Prostate Specific Antigen Quartz Crystal Microbalance Quartz Crystal Microbalance-Dissipation Root Mean Square RiboNucleic Acid Resonance Units or Refractive Unit StreptAvidin Surface Acoustic Wave Self-assembled Monolayers Supported Bilayer Membranes Scanning Electron Microscopy Succinimidyl 4-(N-Maleimidomethyl)Cyclohexane-1-Carboxylate Surface Plasmon Resonance TetraHydroFurane Thin Layer Chromatography Transmission Plasmon Biosensor Transferrin Ultra Violet-visible X-ray Photoelectron Spectroscopy

xiii

Table of Content 1. Introduction 1.1. General introduction 1.2. Transducer aspects of biosensors 1.2.1. Electrochemical biosensors 1.2.1.1. Ion Sensitive Field-Effect Transistor (ISFET) 1.2.2. Optical biosensors 1.2.2.1. Surface Plasmon Resonance (SPR) 1.2.3. Piezoelectric biosensors 1.2.4. Magnetic biosensors 1.2.5. Mechanical biosensors 1.3. Biochemical aspects of biosensors 1.3.1. Immunosensors 1.3.1.1. Antibody and its properties 1.3.1.2. The antibody-antigen interactions 1.3.1.3. The assay formats 1.3.1.4. The limitation of immunosensors 1.3.2. DNA-based sensors 1.3.3. cells-based sensors 1.3.3.1. cell-based detection 1.3.3.2. cells-based sensor using excitable cells 1.4. The interface aspects 1.5. Performance of biosensor 1.6. Summary

1 1 2 2 2 3 4 5 6 7 7 7 7 10 10 12 13 16 16 18 22 22 24

2. The design of the high quality interface layer 2.1. Introduction 2.2. The role of the interface layer 2.3. The immobilization of biological sensing elements in interface layer 2.4. Current available interface layers 2.4.1. Self-assembled Monolayers 2.4.2. Polymer interface 2.4.3. Lipid bilayers 2.4.4. Sol-gel system 2.5. Task and proposal

29 29 29 34 36 36 38 42 43 44

xiv

3. Protein Adsorption at Solid-liquid Interface 3.1. Introduction 3.2. Human Immunoglobulin G Adsorption on hydrophobic surface 3.2.1. Experimental section 3.2.1.1. Materials and preparation of the sensor chip surfaces 3.2.1.2. Methods and Instrumentation 3.2.2. Results and Discussion 3.2.2.1. QCM-DTM and combined SPR/SAW measurements 3.2.2.2. Combined QCM-DTM/AFM measurement 3.2.3. Summary 3.3. Blood Proteins Adsorption on Hydrophobic Surface 3.3.1. Experimental section 3.3.1.1. Materials and preparation of sensor chips surface 3.3.1.2. Method and experimental procedure 3.3.2. Results and Discussion 3.3.2.1. Comparison between blood proteins 3.3.2.2. Comparison between whole IgG and its fragments 3.3.3. Summary 3.4. The influence of surface property on the adsorption 3.5. Conclusion

55 55 56 58 58 59 63 63 75 79 80 81 81 81 81 81 86 89 90 92

4. Construction of a protein-resistance surface 99 4.1. Introduction 99 4.2. Solvent controlled organization of self-assembled polymeric monolayer 101 4.2.1. Experimental Section 103 4.2.1.1. Materials 103 4.2.1.2. Preparation of gold substrates and polymeric monolayer 103 4.2.1.3. Characterization methods 104 4.2.1.4. Synthesis of the PEG-copolymer 105 4.2.2. Results and Discussion 106 4.2.2.1. Electrochemical Characterization of the polymeric monolayer 107 4.2.2.2. The thickness and the wettability of the polymeric monolayers 110 4.2.2.3. XPS characterization of the polymeric monolayers 111

xv

4.2.2.4.

Surface morphology investigated by Atomic Force Microscopy 115 4.2.2.5. Protein adsorption evaluated by Surface Plasmon Resonance 117 4.2.2.6. Patterned polymeric monolayer achieved by lift-off method 120 4.2.3. Summary 122 4.3. Interfacial Architecture and Resistance to Protein Adsorption 123 4.3.1. Experimental sections 125 4.3.1.1. Materials 125 4.3.1.2. Preparation of gold substrates and polymeric monolayer 125 4.3.1.3. Characterization methods 125 4.3.2. Result and discussion 131 4.3.2.1. Architectures of the PEG grafted polymeric monolayers on gold 131 4.3.2.2. Influence of polymeric monolayer organization on protein adsorption 142 4.3.3. Summary 147 4.4. Conclusion 148 5. Immunosensing using Polymeric monolayer as Interface 157 5.1. Introduction 157 5.2. General introduction of SPR sensing platform 159 5.3. Design and characterization of self-assembled polymeric monolayers 162 5.3.1. Synthesis of the PEG grafted copolymers 165 5.3.2. Preparation of polymeric monolayer and characterization method 169 5.3.3. Results and discussion 169 5.3.4. Summary 172 172 5.4. Immunosensing of S100 protein 5.4.1. Materials and methods 173 5.4.2. Results and discussion 174 5.4.3. Summary 186 5.5. Immunosensing of human transferrin (HT) proteins 187 5.6. Comparison of the performance of different interface layers for

xvi

immunosensing applications 190 5.6.1. Experimental and methods 191 5.6.2. Results and discussion 191 5.6.3. Summary 200 5.7. Modification and oriented immobilization of antibodies 200 5.7.1. Antibody immobilization through biotin-streptavidin couple 201 5.7.1.1. Materials and methods 201 5.7.1.2. Results and discussion 202 5.7.1.3. Summary 204 5.7.2. Protein G mediated antibody immobilization 204 5.8. Conclusion 206 6. Construction of microelectronic DNA chips-proof of concept 6.1. Introduction 6.2. Experiment and methods 6.3. Results and discussion 6.4. Conclusion

215 215 219 220 230

7. Controlling the Neuron Cells Organization on Sensor Surface-A Step to Hybrid Neuron-Transistor 235 7.1. Introduction 235 7.2. Methods for neurons cells patterning 237 7.3. Hippocampal neuron growth on a PEG gel pattern 245 7.4. Conclusion 248 8. Conclusion and Outlook 253 8.1. Task and Purpose 253 8.2. Fundamental study of protein adsorption 254 8.3. Constructing novel interface for preventing nonspecific adsorption 255 8.4. Immunosensor based on novel functional PEG grafted polymeric monolayer 257 8.5. A novel method to achieve microelectronic DNA chips 258 8.6. Controlling the organization of neurons cells on field effect transistor using patterned PEG gels 259 8.7. Final conclusion and outlook 260

Chapter 1: Introduction 1.1 General introduction A biosensor is an analytical device made up of a biological sensing element and a transducer. The sensing element detects the presence of an analyte via a specific interaction and generates a signal whose intensity is either directly or inversely proportional to the number of interactions between the analyte and the sensing element over a given period of time. The transducer in turn receives the signal coming from the sensing element and produces an optical or electronic signal that is directly proportional to the intensity of the signal received [1] (Figure 1).

biosensors

Bio-receptor

transducer

diffusion

Interface linker layer

complex sample matrix

Target analyte

Signal conversion

Data recording and display

Figure 1 the schematic illustration of a biosensor.

Chapter 1: Introduction

The biological sensing elements can be: an enzyme or multi-enzyme system; an organelle; a membrane component; a bacterial cell or other whole cell; an antibody/antigen; a nucleic acid; or whole slices of mammalian or plant tissues, etc. These sensing elements are responsible for the recognition of the analyte in the matrix and provide the selectivity and specificity of the final device [2-4]. Biosensors possess several unique features, such as compact size, simplicity of use, ideally one-step reagentless analysis, absence of radioactivity, etc., that make them very attractive alternatives to conventional biological sensing techniques [2]. Specifically, the features of high sensitivity and specificity have made biosensors promising candidates in the application fields of biomedical science, healthcare, agriculture, industrial process control, environmental monitoring, defense and security [5-13]. 1.2 Transducer aspects of biosensors According to the detection principle of transducers, currently employed biosensors can be divided into different types: electrochemical biosensors; optical biosensors; piezoelectric biosensors; magnetic biosensors; and, mechanical biosensors. 1.2.1 Electrochemical biosensors Electrochemical transducers were historically incorporated into the first enzyme-electrode biosensors and constitute the largest area of current research and commercial activity [14]. Of these transducers, the most commonly reported sensors are potentiometric, amperometric, or conductiometric in nature [2,14-16]. In addition to these classical electrode transducers, recent technological progress in this area has resulted in a variety of miniaturized solid state devices and innovative techniques such as Ion Sensitive Field-Effect Transistors (ISFETs) [15,17] and InterDigitated Electrodes (IDE, based on impedance measurement) [18]. 1.2.1.1 Ion Sensitive Field-Effect Transistor (ISFET) The conventional ISFET is a silicon-based MOS transistor, but the metallic gate of the MOSFET (Figure 2) is omitted and replaced by a reference electrode and a solution with certain pH. The transistor monitors the sensing events occurring at the electrolyte/dielectric interface through the field-effect. The main advantage of modified FET biosensors is the potential for producing

2

Chapter 1: Introduction

a multi-sensor chip incorporating an array of biologically sensitive devices with fully integrated signal processing and analysis circuitry. It enables the simultaneous monitoring of a number of analytes or metabolites [2].

Figure 2 the schematic illustration of MOS field effect transistor.

1.2.2 Optical biosensors Various optical methods have been exploited in biosensors including luminescence and fluorescence spectroscopy [19, 20], interferometry (reflectometric white light interferometry [21] and modal interferometry in optical waveguide structures [22]), spectroscopy of guided modes of optical waveguides (grating coupler [23] and resonant mirror [24]), and surface plasmon resonance (SPR) [25, 26]. Sensors based on optical interferometers, grating coupler, resonant mirror, and SPR measure the binding-induced refractive index changes. Compared to biosensors relying on the measurement of luminescence or fluorescence, sensors basing on measuring changes of refractive index provide a key advantage of carrying out detection without the need of extra labeling, thus avoiding the laborious labeling and multi-steps detection protocols. At the same time, conventional labels, such as enzyme, fluorescent probe, or gold colloids, can still be combined with these label-free technologies in order to improve the sensitivities. Among these label-free technologies, the SPR biosensor is particularly attractive due to the successful commercial SPR instrument developed by Biacore [27].

3

Chapter 1: Introduction

1.2.2.1 Surface Plasmon Resonance (SPR) SPR is based on the transfer of light energy (photons) to a group of electrons on a metal surface [27]. Gold is the preferred metal as it is compatible with a number of surface modification methods and will not be oxidized over time. A standard SPR set up is based on “Kretchman” configuration as shown in Figure 3.

Figure 3 the schematic illustration of “Kretchman” configuration of the SPR set-up.

It is based on total internal reflection in a glass prism onto which a thin film, e.g. gold, is deposited. Part of the light, the so-called evanescent field, penetrates outside the glass and the metal when total internal reflection occurs. The penetrated evanescent field sets up a surface plasmon, i.e. a wave that propagates along the interface between the metal and ambient medium. Under a specific angle of the incident light (SPR angle), most of the incident light is transferred into the surface plasmon wave, i.e. plasmon resonance, and a minimum in the reflected light intensity is observed. The change in the index of refraction in the ambient medium close to the sensor surface will induce a change of the SPR angle. As the evanescent wave has a short penetration depth, around several hundreds nanometers, processes occurring in the bulk ambient medium will have little influence on the SPR angle. Therefore, the SPR angle change can be used for real-time monitoring of the sensing events occurring at the sensor surface.

4

Chapter 1: Introduction

1.2.3 Piezoelectric biosensors Piezoelectric sensors typically use piezoelectric materials such as quartz which vibrate at specific frequencies depending on the sensor configuration and electric oscillator frequency. The simplest and most widely used piezoelectric sensor is the Quartz Crystal Microbalance (QCM) [28]. The AT-cut quartz crystal is coated with a pair of electrodes on both sides (Figure 4). An Alternating Current (AC) voltage is applied across the quartz to make the quartz crystal vibrate at a certain frequency, i.e. resonance frequency, normally between 1 to 10MHz. The resonance frequency is sensitive to the mass deposited on the sensor surface [28]. The mass deposited onto the sensor surface can be related to the shift of resonance frequency with the classic Sauerbrey equation under simple condition [28]. A more advanced QCM set-up combines dissipation measurement (QCM-D) at the same time. In QCM-D measurement, both the shift of vibrating frequency and the damping of the vibration are monitored. Thus not only the mass of the adlayer deposited on the sensor surface but also the visco-elastic property of the adlayer can be monitored in real time during a QCM-D measurement.

N Figure 4 the schematic illustration of a QCM set-up

The Surface Acoustic Wave (SAW) sensor differs from the QCM sensor by the layout of the electrodes and the operation frequency (Figure 5). The SAW quartz is vibrated at a higher range of frequencies, thus yielding higher sensitivities [29, 30]. In a SAW sensor, two sets of interdigitated electrodes are patterned on both sides of an AT-cut quartz substrate to set up input Interdigital Transducers (IDTs) and output IDTs. A guiding layer, normally SiO2, is growth on top of the quartz together with IDTs. A Shear-Horizontal wave is propagated through the guiding layer from the input IDTs to output IDTs. The interactions

5

Chapter 1: Introduction

between the acoustic wave, the mass and the elastic stiffness of the deposited adlayer, and the electric/dielectric properties of the guiding layer give rise to a sensing response.

input/output interdigitated electrodes

quartz Figure 5 the schematic illustration of a SAW set-up.

1.2.4 Magnetic biosensors Over the past several years, magneto-electronics has emerged as a promising new platform technology for biosensor development. The techniques are based on the detection of the magnetic fringe field of a magnetically labeled analyte interacting with a biological sensing element bound to a magnetic field sensor [31] (Figure 6). This detection platform provides a robust, inexpensive sensing technique with high sensitivity and considerable scope for quantitative signal data in comparison with fluorescence-based detection platform.

signal

“Spin-valve” sensors Figure 6 the schematic illustration of a spin-valve magnetic sensor.

6

Chapter 1: Introduction

1.2.5 Mechanical biosensors The merging of silicon micro-fabrication techniques offers new opportunities in developing microscopic devices, e.g. microcantilever, for biosensor applications. A microcantilever-based sensor relies on the mechanism that the cantilever bends upon the adsorption of the molecules onto the sensor surface [32] . In fact, the microcantilevers can transduce a number of different signal domains, e.g. mass, temperature, heat, electromagnetic field, and stress, into a mechanical deformation: either a bending or a change in the resonance frequency. Hence, a versatile detection protocols can be established on the microcantilever-based biosensors [33, 34]. 1.3 Biochemical aspects of biosensors From application point of view, a more popular classification of biosensors will be based on the biological sensing elements incorporated on the sensor surface, such as enzyme-based biosensors [4, 6, 11, 16], antibody-based biosensors (immunosensor) [4, 7, 10, 38-40], DNA-based biosensors [7, 35-37], and cells-based biosensors [4, 41-45]. Among them, the immunosensor, DNA-based biosensors and cells-based biosensors are specifically interests for this PhD research. Therefore we will give a brief review of the state-of-the- art of these three types of biosensors. 1.3.1 Immunosensors Immunosensors are a type of affinity biosensors. They are based on the binding interactions between an immobilized biomolecule (antibody/antigen) on the transducer surface and the analyte of interest (corresponding antigen/antibody), resulting in a detectable signal [4, 10, 40]. The sensor system takes advantage of high selectivity provided by the molecular recognition characteristics of an antibody, which binds reversibly with a specific antigen. The antigen encountered in immunosensor test can be versatile molecules, including hormones, drugs, bacteria, clinical disease markers, and environmental pollutants such as pesticides. 1.3.1.1 Antibody and its properties Antibodies are specialized proteins capable of recognizing antigens with high specificity. The quality of the antibody employed greatly influences the

7

Chapter 1: Introduction

specificity and sensitivity of an immunosensor. Polyclonal antibodies (pAbs) consist of a wide variety of antibody molecules of different specificities and affinities. The use of pAbs in immunosensors sometimes results in appreciated cross-reactivity and decreased sensitivity and reproducibility due to the intrinsic heterogeneous properties of pAbs [10]. Thanks to the hybridoma technology, an excellent alternative, monoclonal antibodies (mAbs) can be produced with constant characteristics. In addition, mAbs of desired affinities can be selected. Recombinant DNA technology has also been exploited for the engineering of the antibodies, which enables to further improve the sensitivity and reproducibility of the immunosensors. Virtually any (bio)chemical compound can serve as an antigen if it is able to trigger an immune response. Antibodies against the large molecules such as proteins are generated by immunizing an animal with an injection of the protein (antigen), while antibodies against the small chemicals (haptens) are generated by conjugating the hapten with an immunogenic substance before immunizing. The resulting immunological response observed in a mammal immunized with a particular antigen is a cascade of different classes of antibodies with typically increasing affinity for antigen [7]. The development of molecular biotechnology further allows generating antibodies with enhanced affinity and quantities without involving the immune system of an animal and the laborious culture and selection of hybridomas. In theory, the type of an analyte (antigen) that can be detected by the immunosensors are infinite, since there is always possibility to generate a specific antibody against the exist antigen. It implicates that the potential application fields of the immunosensors are probably limitless. An antibody is distorted Y-shaped immunoglobulin made up of four polypeptide chains linked by disulfide bridges (Figure 7). The two heavy chains (H) are identical to each other, as are the two light chains (L). The molecule can be cleaved by papain to produce two fragments. The fragment containing the part of the molecule with the arms of the Y shape that binds the antigen is called Fab (fragment antigen binding, or antigen-binding fragment). The remaining fragment is designated as Fc because it is readily crystallizable. Both L and H chains consist of a number of subunits or domains, some of which are similar (i.e. constant domains) whereas others are variable (i.e. variable domains). The L chains have one constant and one variable domain (CL and VL), whereas the H chains have one variable and three or four constant domains (VH; CH1; CH2; etc.) [46, 47].

8

Chapter 1: Introduction

Figure 7 the schematic illustration of the components of a antibody.

Some variable domains on the L and H chains fold differently. The close proximity of these differently folding domains, three from the VL and three from the VH chains on each arm of the Y-shaped immunoglobulin, are clustered in space to form the antigen-combining site (ACS) with its six hypervariable loops forming the complementarity determining regions (CDRs). The CDRs determine the specific chemistry, nature and shape of the antigenic determinant, i.e. the contact surface or epitope on the antigen that binds to this site. The contact surface on the antibody at which the epitope of an antigen docks is called a paratope. The paratope is part of an area of the surface conformation which is known as the idiotope. The three-dimensional shape of the epitope and its complementary paratope influences the interaction between the antigen and antibody. However, it is the complementarity of the electron cloud and the overall configuration of the outer electrons that govern the epitope/paratope interaction, and not the chemical nature of constituent residues of the epitope and its complementary paratope. It is now accepted that the `induced fit’ model constitutes the general mechanism for antigen-antibody recognition. The mutual or `induced fit’ model suggests that both antigen and antibody change their conformation upon complexation. There are significant structural changes at the binding site when binding of the antigen- antibody occurs, e.g. small side-chain rearrangements, segmental and relatively large movements of CDRs, or alterations in the relative disposition and rotation of variable light and heavy chains. The magnitude of this inducibility ultimately affects the specificity of antigen-antibody interaction: the better the fit between the epitope and the paratope the stronger the non-covalent bonds formed. The

9

Chapter 1: Introduction

strength of these interactions is referred as affinity if the binding sites are homogenous (e.g. monoclonal antibody) and avidity if the antigen binds to a variety of paratopes (e.g. polyclonal antibodies). 1.3.1.2 The antibody-antigen interactions In solution phase, antibody molecule (Ab) interacts specifically and reversibly with an antigen (Ag) to form an immune-complex (Ab-Ag) according to the following equilibrium equation: Ka / Kd Ab + Ag ← → Ab − Ag (1.1)

where Ka and Kd are the association and dissociation rate constant, respectively. The equilibrium constant (or the affinity constant) of the reaction is expressed as following: K=

K a [ Ab − Ag ] = K d [ Ab][ Ag ]

(1.2)

The equilibrium kinetics of antibody binding with antigen in solution suggests that both association and dissociation are relative rapid. The direction of the equilibrium depends on the overall affinity, which is basically the summation of both the attractive and repulsive non-covalent forces between the Ab and Ag. An immuno-complex usually shows low Kd values (in the range of 10-6 to 10-12 M) and also displays high affinity, i.e. K value, (typically 104). However, the Ab and Ag interactions in the immunosensors normally do not show the same kinetics as those established for the reaction in solution phase, since the immobilization of Ab (or Ag) onto the solid surface can either alter the properties of the biomolecules (Ab or Ag) or the accessibility of the analytes ( Ag or Ab) to the immobilized biomolecules. 1.3.1.3 The assay formats The detection protocol of an immunosensor can be adapted depending on the properties of the analytes and the transducers as well as the desired performance: such as sensitivity, detection speed, and cost. Generally, there are three kinds of assay formats, including direct assay, sandwich assay, and indirect competition assay (Figure 8).

10

Chapter 1: Introduction

(a) direct assay

(b) sandwich assay

(c) competition assay

(d) inhibition assay

antibody

antigen

tracer

Figure 8 the schematic illustration of different assay formats.

In a direct assay format, the Ab or Ag is directly immobilized on the transducer surface, and the affinity binding of the analyte of interest (Ag or Ab) generates a measurable signal on the transducers (Figure 8a). The generated signal is directly proportional to the concentration of the analyte. A major advantage of the direct assay is its simplicity. However, it also suffers from several limitations: due to the moderate sensitivity of the transducers, only analytes with large Molecular weight (Mw), which can generate sufficient signal during affinity binding, are preferably to be detected; the nonspecific interactions between the analytes to the transducer surface and the immobilized biomolecules cannot be easily distinguished from the specific Ab-Ag interaction, thus resulting in possible false signals in the test. The interference of nonspecific interactions in immunosensor can be eliminated by applying a sandwich assay format, where the analye is “sandwiched” between the surface immobilized primary antibody and a secondary antibody (Figure 8b). Since the primary antibody and secondary antibody interact with different epitopes on the analye, it requires the analyte presenting at least two distinct epitopes. The binding of the secondary antibody not only reduces the background signal induced by nonspecific interaction, but also enhances the sensing signal, i.e. improving sensitivity. On the other hand, the detection time is prolonged and complexity in the sandwich assay will be increased due to the extra secondary antibody binding step.

11

Chapter 1: Introduction

Indirect competitive assay is most popularly exploited for the detection of low Mw analyte. There are two varied formats in such assay: competition assay and inhibition assay. In the competition assay, the antibody is immobilized on the surface, and the analyte and its corresponding tracer molecule competes for the available binding sites on the immobilized antibody (Figure 8c). In contrast, in inhibition assay, the tracer molecule (antigen) is immobilized on the surface and the free analyte in solution competes with the available binding sites on the antibody in the solutions (Figure 8d). For the indirect competitive assay, the sensor signal is inversely proportional to the concentration of the analyte. Thus the lowest concentration of the analyte is measured as a maximal change of the sensing signal, which sometimes gives rise to the relative noise level in the test. Obviously, the required test time and complexity remain drawbacks for the indirect assay format. In addition, it takes extra cost and time to prepare the appropriate tracer molecules before applying the indirect competitive assays. 1.3.1.4 The limitation of immunosensors Cross-reactivity of an antibody to unexpected analyte, i.e. interference, is a major problem in the immunosensor test [10, 46]. Depending on the antigen (or conjugated hapten) used for immunization and the composition of the mixture samples under investigation, cross-reactivity of the antibodies with the unexpected analyte, which is similar to the antigen, are frequently observed [10]. The polyclonal antibodies with a variety of specificities normally give rise to a considerable interference during immunosensors test. However, even monoclonal antibody presenting mono-specificity has been shown to recognize more than one epitope. Moreover, if the test samples are pool of human serum, the interference is readily expected due to the existence of anti-(antibody) antibody, which is generated by part of the normal immune responds known as “idiotypic network” [46]. These interference effects always complicate the accurate measurement. Another general problem with the immunosensors is the difficulty to reversibly regenerate the sensing surface for continuous monitoring, or for repeated usage. However, in practice, adequate analytical sensitivity can only be achieved if antibodies with increased affinity (>1010 M-1) are used [48]. Therefore, a high-affinity constant and a labile immobilized antibody sensitive to the harsh conditions are often required, which make regeneration of the surface difficult. Currently available regeneration strategies, e.g. treating surface with acid, base, or detergent, either results in the partial loss and

12

Chapter 1: Introduction

denaturation of immobilized antibody/antigen, or leads to incomplete regeneration [4]. In addition, the measurable physical-chemical signal change during direct antibody-antigen binding is very small. Under a lot of circumstances, the binding event is “visualized” via an auxiliary reaction, i.e. using a label with the antibody or antigen [49]. It is thus normally associated with complications of multi-reagent or multi-step protocols in order to achieve high sensitive detection [4, 48, 49]. Moreover, factors, such as oriented immobilization, bioactivity of the immobilized antibody (or antigen), steric hindrance between the immobilized antibody (or antigen), and the nonspecific interaction, are crucial in determining the performance of the immunosensor. We will discuss them later (interface section) in this chapter. 1.3.2 DNA-based sensors Nucleic acid based biosensors for the identification of bacteria, viruses, and their products (antigen and toxins) play an increasing important role in applications such as clinical diagnostics, food, agriculture and environmental monitoring [7, 50]. Nucleic acid biosensors are based on the process of hybridization: the matching of one stand of DNA with its complementary stand [48] (Figure 9). These biosensors are also known as “gene chips”. Since the hybridization is dependent upon the formation of stable hydrogen bonds between the two nucleic acid strands, it contrasts with interactions of the antibody-antigen complex formation where hydrophobic, ionic and hydrogen bonds play a role. The bonding between nucleic acids takes place at regular intervals (nucleotide) along the length of the nucleic acid duplex, whereas the antibody-antigen bonds occur only at a few specific sites (epitopes). The frequency of bonding is reflected by the higher association constant for a nucleic acid duplex in comparison with an antibody-antigen complex and indicates that highly specific and sensitive detection systems can be developed using nucleic acid probes [50]. The specificity of nucleic acid probes relies on the ability of different nucleotides to form bonds only with an appropriate counterpart, i.e. the Waton-Crick rules of base pairing. Virtually, any self-replicating biological entity can be discriminated on the basis of nucleic acid sequences unique to that particular organism. However, as a prelude to the development of DNA-base biosensors, specific sequences

13

Chapter 1: Introduction

unique to the interesting organism have to been identified first. Since the nucleic acid recognition layers are very stable, an important advantage of nucleic acid ligands as immobilized sensors is that they can easily be denatured to reverse binding and then be regenerated simply by controlling buffer-ion concentrations.

T

T G A C T

G A C T C G C C G C C C G A T A T C G T A T A C G

C C

G C

G C

G

C A A C T T

G T T

G A A C

G

modulated current

© KdK2003

point mutation

modulated current

© KdK2003

Figure 9 a schematic illustration of DNA sensing on a electronic transducer.

The DNA-based biosensors can be broadly classified into two categories: direct target probing with/without signal amplification and target amplification. In the target amplification protocol, the quantity of the target DNA is firstly amplified by Polymerase Chain Reaction (PCR) before the target probing. Direct target probing is faster and better in quantifying the target compared to PCR, but it has limited sensitivity due to the normally considerable low amount of target DNA exist in the samples. In general, most DNA based detection systems require at least 105-106 targets. The detection of DNA-based sensors can be significantly improved by PCR processes which effectively increase the amount of the target DNA in the sample. In theory, the DNA-based sensor should be able to detect a single interested organism by first generating enormous copies of the specific target nucleic acid unique to the organism. However, the greatest challenge in the development of DNA-based biosensors

14

Chapter 1: Introduction

is to make them totally inclusive and thus make them recognize all potential biological variants. By spatially arranging DNA probes on the solid surface, a DNA microarray can be created. The advantage of a DNA microarray is that it is able to simultaneously detect multi-analytes. Spotting is most popularly used to place a series of DNA probes on the solid surface to form a DNA array. Photolithography has been combined with in situ DNA synthesis to achieve micro-addressable DNA arrays, which is so-called light-directed chemical synthesis. The photosensitive linker molecules, which include amino acids or hydroxyl-photoprotected deoxynucleosides, are covalently attached to silicon wafers. When pinpoint selected areas are illuminated through a photolithographic mask, the linker molecules are photo-deprotected. Consequently, the linker is reacted with nucleotide in solution to form oligonucleotide probe, but only at the positions exposed to light [12]. Another method of constructing micro-addressable array is the directional placement of off-chip-synthesized oligonucleotide probes by electrostatic attraction. A dense array of electrodes is fabricated on the chip. A positive electrical bias applied to the electrodes attracts negatively charged oligonucleotide probes to a precise location on the chip surface [12]. For signal generation, the probe can be labeled with a variety of reporter molecules including radioisotopes, fluorophores, enzymes, or haptens (to which antibodies are available). Different transducers, including optical, piezoelectric, and electrochemical, have been used to monitor the hybridization. In addition, direct, i.e. label-free, monitoring of the hybridization have been demonstrated using evanescent wave technology, piezoelectric acoustic wave devices, and electrochemical impedance spectroscopy [48, 50-51]. As the PCR renders the DNA sensors with extremely high sensitivity, it also brings serious limitations [13]. Firstly, the PCR requires a clean sample, since the compounds in the complex sample matrices will inhibit the PCR or decrease its efficiency. For this reaction, target analytes usually must be isolated or purified from the complex sample matrices prior to analysis, which results in labor-intensive and time-consuming work; Secondly, the contaminating or carryover DNA has the chance to be amplified also during PCR, resulting in false-positive signals. Concerning the limitation of the PCR, direct target detection without the PCR is preferred. However, indirect signal amplification has to be introduced to enhance the sensitivity of the direct target detection, which sometimes also involves labor-intensive and time-consuming

15

Chapter 1: Introduction

procedures and leads to poor reproducibility [52]. Thus, a critical need exists for a sophisticated, ultrasensitive, and quantitative detection approach without “amplification” step. 1.3.3 cells-based sensors Cells-based biosensors are devices that contain living biological cells. They monitor physiological changes of living cells induced by exposure to environmental perturbations such as toxicants, pathogens, or other agents [44]. 1.3.3.1 cell-based detection Biological cells are small membrane-bounded structures, which contain a high concentration of chemicals including enzymes, nucleic acids, ions, many different types of proteins, small organic molecules and many others [4]. As a complex system presenting specific functionalities, whole cells serves as a kind of sophisticated bioreceptors that can be used in the development of biosensors [53] . This kind of so-called cells-based sensors use the intrinsic response of a specific cell type to a potential toxic or infectious foreign substance to identify a target analyte [13]. A variety of types of cells have been used to construct biosensors, including microorganisms (bacteria and fungi), mammalian cells (lymphocytes, endothelial cells, hepatocytes, neurons, and myocyte) [13, 41, 43, 44, 53]. Different kinds of response, such as cells metabolism (growth inhibition, cell viability, substrate uptake), cells’ bioluminescence, and action potential, have been monitored as the signal for target analyte detection. In fact, the cell-based biosensors are composed of two transducers. A primary transducer is the cell that convert the presence of a bioactive substance into a cellular signal, whereas a second transducer is a physical device that converts the resulting cellular signal into an electrical signal that can be processed and analyzed [44]. Both the optical or electrochemical methods have been exploited to achieve signal transduction. The most commonly used optical method is to monitor the intracellular calcium transient with fluorescent labels. The Ca2+ ions are involved for instance in the contraction of muscles, release of neurotransmitters, metabolic processes of the cell by the activation of biochemical signal cascades or the activity of ion pumps [3]. Fluorescent labels which are derived from the well-known Ca2+ chelating molecules. The binding of Ca2+ with the fluorescent label changes the fluorescent properties. In this

16

Chapter 1: Introduction

way, information about the intracellular calcium level may be obtained from the fluorescence intensity of the label. Electrochemical methods are frequently used to directly record intra- and extra-cellular electrical signals from cells in different analytes. In addition, the products, such as oxygen, glucose, lactate, and different ions, released from metabolism or cellular activities upon analyte stimulation can also be followed by the electrochemical methods [3-4]. The potential advantages to be gained by exploiting whole cells in a sensor format include [41, 43, 53, 54]: (1) unlike other biosensors such as DNA- or antibody-based sensors, cells-based biosensors are not specific for certain compounds but are capable of responding to a wide range of biological active compounds and offer the potential to gather greater information content than biomolecular-based sensors; (2) cells-based biosensors are able to simultaneously detect large numbers of possible health or environmental threats, especially unknown or unanticipated ones. The molecular biosensors (DNA sensors and antibody-based sensors) are generally designed to respond only to certain target molecules and have moderate sensitivity, but these characteristics limit their ability to detect relevant molecules with similar function. Cells-based sensor overcomes the limitations and may detect unexpected molecules; (3) cells-based biosensors can detect the analytical information quantitatively. They verify not only the presence but also the concentration of the substance; (4) the more important fact is that cells-based biosensors are able to measure the functional information. For example, cell-based biosensors can be used to analyze the effect of the pharmaceutical compound on a giving physiological system. Sometimes, people make cells-based biosensors as detector to study the mechanism of action of second messenger; (5) cells-based biosensors can be used to predict human performance decrements caused by low level of agents or synergistic effects of environmental toxicants. They are less complex than whole organism, but still provide detailed, physiologically relevant information. At the same time, the cost of growing cells in culture is low. For evaluating toxicology and efficacy of potential pharmaceuticals, the cell-based sensor can substitute for the use of live animals, which is expensive and cumbersome; (6) since cell organelles are essentially closed systems, they can be used over long periods of time with good reproducibility; (7) the cell lines can be genetically engineered and selected to produce desired properties and levels of sensitivity to respond to versatile target analytes. Despite the attractive features presented by the cells-based biosensors system, there are many complex problems when living cells were treated as the sensing elements, including the selection, the culture and the maintenance of living

17

Chapter 1: Introduction

cells. The stability of biological cells is of critical importance for biosensor applications. In general, cells derived from higher organisms are more difficult to isolate and maintain in a viable form for extended periods of time. The most biologically robust cells are derived from less complex organisms, particularly yeast, bacteria and algae. Thus they enable effectively development of biosensors incorporating a truly “living” biological component, but natural cell differentiation and growth characteristics of such whole cell biosensors can cause problem associated with signal drift over a period of time. Carefully choice of operating conditions, such as pH, temperature and light, or application of chemical growth inhibitors can reduce this problem [4]. Moreover, the coupling of living cells and the physical transducer constitutes one of biggest challenges. In addition, some kind of cells-based sensors also subject to the same kind of cross-reactivity problems as other antibody-based technologies. 1.3.3.2 cells-based sensor using excitable cells Excitable cells, such as neural cells, myocytes, are evoked action potential when subjected to stimulus. For those cells, an electrical voltage, or potential, always exists between the inside and outside of the cell membranes. The voltage of an inactive cell stays at a negative value (inside relative to outside the cell), and the membrane is said to be polarised. The potential difference across the membrane at rest is called the resting potential and this is about -70 mV. The establishment of this potential difference is due to: (1) negatively charged organic ions do not pass across the membrane and so remain trapped inside the cells; (2) sodium ions are actively transported out of the cells by sodium-potassium pumps (specialised carrier proteins); (3) potassium ions are actively transported into the cells by sodium-potassium pumps; (4) three sodium ions move out for every two potassium ions that are pumped in; the increased movement outward of the sodium compared with the potassium inward leads to more positive ions outside the cells than in cells; (5) the membrane is 100 times more permeable to potassium ions, which therefore diffuse back out of the cells faster than the sodium ones diffuse back in. This further increases the potential difference between the negative inside and the positive outside of the cells; (6) apart from the chemical gradient that causes the movement of the potassium and sodium ions, there is also an electrical gradient. As more and more potassium ions diffuse out of the neurone, so the outside becomes more and more positive. Further outward movement of potassium ions is therefore made difficult because, being positively charged,

18

Chapter 1: Introduction

they are attracted back into the cells by its overall negative state and repelled from moving outwards by the overall positive state out of the cells. Under the above conditions, an equilibrium is established whereby there is no net movement of ions and there is a balance between the chemical and electrical gradients. When a stimulus arrives, its energy causes a temporary reversal of the charges on the cell surface membrane. As a result, the cell will undergo (or "fire") an action potential, and in this condition the membrane is said to be depolarised (often called a "spike"). This depolarisation occurs because ion channels (normally called voltage-gated ion channels) in the membrane change shape, and hence open or close, depending on the voltage across the membrane. The energy of the stimulus causes the sodium voltage-gated channels in the cell surface membrane to open and therefore sodium ions diffuse in through the channels along their electrochemical gradient. Being positively charged, they begin a reversal in the potential difference across the membrane. As sodium ions enter, so more sodium channels open, causing an even greater influx of sodium ions. As a result, the negative charge of 70 mV inside the membrane becomes a positive charge of around +40 mV. This is known as the action potential. Once the action potential of around +40 mV has been established, the voltage gates on sodium channels close (so further influx of sodium is prevented) and the voltage gates on the potassium channels begin to open. With some potassium voltage-gated channels now open, the electrical gradient that was preventing further outward movement of potassium ions is now reversed, causing more potassium channels to open. This means that yet more potassium ions diffuse out, causing repolarisation of the cells. The outward movement of these potassium ions causes the temporary overshoot of the electrical gradient, with the inside of the neuron being more negative (relative to the outside) than usual. This is called hyperpolarisation. The gates on the potassium channels now close and the activities of the sodium-potassium pumps cause sodium ions to be pumped out and potassium ions in, once again. The resting potential of -70 mV is re-established and the neurone is said to be repolarised. The action potential does not dwell in one location of the cell's membrane, but travels along the membrane. Taking neuron cells as example, the action potential can travel along an axon for long distances. Action potentials carry fast internal messages between tissues, and are an essential feature of animal life. They are used most extensively by the nervous system to send messages between nerve cells and from nerve cells to other body tissues such as muscles and glands. Action potentials are measured with the recording techniques of electrophysiology. The action potential activities of excitable cells depend on

19

Chapter 1: Introduction

the cellular functional information. The ability to monitor the activity of these cells in a noninvasive manner, i.e. extracellular recording, makes the neurons and cardiomyocytes particular useful in constructing cells-based biosensors. Over the past 20 years, “neuronal biosensors” have been intensively studied due to the specific properties of the neuron cells and the important biological role of neuron network (Figure 10).

Figure 10 (left) an illustration of a neuron cell; (right) an illustration of the major elements in a prototypical synapse.

It has been found that neuronal cells bind specifically and sensitively with odors, drugs, and toxins. With the binding of those substances, they generate electric signals in a substance-specific and concentration-dependent manner, and the response profiles can be monitored by electrochemical transducers [15, 44] . Thereby, electrochemical transducer coupled with neurons network serves as an efficient biosensor to quantitatively identify the target analyte through the generated signal patterns upon the stimulation of different substances. In addition, the neurons networks underlie memory storage and information processing in the human brain, and ultimately participate in what Eccles referred as “the creation of consciousness”. Recording the response of the neuron networks upon different stimulation and collecting the information about the mechanism of the neuron network communication can eventually provide invaluable information for the diagnosis and treatment of the neuron related disease [56]. In addition, the information are specifically useful in the attempt of constructing “brain-computer interfacing” [57].

20

Chapter 1: Introduction

Three categories of chemical compounds are known to influence the electrophysiological behavior of the neuron network. They are called “neuroactive compounds”. The first category are compounds that cause substance-dependent major change in the spontaneous native activity pattern of the neuron networks, i.e. increase, decrease or stop the activity. The compounds include all synaptically active agents (e.g. glutamine, strychnine, N-methyl D-aspartic acid) and metabolic poisons. Thus the change of the active pattern serves as detection for the above neuroactive compounds. The second category are compounds that cause substance-dependent changes in network oscillations. The oscillation is the condition where all the bursts have the same shape and most channels are synchronized. This ideal condition was achieved through disinhibition, i.e. via the blocking of the inhibitory GABA and glycine synapses with bicuculline and strychnine, respectively. It is known that synaptically and metabolically active substance, ion channel blockers and toxins will alter the regularized, oscillatory activity of the disinhibited neuron network, and that such substance- and concentration-specific response may be used as primary readout of a neuronal biosensors. The third category are compounds that cause substance-dependent paroxysmal response. For example, the gp120 protein of AIDS virus produces massive, unique paroxysmal discharges that may last as long as 2 min. Hence, this specific paroxysmal response can be used to indicate the presence of relative neuroactive compounds. Basically, the neuronal biosensors detect the neuroactive compounds depending on the alteration of the regular electrophysiological behaviors [44]. As for the signal recording, planar metal electrodes are widely used [45, 58]. An array of metal electrodes is fabricated on the insulating substrates, such as glass, using lithographic processes. The electrodes are open to the solution containing the cells. The electrodes detect local variation of the electrical potential that spans the membranes of cells in their vicinity. The sources of the recorded signals are thus those compartments of neuronal cells rich in ion channels, such as axon hillocks [55]. Despite the great perspective of “neuronal biosensor”, there are considerable challenges remained to be overcome before going to real application. For example, precisely aligning the neuron network with the geometric pattern of sensing spots on the transducer is a big obstacle for achieving efficiently stimulating and recording of neuron cells [45, 54, 58]. Moreover, the confining of the desired neuron network pattern for long term monitoring obtained only limited progress. In addition, in traditional 2D culture, neuronal survival and

21

Chapter 1: Introduction

viability are often compromised and the system might be limited in its ability to accurately predict the behavior of the same cell types in vivo, new strategy for 3D culture is expected to obtain neuron network more comparable with that in vivo [43]. 1.4 The interface aspects The basic requirement of a biosensor is that the biological sensing elements should be brought in close proximity of a transducer and thus the physico-chemical changes can be efficiently “sensed”. In this direction, the interface layer that links the biological sensing elements and the transducer plays a major role. Obviously, only when the biological sensing elements can be attached onto the transducer and stay there during the course of the test, will the biosensor be able to sensitively specifically, and reproducibly detect the target analytes. The interface layer serves to anchor and retain, normally called “immobilize”, the biological sensing elements on the transducer. From this sense, the interface layer has crucial importance. However, the role of the interface layer is more than “immobilization”, it will also influence the characteristics of the biosensors, e.g. sensitivity, specificity, stability, and reproducibility, which will be discussed in chapter 2. 1.5 Key attributes of a biosensor Biosensors should meet or exceed certain requirements so as to make them comparable or even better than the traditional analytical systems. They must be simple to handle, small, cheap, and able to provide reliable information in real-time. They also need to be sensitive and selective for the analyte of interest, and suitable for in-situ monitoring. Sometimes, they even need to be able to detect multi-analytes either simultaneously or with good temporal/spatial resolution [59]. Biosensors are unique and innovative tools that may provide complementary data and information compared with more “classical” analytical instrumentation. Their uniqueness relies clearly on the presence of an immobilized biological sensing element closely connected to a physical transducer. The biological sensing element of a biosensor interacts selectively with the target analytes, assuring the selectivity of the sensor. The transducer converts the biological response, resulting from the interaction between the biological sensing element and the target analyte, into a quantifiable signal. The major attributes of a good biosensor are its specificity (or selectivity),

22

Chapter 1: Introduction

sensitivity, stability and reproducibility. The specificity (or selectivity) of the biosensor for the target analyte is mainly determined by the biological sensing element because that is where the target analyte interacts with the sensor. In comparison with chemical sensor, an inherent advantage that can be exploited in biosensor is the significantly higher specificity that can generally be achieved as a direct result of biologically-optimized molecular recognition [4]. That is best typified by the antibody-antigen interaction, where an antibody can recognize and bind its antigen with extremely high specificity. Minor modification of the molecular structure of the antigen can dramatically lower its affinity for the original antibody. The sensitivity of the biosensor and limit of detection (LOD) are a function of the molecular recognition element and the physical design, i.e. the sensitivity of the biosensor is dependent on both the biological sensing element and the transducer. That is because there must be a significant biological sensing element-analyte interaction and a high efficiency of subsequent detection of this reaction by the transducer. The required sensitivity for a particular analyte is determined by the concentration levels found in the measurement environment of interest [59]. In this respect, the LOD for a particular biosensor will depend on the intended application. For pharmacological or physiological analytes, the concentration ranges of the target analytes are fairly well defined around the nanomolar to micromolar levels. For food or environmental applications, the LOD will be set by regulatory levels. These levels can be rather low, for example, the EEC has set regulations for pesticides in drinking water at 0.1 g/L [14]. There is a common bottleneck in the development of biosensors, which is the insufficient stability and reproducibility. The intrinsic instability of biological sensing element after isolation significantly reduces the life-time of the biosensor as well as the storage time of the product. In addition, the deterioration of the performance of the transducer as well as the loss of the biological sensing elements during the test are another source of instability. The reproducibility is influenced by the stability. All the above instable factors will give rise to the poor reproducibility. Meanwhile, the controllability on the construction of biosensor, including transducer processing and biological sensing elements loading (qualitatively and quantitatively), will also affect the reproducibility of the biosensor. Another important issue, which is frequently overlooked, is the interference

23

Chapter 1: Introduction

during the test. The interference will occur at different levels. It may generate false positive signal on the transducer, whereas it may also lead to background noise signal. It may deteriorate the performance of transducer, furthermore it may even destroy the activity of the biological sensing element. There are several sources responsible for the generation of the interference. For example, in the enzyme based biosensor, the existence of non-target electroactive species frequently interferes with the measurement of target electroactive species [59]. While in the immunosensor, the cross-reactivity between the antibody and antigen-similar non-target can interferes with the recognition of target antigen. A major source of the interference is the nonspecific adsorption of target or non-target biological species on the sensor surface. The nonspecific adsorption occurs not only on transducer surface but also on immobilized biological sensing elements with non-optimal orientation. 1.6 Summary The biosensors have attracted intensive interests due to their unique properties including fastness of response, compact size, simplicity of use, ideally one-step reagentless analysis. They are expected to be promising candidates in a variety of applications such as biomedical science, healthcare, agriculture, industrial process control, environmental monitoring, defense and security. However, significant efforts should be invested to improve the performance of the biosensors in order to meet the requirement and challenge in real applications. Collaboration between chemists, biochemists, immunologists, pharmacologists, biologists, and engineer is required in order to obtain stable, isolated, and immobilized biological sensing elements that possess the desirable characteristics of biosensing, at the same time to achieve advances in both instrumentation and detection procedure for the optimal detection of the binding event. When designing a biosensor, it is important to understand the multiple factors that will eventually influence the performance of the sensing system. For example, the properties of an interface layer can dramatically alter the performance of the biosensor (see chapter 2). The major interest of this PhD study is to establish and optimize some novel interface layers and eventually to use these interface layers for constructing high performance biosensor.

24

Chapter 1: Introduction

Reference: [1] Biosensors and their Applications, Yang, V.C.; Ngo, T.T., Ed.; Kluwer: New York, 2000 [2] Sethi, R.S. Biosens. Bioelectron. 1994, 9, 243-264 [3] Gopel, W.; Heiduschka, P. Biosens. Bioelectron. 1995, 10, 853-883 [4] Byfield, M.P.; Abuknesha, R.A. Biosens. Bioelectron. 1994, 9, 373-400 [5] Taylor, R.F., Chemical and biological sensors: markers and commercialization, In: Handbook of Chemical and Biological Sensors; Taylor, R.F., Schultz, J.S., Ed.; IOP publishing: Philadelphia, 1996 [6] Karube, I.; Nomura, Y. J. Mol. Catal. B-Enzym. 2000, 10, 177-181 [7] Iqbal, S.S.; Mayo, M.W.; Bruno, J.G.; Bronk, B.V., Batt, C.A.; Chambers, J.P. Biosens. Bioelectron. 2000, 15, 549-578 [8] Giardi, M.T.; Koblizek, M.; Masojidek, J. Biosens. Bioelectron. 2001, 16, 1027-1033 [9] Mulchandani, A.; Chem, W.; Mulchandani, P.; Wang, J.; Rogers, K.R. Biosens. Bioelectron. 2001, 16, 225-230 [10] Suri, C.R.; Raje, M.; Varshney, G.C. Crit. Rev. Biotechnol. 2002, 22, 15-32 [11] Wilson, G.B.; Hu, Y. Chem. Rev. 2000, 100, 2693-2704 [12] McGlennen, R.C. Clin. Chem. 2001, 47, 393-402 [13] Lim, D.V.; Simpson, J.M., Kearns, E.A.; Kramer, F. Clin. Microbiol. Rev. 2005, 18, 583-607 [14] Rogers, K.R. Mol. Biotechnol. 2000, 14, 109-129 [15] Leech, D. Chem. Soc. Rev. 1994, 23, 205-213 [16] Chaubey, A.; Malhora, B.D. Biosens. Bioelectron. 2002, 7, 441-456 [17] Bartic, C., Organic-Based Field-Effect Devices as Transducers for Bioanalytical Applications, Ph.D Thesis, K.U.Leuven, Leuven, Belgium, 2002 [18] Bonroy, K., Optimization of Biosensor Interfaces towards the Direction of Low Molecular Weight Compounds, Ph.D Thesis, K.U.Leuven, Leuven, Belgium, 2005 [19] Gamiz-Gracia, L.; Garcia-Campana, A.M.; Soto-Chinchilla, J.J.; Heurtas-Perez, J.F.; Gonzalez-Casado, A. TRAC-trends in analytical chemistry 2005, 24, 927-942 [20] Rowe-Taitt, C.A.; Hazzard, J.W.; Hoffman, K.E.; Cras, J.J.; Golden, J.P.; Ligler, F.S. Biosens. Bioelectron. 2000, 15, 579–589 [21] Piehler, J.; Brecht, A.; Gauglitz, G. Anal. Chem. 1996, 68, 139–143 [22] Heideman, R.G.; Kooyman, R.P.H.; Greve, J. Sens. Act. B 1993, 10, 209–217 [23] Clerc, D.; Lukosz, W. Sens. Act. B 1994, 19, 581–586

25

Chapter 1: Introduction

[24] Cush, R.; Cronin, J.M.; Stewart, W.J.; Maule, C.H.; Molloy, J., Goddard, N.J. Biosens. Bioelectron. 1993, 8, 347–353 [25] Homola, J.; Yee, S.; Gauglitz, G. Sens. Act. B 1999, 54, 3–15 [26] Homola, J.; Yee, S.; Myszka, D. Surface plasmon biosensors. In: Optical biosensors: present and future; Ligler, F.S., Taitt, C.R., Ed.; Elsevier, 2002 [27] Cooper, M.A. Anal. Bioanal. Chem. 2003, 377, 834-842 [28] Marx, K.A.; Biomacromolecules 2003, 4, 1099-1120 [29] Du, J.; Harding, G.L.; Ogilvy, J.A.; Dencher, P.R., Lake, M. Sens. Act. A 1996, 56, 211-219 [30] Gizeli, E.; Goddard, N.J.; Lowe, C.R.; Stevenson, A.C. Sens. Actuators B-Chem. 1992, 6, 131-137 [31] Graham, D.L.; Ferreira, H.A.; Freias, P.P. Trends Biotechnol. 2004, 22, 455-462 [32] Raiteri, R.; Grattarola, M.; Butt, H.; Skladal, P. Sens. Actuator B-Chem. 2001, 79, 115-126 [33] Hansen, K.M.; Thunder, T. Methods 2005, 37, 57-64 [34] Ziegler, C. Anal. Bioanal. Chem. 2004, 379, 946-959 [35] Hahn, S.; Mergenthaler, S.; Zimmermann, B.; Holzgreve, W. Bioelectrochemistry 2005, 67, 151-154 [36] Kuswandi, B.; Tombelli, S.; Marazza, G.; Mascini, M. Chimia 2005, 59, 236-242 [37] Piunno, P.A.E.; Krull, U.J. Anal. Bioanal. Chem. 2005, 381, 1004-1011 [38] Franek, M.; Hruska, K. Vet. Med-Czech 2005, 50, 1-10 [39] Lin, J.H.; Ju, H.X. Biosens. Bioelectron. 2005, 20, 1461-1470 [40] Scheller, F.W.; Wollenberger, U.; Warsinke, A.; Lisdat, F. Curr. Opin. Biotechnol. 2001, 12, 35-40 [41] Durick, K.; Negulescu, P. Biosens. Bioelectron. 2001, 16, 587-592 [42] D’Souza, S.F. Bisens. Bioelectron. 2001, 16, 373-353 [43] Stenger, D.A.; Gross, G.W.; Keefer, E.W.; Shaffer, K.M.; Andreadis, J.D.; Ma, W.; Pancrazio, J.J. Trends Biotechnol. 2001, 19, 304-309 [44] Park, T.H.; Shuler, M.L. Biotechnol. Prog. 2003, 19, 243-253 [45] Jenkner, M.; Tartagni, M.; Hierlamann, A.; Thewes, R. IEEE J. Solid-State Circuits 2004, 39, 2431-2437 [46] Ismail, A.A.A.; Walker, P.L.; Cawood, M.L.; Barth J.H. Ann. Clin. Biochem. 2002, 39, 366-373 [47] Kuby, J; Immunology, 3rd ed, W.H. Freeman and Company, New York 1997 [48] D’orazio, P. Clin. Chim. Acta 2003, 334, 41-69 [49] Diaz-Gonzalez, M.; Gonzlez-Garcia, M.B.; Costa-Garcia, A. Electroanal.

26

Chapter 1: Introduction

2005, 17, 1901-1918 [50] Ivnitski, D.; Abdel-Hamid, I.; Atanasov, P.; Wilkins, E. Biosens. Bioelectron. 1999, 14, 599-624 [51] Tombelli, S.; Minunni, M.; Luzi, E.; Mascini, M. Anal. Lett. 2004, 37, 1037-1052 [52] De Paul, S.M.; Falconnet, D.; Pasche, S.; Textor, M.; Abel, A.P.; Kauffmann, E.; Liedtke, R.; Ehrat, M. Anal. Chem. 2005, 77, 5831-5838 [53] Vo-Dinh, T.; Cullum, B. Fresenius J. Anal. Chem. 2000, 366, 540-551 [54] Wang, P.; Xu G.; Qin, L.; Xu, Y.; Li, Y.; Li, R. Sens. Actuator B-Chem. 2005, 108, 576-584 [55] Kandel, E.R.; Schwartz, J.H.; Jessell, T.M. Principles of Neural Science, 4th ed, McGraw-Hill, New York, 2000 [56] Fernandez, E.; Pelayo, F.; Ahnelt, P.; Ammermuller, J.; Normann, R.A. Restor. Neurol. Neurosci. in press [57] Fromherz, P. Neuroelectronic interfacing: Semiconductor Chips with Ion Channels, Nerve Cells, and Brain, In: Nanoelectronics and information Technology; Waser, R., Ed.; Wiley-VCH: Berlin, 2003 [58] Morin, F.O.; Takamura, Y.; Tamiya, E. J. Biosci. Bioeng. 2005, 100, 131-143 [59] Wilson, G.S.; Gifford, R. Biosens. Bioelectron. 2005, 20, 2388-2403

27

Chapter 1: Introduction

28

Chapter 2: Design of a high quality interface layer 2.1 Introduction We have introduced in Chapter 1 the different aspects of a biosensor, including the physical transducer, biological sensing elements, and interface layer. Apparently, the performance of a biosensor will be determined by the collective properties of these components. Among them, the properties of the interface layer play an important role. Both the sensitivity and reliability of the biosensor will be ultimately dictated by the properties of the interface layer. 2.2 The role of the interface layer As a linker layer, the interface layer interacts not only with the biological sensing elements but also with the physical transducer. However, due to the specific properties of the biological sensing elements and the rather complex interaction between the sensing elements and the interface layer, the effect of the interface layer on the transducer aspect is frequently overlooked. It has been proven that both the stability and the sensitivity of the transducer device will also be influenced by the interface layer. First, the interface layer shields the sensing area of the transducer from the attack of harmful compounds present in the sensing environment. For instance, in a silver-coated QCM based biosensor, the biologically activated silver electrodes cannot retain long-term stability and reactivity due to the continuous oxidation progress. Similarly, in a SAW device, the aluminum electrodes show high chemical instability in buffers containing corrosive electrolytes. Shielding

Chapter 2: Design of high quality interface layer

polymers, e.g. polyimide and polystyrene, etc., have been found to be able to protect QCM and SAW devices from undesirable corrosive/oxidation, and in addition being used for biological sensing elements immobilization [1]. For FET devices used in aqueous environment, the degradation of the gate insulator resulting from hydrolysis and adsorption of unknown species limits their application [2]. The protection of the gate area with a self-assembled monolayer showed efficiency in improving the stability of the transistor [3]. Secondly, the physical properties of the interface layer can decrease or increase the sensitivity of the transducer. For example, the SPR transducer works basing on the interaction between the surface plasmon wave and the analytes. However, the surface plasmon only propagates several hundred nanometers above the surface, and the sensitivity of the SPR device decreases exponentially with the distance between the analyte and the surface [4]. Thus, a thinner layer will be preferred in surface modification, which will bring the interaction between the analyte and sensing element closer to the surface. Similarly, the FET transistor senses the charges over a distance of the order of 1 nm, and thus operation of these devices relies on having the molecules of interest bind as closely as possible to the semiconductor surface [5]. Obviously, a thinner interface layer will result in less interference on the sensitivity of the transducer. Besides the thickness, other properties of the interface layer will also affect the sensitivity of the transducer. For instance, a polymer with conductive properties is preferably used in an amperometric biosensor, because the conductive interface layer will facilitate the electron transfer on the surface, thus improving the sensitivity of the transducer [6-8]. In addition, it is well known that the transportation of analytes towards the sensor surface is controlled by the properties, such as thickness and permeability, of the interface layer. It will eventually influence the response time and the linear range of a transducer [2, 6, 8]. As mentioned previously, the crucial role of the interface layer is justified by its ability to immobilize the biological sensing elements onto the sensor surface. Most importantly, it has been recognized that the affinity, specificity, and stability of the biological sensing elements are directly related to the characteristics of the immobilization. Significant efforts have been exerted to achieve “optimal” immobilization of biological sensing elements on the sensor surface, since the affinity, specificity, and stability of the biological sensing elements will finally determine the sensitivity, specificity and stability of the biosensor.

30

Chapter 2: Design of high quality interface layer

In summary, the interface layer and the immobilization of biological sensing elements on the interface layer will affect the properties of the biological sensing elements as well as the performance of biosensors in the following senses: (1) It has been found that the direct adsorption of biological sensing elements on the unmodified transducer surface resulted in denaturation of the biological sensing elements as well as irreproducible binding. For unmodified transducer surfaces, there are no functional groups available to specifically interact with biological sensing elements. The main forces attracting the biological sensing elements onto the sensor surface are hydrophobic interaction and electro-static interaction [9]. It is well known that this kind of direct adsorption results in the conformational change of the biological materials. Since the bioactivity of the biological sensing element is determined by its native conformation, a conformation change, i.e. denaturation, normally lead to the loss or decrease of the bioactivity of the biological sensing elements. Therefore, it is necessary to introduce an interface layer, which can mimic the properties of the natural environment of the biological sensing elements, to avoid or reduce denature of the sensing elements. Meanwhile, for the direct adsorption of biological sensing elements on the unmodified sensor surface, the amount of biological elements binding and the degree of activity loss will vary greatly depending on the experimental conditions. That is because a lot of experimental conditions, such as pretreatment of surface, ion concentration, pH value, etc., will affect the interaction between the unmodified surface and the biological sensing elements. These variant will result in poor reproducibility. Hence, a well presented interface layer is desired not only to retain the bioactivity of the biological sensing elements but also to achieve a quantitatively controllable binding of the biological sensing elements. (2) From the application point of view, the biological sensing elements should retain on the sensor surface during the course of the test. Specifically, for long-term or continuous measurements, the retention of biological sensing elements on the sensor surface is essential for attaining a stable and reproducible signal. A major goal in introducing an interface layer is to obtain strong interaction, e.g. covalent binding or other high affinity linking, between the surface and the biological sensing elements [6]. (3) For the biosensors, the biological sensing elements have to be partially or wholly isolated from the biological source in order to integrate these

31

Chapter 2: Design of high quality interface layer

sensing elements onto the transducer surface. However, those isolated biological sensing elements suffer greatly from the inherent instability. Taking enzyme and membrane proteins for example, the isolated enzyme will lose activity in hours or several days, while the isolated membrane proteins will even totally lose their functionality immediately after placing on the unmodified transducer. Studies have shown that the stability of the isolated enzyme can be greatly improved by immobilizing it in a suitable interface layer [2, 10]. Similarly, membrane protein based biosensor has been successfully constructed through immobilizing the membrane protein in a suitable interface layer, such as an artificial lipid membrane [11]. (4) Because the binding of an analyte to a biological sensing element is stoichiometric in nature, and because there is a finite sensor surface area available, the density of immobilized biological sensing elements is critical for the sensitivity of the biosensor. It has been shown that the amount of immobilized biological sensing elements can be dramatically increased by using a suitable interface layer [12-13], and consequently the sensitivity of the resulted biosensor is greatly enhanced [13]. Whatever the amount of the immobilized biological sensing elements, these immobilized elements should still have enough degree of freedom to freely interact with their specific ligand in the environment. It requires sufficient spatial separation among immobilized biological sensing elements in order to secure the easy access of target analyte. In another word, the steric hindrance between the binding sites of the biological sensing elements should be limited. Obviously, for an unmodified sensor surface, neither a high amount of immobilization nor a controllable spatial separation of the biological sensing elements can be obtained. Introducing an interface layer with delicately tunable properties presented a promising way to fulfill those requirements [13]. (5) Since only the “binding sites” on the biological sensing elements specifically interact with the target analyte, the attachment of the biological sensing elements onto the surface should not shield the “binding sites” from the analyte. In other words, the biological sensing elements should be attached onto the surface with good orientation. Under most circumstance, the direct adsorption of biological sensing elements onto an unmodified surface can not give controllable orientation due to the intrinsically complex and, sometime, un-predictable surface properties. Thus, introducing an interface layer with well-controllable surface properties is essential to guarantee a controllable immobilization of the biological sensing elements.

32

Chapter 2: Design of high quality interface layer

(6) Moreover, the growing and increasing use of miniaturized sensor, DNA and protein chips, and cells-based sensors requires reproducible, precise, and localized deposition of the biological sensing elements on pre-defined areas [6, 14-17] . The most efficient way to meet this requirement is using an interface layer with well-defined surface properties (chemical cue) or topography (topographic cue). These chemical and topographic cues can be exploited to selectively interact with biological sensing elements and localize the attachment of on desired areas. In conclusion, the interface layer is exploited in order to attach the biological sensing elements onto the sensor surface with fully retained activity and stability, and additionally in a quantitatively, spatially and geometrically controllable manner. In comparison with the considerable concerns about the role of the interface layer in the immobilization of biological sensing elements, the value of the interface layer in preventing non-specific interference is often overlooked. Indeed, a major problem associated with biosensors is the high degree of nonspecific binding to the sensor surface that occurs when the sensor is presented with real (and therefore multi-component) samples, particularly whole serum, which alone contains many thousands of proteins [18-19]. The nonspecific adsorption will lead to high background signal, i.e. noise level, decreasing sensitivity of the biosensor [7, 18]. Sometimes the nonspecific adsorption will also generate false positive signal. Moreover, the “fouling” of the sensor, i.e. the nonspecific adsorption of the biological materials on the sensor surface, will also result in the failure of the whole sensor [8, 21]. Since the unmodified sensor surface will interact with all the biological materials in a nonspecific manner due to the intrinsic hydrophobic and static electric interaction, the sensor surface has to be covered by an interface layer to prevent this kind of nonspecific interaction. Of course, the interface layer itself should be able to prevent nonspecific adsorption, which can be readily obtained by control the properties of the interface layer. One may argue that if the sensor surface was fully covered by the biological sensing elements, this biological sensing layer can serve as barrier to prevent nonspecific adsorption, thus the interface layer is not necessary anymore. However, an ordered and dense packing of biological sensing elements on the unmodified sensor surface is hardly available due to the very complex and unpredictable properties of the unmodified sensor surface. Even if such kind of packing was available, the small size biological materials can still penetrate the barrier layer of biological sensing elements and adsorb nonspecifically onto the sensor surface. In

33

Chapter 2: Design of high quality interface layer

addition, the direct adsorption of biological sensing elements onto the unmodified sensor surface always lead to certain denaturation of the biological materials as mentioned above, while the denatured biological sensing elements themselves will become the target for nonspecific adsorption. From this sense, the interface layer will prevent nonspecific adsorption not only by shielding the sensor surfaces but also by decreasing the degree of denature of the biological sensing elements. Indeed, not only the denatured biological sensing elements will result in nonspecific adsorption but the wrong orientation of the biological sensing element will allow nonspecific adsorption as well. For example, it is well know the F(ab) portion of the whole antibody almost exclusively interacts with the specific ligand (antigen), but the Fc portion of the whole antibody will interact considerably with a lot of nonspecific biological compounds [18]. Considering the antibody was attached on the surface with an orientation of most Fc potions pointing into solution, a significant amount of nonspecific adsorption on the antibody layer itself will be expected. From this aspect, the interface layer can further prevent nonspecific adsorption by controlling the orientation of the biological sensing elements. 2.3 The immobilization of biological sensing elements in interface layer A variety of approaches have been exploited for immobilizing biological sensing elements onto the transducers, including: 1) Physical adsorption: physical adsorption [22-27] onto a surface of substrate can attach biological materials for a single, non-repeatable measurement. This method of protein immobilization depends on the non-specific interaction of the proteins with the substrate. These interactions include various non-covalent bridges, such as ionic and hydrogen bonds, hydrophobic interactions and Van Der Waals forces. The number and extent of these bridges largely depend on the chemical properties of the solid surface and the surface characteristics of the proteins, as determined by the solvent. This method is the easiest and quickest way to immobilize proteins onto a substrate, however this form of binding cannot achieve oriented immobilization of proteins and a serious limitation of this method is the leaching of immobilized proteins during the measurement procedure [22-23]. 2) Entrapment: one of the most popular methods of immobilization is entrapment in a polymer matrix or membrane [28-35]. The pores of the matrix are large enough to allow the diffusion of the test species and

34

Chapter 2: Design of high quality interface layer

product, but small enough to prevent the loss of the active, high molecular weight biological material. Nevertheless, the oriented immobilization cannot be achieved by this method either and this method also suffers from the problem of the leakage of the entrapped proteins during measurement. 3) Covalent binding: covalent binding provides a longer-term stable immobilization of proteins compared to the above two physical immobilization methods. Chemical groups (-OH, -COOH, -NH2, -SH) on the biological components, which are not essential for biological activity, may be attached covalently to chemically activated supports [36-49] . Bifunctional agents that induce intermolecular chemical cross-linking can bind biological materials to solid supports [50-51]. A potential risk of this kind of method is that sometimes the activity of the biological sensing elements will be decreased or altered due to the probably harsh treatment condition [22, 52]. 4) Affinity binding: biological binding proteins such as protein A or protein G or the avidin-biotin system have been used to immobilize the biological sensing elements [7, 53-55]. The advantage of using this kind of affinity interaction for immobilization is that the binding will have the least effect on the native activity of the biological sensing elements, and at the same time facilitate the orientated immobilization. A major benefit of oriented immobilization is that the binding sites on the sensing elements are not interfered by the immobilization condition and at the same time maximal degree of freedom is retained to interact freely with target analyte [53]. To achieve oriented immobilization of proteins, a special site, which is not used for molecular recognition, of the biologically active protein should be chosen for immobilization. A biospecific complex formation for oriented immobilization of biologically active proteins was used first in 1978 with covalently attached protein A, which has the unique capacity to bind the Fc portion of mammalian immunoglobulin G (IgG) [56-58]. Janis and Regnier used the other bacterial Fc binding protein-protein G- instead of Protein A, which has the advantage of binding to a wider range of IgG species and subclasses [59-61]. Meanwhile, antibodies against the Fc portion of the IgG can also be used for oriented immobilization after first attaching the former antibody onto the sensor surface [53]. The carbohydrate moieties in the glycoproteins are often used as binding sites for oriented immobilization. For example, the carbohydrate moieties in antibody can be oxidized with periodate

35

Chapter 2: Design of high quality interface layer

to form aldehyde groups, which are then chemically bound to hydrazide-activated sensor surface [55]. The histidine-tagged biological sensing elements can be immobilized onto the surface in an oriented way though metal complex. The metal cation, such as Ni2+, is first bound to the chelator nitrilotriacetic acid present on the sensor surface, and then the histidine tag on the biological sensing elements is further chelated with the metal cation [53]. The biotin-avidin system is frequently exploited to achieve oriented immobilization, where the biotin labeled biological sensing elements specifically bind to avidin functionalized sensor surface through high affinity biotin-avidin interaction [53]. In addition, F(ab’) fragment containing free sulfhydryl can be immobilized onto the sensor surface in an oriented way using the free sulfhydryl as anchor groups [18]. However, the reduction of whole antibody to its F(ab’) fragment will sometimes harm the native activity of the antibody. To overcome this drawback, site-directed mutagenesis strategy has been explored to attach anchor groups on to a predefined position, located away from the specific binding site, on the biological sensing elements. Consequently, the biological sensing elements will bind onto sensor surface using the introduced anchor groups while leaving the binding site freely accessible to the target analytes [53]. 2.4 Current available interface layers Different approaches have been exploited to construct interface layers for biosensor applications. All the interface layers have their specific advantages and show their merits in specific biosensor applications. However, by now, there is no universal interface layer that is suitable for all kinds of biosensors architectures. Generally, there are four types of interface layers: Self-assembled Monolayers (SAMs), polymer layer, Lipid Bilayers (LBs), and Sol-gel system. 2.4.1 Self-assembled Monolayers Self-assembled Monolayer (SAM) formation is induced by the strong chemisorption between the substrate and head groups of selected organic molecules. The attractive feature of self-assembly is that the uncorrelated molecules will spontaneous form an organized superstructure with no need for external intervention. SAMs provides one of the most elegant approaches towards making ultrathin organic films of controlled thickness. Frequently used SAMs are alkanethiolates on Au and silanes on various oxide surfaces. The properties, such as easy and fast formation, and readily controllable surface

36

Chapter 2: Design of high quality interface layer

properties, make SAMs promising for biosensor applications [62-66]. Normally, the monolayer can be tailored with versatile functional terminal groups for immobilization purposes. The covalent binding of biological materials to these SAMs will result in a biolayer that can be used for biosensor applications. The functional alkane thiols that have been used for biosensors applications include: carboxy functional thiols [67-68], pre-actived carboxy functional thiols [69-70], amine functional thiols [71-72], aldehyde functional thiols [73], biotin functional thiols [74-78], maleimide functional thiols [79], and nitrilotriacetic acid functional thiols [80]. On the other hand, versatile functional silanes are also used for biosensor applications, such as amine functional silane [81-82], epoxy functional silane [82] and mercapto functional silane [82]. Indeed, chemical modifications to produce a self-assembled monolayer offer a number of important advantages: the diversity to generate a wide variety of surfaces (hydrophobic/hydrophilic) via the incorporation of different groups into the alkyl chain and chain termini of a SAM; a wide variety of chemical manipulation procedures which result in the formation of a covalent bond between the protein and the substrate surface under non-denaturing conditions; the possibility of controlling the density and the environment of the immobilized species, i.e. mixed SAMs; the generation of uniform structures; higher coverage of the substrate surface; and, the reduction of the number of possible random orientations that the proteins can have on the surface. These advantages make the SAMs the most popular interface layer for biosensor applications. However, SAMs also suffer from some limitations. An important limitation of alkane thiol systems lies in the relatively poor stability of chemisorbed alkanethiolates. It has been shown that under ambient conditions substantial oxidation and subsequent loss of stability of alkanethiol SAMs takes place within days and the integrity of the adlayer is readily compromised [83-87]. To control the surface density of the biological sensing elements, mixed SAMs, which consist of binary thiols with different end functional groups, have been frequently used. Nevertheless, phase separation of the binary thiols are sometimes observed, which damage the quality of the monolayer and lose the control of the density of the biological sensing elements [75, 80, 88, 89]. As for silane, the stability is not a problem, but the ability to obtain a homogeneous monolayer with good reproducibility remains a challenge due to the intrinsic risk of intermolecular polymerization. Moreover, there is no readily available approach to form “good” mixed silanes. Attempts to control the functional group density either consist of complex surface modification procedures or result in non-uniform molecular structure [90]. In addition, due to the intrinsic

37

Chapter 2: Design of high quality interface layer

2D nature of the SAMs, the binding capacity of biological sensing elements is limited in comparison with 3D interface layer, while a higher amount loading of biological sensing elements are sometimes beneficial for achieving higher sensitivity of the biosensors [91]. 2.4.2 Polymer interface A major advantage of a polymer interface is the improved mechanical and chemical stability in comparison with SAMs. A broaden type of polymers has been used for biosensor applications. Generally, the polymers can be categorized as conducing polymer and non-conducting polymer according to their native properties. Conducting polymers, such as polypyrrole, polythiophene, and polyaniline, are capable of conducting electronic charge. Thus they are incorporated in electrochemical transducer based biosensors to facilitate electron transfer from the biocomponent to the electrode surface [8, 17, 22]. The biological sensing elements are immobilized in the conducing polymer films either by entrapment or by covalent binding. A limitation of a conducting film is that it may interact with chemical species or oxygen to produce deleterious effects [17]. Non-conducting polymer is incorporated into biosensors to prevent interferences, to prevent fouling of the electrodes surface by proteins or other biological substances, and to immobilize the biological sensing elements. A lot of non-conducting polymers, such as polyphenylenediamines [8], polyphenols [17] , agarose [14], collagen [14], polycarbonate [52], polyphthalate [52], polyvinyl alcohol [52], polyacrylonitrile [52], alginate [52], chitosan [52], dextran [92], polystyrene [93-94], polyethyleneimine [95-97], polyurethane [98], poly(dimethyl siloxane) [99], polyethylene [100], poly(2-hydroxyethyl methacrylate) [101-102], polyethersulfonate [103], and polyacrylamide [104-105] have been used as interface for biosensor construction. In most cases, direct entrapment approach is used to immobilize the biological sensing elements. However, the entrapment method suffers from the potential problem of leaking of biological sensing elements. Therefore covalent binding is preferred for stable immobilization. The polymers are either pre-derived with functional binding groups or post-treated to introduce functional entities. Many conventional polymer surface modification techniques are available, such as plasma treatment [106-109], ion implantation [110], photochemical immobilization [111-114] and graft polymerization [115-121], to post-treat the polymer film in order to generate suitable binding groups for the biological sensing elements. Because it is not

38

Chapter 2: Design of high quality interface layer

always feasible to pre-deriving the polymer with functional groups, and sometimes the pre-derive procedure involves labor-intensive and time-cost steps, the post-treatment approaches turn to be faster and more straightforward. However, a main drawback of these post-treatment approaches is the low efficiency of surface functionalization and poor reproducibility. Consequently, only limited amount of biological sensing elements can be immobilized, frequently in a quantitatively irreproducible manner. Several techniques are available to produce polymer films on the sensing area of the transducers, including spin-coating, electropolymerization, plasma-polymerization, grafting, and self-assembly. The spin-coating [8] is simplest and fastest, and at the same time allows control over the thickness, but the solvent residue sometimes is harmful for the biological elements. Both electropolymerization [62] and plasma-polymerization [122-123] utilize the specific treatment (electrochemical or plasma) to initiate the chain polymerization of monomers. The resulted polymer is deposited onto the substrate during the course of polymerization. The electropolymerization can only be used on conductive electrode surface, while the plasma-polymerization doesn’t have this limitation. However, both approaches have only been applied for limited species of polymers. Most importantly, the reproducibility of both approaches can only be realized in a moderate degree. The polymer films deposited onto the sensor surface through spin-coating, electropolymerization, and plasma-polymerization do not have covalent links with the beneath sensor surface. Thus the stability of the polymer films on the sensor surface is limited, especially under the circumstance where strong shear force exists. The grafting method can provide covalent binding between the surface and the polymer films. Nevertheless, the sensor surface has to be pre-treat to generate suitable anchor groups before the polymer can be grafted onto the anchor sites. It means the grafting method normally needs multi-steps surface modification. In addition, these multi-steps procedure leads to either low efficiency of grafting or poor reproducibility. Recently, the self-assembly principle has been applied to attach polymer film onto sensor surface. These methods are based on thiol-functional polymers [4, 124] or polyelectrolytes [125, 126]. Linear polymers with carboxy or aldehyde groups at one end are thiolated at the other end, and the resulting polymer can chemisorb onto gold surface. However, due to the large steric hindrance between the polymer chains, the packing density and the chain order in the resulted polymer film are rather poor. Meanwhile, the limited stability

39

Chapter 2: Design of high quality interface layer

associated with alkane thiol also exists for the thiol-functional polymers. For the polyelectrolytes based film, the electro-static interaction is the drive force to achieve self-assembly. The surface is first treated to introduce positive or negative charge, and then the polyanion or polycation molecules self-assemble onto the surface through the electro static adsorption. Multilayer of alternative polyanion/polycation films can be constructed onto the surface through layer by layer procedure, and the biological sensing elements are entrapped or covalently bound in the multilayers. A significant advantage of the multilayer of polyelectrolytes films is that a large amount of biological sensing elements can be immobilized, but the construction of the multi-layers involves time-cost procedure and sometimes results in poor-reproducibility [126]. A more simple procedure has been developed recently, where a polycation polymer (Figure 1) can directly self-assemble onto a series of oxide surfaces, such as Ta2O5, SiO2, TiO2, which present negative charge under physiology pH condition [170].

Figure 1 the schematic illustration of the polycation polymer.

This polylysine-grafted poly(ethylene glycol) copolymer is very promising for biosensor applications. The grafted PEG can be end functionalized with functional groups, such as biotin, and the biological sensing elements can bind onto the biotin moieties. The density of the bound biological elements can be readily tuned through the graft ratio of the functional PEG chains. Meanwhile, the grafted PEG chains can efficiently prevent nonspecific adsorption. Despite the attractive advantages of this copolymer, it suffers from the limited stability of electro-static interaction. Moreover, the degree of both negative charge on surface and positive charge on the polymer is influenced by the pH condition. Extremely lower or higher pH values will decrease the relative negative charge

40

Chapter 2: Design of high quality interface layer

and positive charge at the interface, as such destroying the interaction between the surface and the polymer. A new trend in using polymers for biosensor application is the use of dendrimers to construct a high performance biosensor. Dendrimers constitute a unique class of polymers that are distinguished from all other synthetic macromolecules by their globular shape resulting from their perfectly branched architecture and their mono-disperse nature. The size, molecular weight, and chemical functionality of dendrimers can be easily controlled through the synthetic methods used for their preparation both by divergent and by convergent methods. The highly branched dendritic architecture causes the molecules to adopt three-dimensional globular conformations to minimize their free energy, with an immediate effect on macromolecular properties [127-129]. These polymers, if grafted properly to a surface, will form well-ordered and uniform layers with the majority of terminal functional groups located at the surface layer and with molecules being in either globular or compressed globular conformation. The presence of multiple anchoring groups in the terminal branches chemically attached to one core makes these molecules unique in their multiple grafting abilities. One can expect the formation of the uniform surface layer with suppressed tendency toward microphase separation of dissimilar multifunctional arms due to chemical constraints imposed by the branched microstructure, research has demonstrated that about 40% of the surface area can be occupied by epoxy groups distributed within alkyl branches without any signs of microphase separation usually observed for two-component SAMs (mixed SAMs) [130]. Variation of chemical functionalities of the different branches can be an effective route to tailor surface properties of such layers without a risk of having heterogeneous surface. On the other hand, having molecules with higher molecular weight as a single building block, one can fabricate thick (3-10nm) layers, which can possess enhanced micromechanical properties characteristic of macromolecular materials rather than low molar mass organics. Yoon et al [131-134] have developed biofunctionalizable monolayers constructed with poly(amidoamine) dendrimers, whose surface chain-end groups were double-functionalized with biotinyl ligands and ferrocenyl groups for biospecific recognition and electron transfer reactions, respectively. The biosensor basing on these monolayers showed greatly enhanced sensitivity and stability. However, a limitation in their design is that the surface has to be activated with avidin groups first and the loading of dendrimer onto the surface is not well controllable. Due to the poor control of the dendrimer loading, the density of the functional groups for

41

Chapter 2: Design of high quality interface layer

biological sensing elements binding is also out of control. In summary, from the point of view of thermal and mechanical stability, the polymer materials are good candidates for biosensor application. However, as we discussed above, currently used technologies for constructing polymer films bear several shortcomings: 1) the density of the binding sites in the polymer film can not be easily controlled and sometimes, because of the low efficient surface functionalization, only very limited numbers of functional groups can be introduced onto the surface, and consequently the immobilization capacity of the resulting polymer film is quite low; 2) the interaction of the substrate and polymer film is not so strong, because normally the polymer film is formed by spin cast directly on the solid substrate. To enhance the interaction between the substrate and the polymer film, people have used covalent binding method to attach the polymer film to the solid substrate, but this means several extra modification steps will be required to achieve a stable functionalized polymer surface. Meanwhile, the multi-steps procedure gives rise to the chance of poor reproducibility; 3) the biggest drawback of the polymer film is that the film structure can not be controlled at molecular scale. The potential applications of biosensors devices could be expanded to novel horizons if biomolecules could be organized in ordered microstructure arrays on surfaces [135-136]. To achieve development of nano-technology in biosensor device, an elaborate control of the interface structure in molecular scale is required. The reason why SAM technology is widely used for surface modification lies in the fact that this technology provides a possibility to control the surface structure at a molecular scale. 2.4.3 Lipid bilayers Recent advances in supported bilayer membranes (SBMs) have exciting implications for bioanalytical research [137-138] because of their increasing capacity to mimic cell membranes, in which vital biological processes such as energy transduction, signaling, and transport occur [139]. They have been used as effective sensing platforms for the analysis of glucose [140], neurotransmitters [141] , artificial sweeteners [142], pharmaceutical drugs [143] and bacteria [144]. In addition, SBMs have been applied in the field of proteomics to investigate peptide and protein functions [145-149]. Supported membranes are held together by a delicate balance of electrostatic, hydration, and van der Waals forces [150], and it has been suggested that exposure to air and even shear flow may damage or destroy them [145, 151].

42

Chapter 2: Design of high quality interface layer

Therefore, a fundamental challenge in SBM design is to reconcile mechanical stability and lateral fluidity in one structure. Several approaches have been developed to improve the stability of the SBM layers: (1) hybrid bilayer membranes, consisting of a SAM underlayer and a single membrane leaflet on top has been constructed. They exhibit much higher stability than bilayer membranes on solid substrates; (2) polymerization [152] and “heat stabilization” [190] have been employed in SBM fabrication. However, significant reduction of lateral mobility is often observed in these membranes. Biological processes and binding events that require free movement of lipid constituents in the membrane plane can be unfavorably affected; (3) the development of micropatterned SAMs has made it possible to construct arrays of hydrophilic “wells” surrounded by hybrid bilayer walls. The bilayer “patches” assembled in these wells are fluid and have been shown to have improved stability [153]; (4) electrostatic forces were used to enhance surface function and improve membrane-substrate interactions, thereby strengthening membrane stability for biosensor applications [154]. Polyelectrolyte multilayers as a “cushion layer” for generating stable bilayer membranes have been intensively studied [155-156]. Hydrogels tailored with charged functional groups were found effective to increase membrane robustness in an ion channel device [157]. Despite the appreciated success of the above strategies, a robust SBM feasible for stable biosensor application still remains a challenge. 2.4.4 Sol-gel system The sol-gel method enables the powderless processing of glasses, ceramics and thin film, directly from solution [158]. Precursors are mixed at the molecular level and variously shaped materials may be formed at much lower temperature than it is possible by traditional methods of preparation. Since the inorganic polymerization process takes place in mild conditions, it allows the association of inorganic phase (glass, ceramics, silica) with biological system. The possibility to immobilized biological sensing elements without loss of their activities led to the development of biosensors [158-161]. One of the major advantages of sol-gel system is the wide variety of chemical compositions, dimensions and forms that can be achieved via sol-gel chemistry. The possibility of synthesizing hybrid organic-inorganic materials facilitates the design of new engineering materials with exciting properties for biosensor applications. For example, multifunctional “smart” devices can be achieved through the sol-gel method by combining biocompatibility, biological activity

43

Chapter 2: Design of high quality interface layer

and stimuli-response materials [160]. The drawback of sol-gel system is that the sol-gel layer is physical attached onto the sensor surface with limited stability, whilst the leakage of entrapped biological sensing elements may result in the loss of stability and reproducibility. 2.5 Task and proposal As we mentioned above, the properties of the interface layer will influence the performance of the biosensor from several aspects. Firstly, both the stability and the sensitivity of the transducer are influenced by the interface layer. The coverage of the transducer with the interface layer prevents the sensor surface from the attack of potential harmful species, but at the same time the property of the interface should be designed in a way that is compatible with the working mechanism of the transducer. If an electronic transducer is to be utilized, the direct electronic coupling of the biomolecular function units to the conducting electrodes is essential. Because of the spatial limitation due to electronic tunneling, this requires specific geometric arrangement with a control down to the nanometer scale. Consequently, an interface layer with limited thickness as well as well controlled molecular order will be preferred. Secondly, the interaction between the interface layer and the biological sensing elements is the most crucial factor in the development of biosensors. The stability of the interface layer and the binding manner of the biological sensing elements on the interface will first dictate how stable the biological sensing elements will be retained on the sensor surface during the course of the test. The property of the interface layer and the procedure of immobilization will again determine how the biological elements will retain their native activity. Finally, the characteristic of the immobilization, such as the amount of immobilized biological sensing element, the steric hindrance between the immobilized biological sensing elements, and the orientation of the immobilized biological sensing elements, will determine the affinity, and sensitivity of the binding events. Lastly, the nonspecific adsorption remains a big challenge for real biosensor applications, because the sensor has to resolve signal from background noise in complex sample such as blood. The concentration of certain target analyte under test in a mixture may be as low as 1 g/L, or even less, compared with a total background concentration of other biological materials of 70 g/L. An ideal

44

Chapter 2: Design of high quality interface layer

biosensor must therefore display a very high degree of discrimination, while the prevention or reduction of nonspecific adsorption will greatly help to enhance the discrimination as well as sensitivity. An interface layer that is able to prevent the nonspecific adsorption will provide added value in improving the performance of the biosensor. Regarding the important role of the interface layer, keeping in mind the limitations of the currently available interface layers (see section 2.3), we are aiming to develop novel interface layers that might cope with features such as: 1) Result in a stable film on the transducer surface; 2) Result in a film whose structure can be controlled at the molecular scale; 3) Result in a reproducible amount of protein immobilization on the surface. This includes: to control the density of anchored proteins to limit steric interactions of neighboring proteins and to ensure that the amount of immobilized protein is reproducible; 4) Maintain the activity of the antibody (such as orientation); 5) Form a 3-D like platform for protein immobilization; 6) Reduce the non-specific binding reactions, in order to enhance the sensitivity. Consequently, the scope of this PhD study will involve: 1) The investigation of interactions between biological species, such as proteins, and the interface layer: we have verified the important role of the interaction between the biological sensing elements and interface layer on the biosensor development. Briefly, the requirement to control the interaction between the biological sensing element and the interface layer is promoting the specific binding of biological sensing element, while at the same time preventing the nonspecific binding of interference species. In order to achieve such goals, the knowledge of the behavior of the biological species on interface layer is essential. Basing on the collected information of the behavior of biological species on interface layer, we can deduce useful guidance for future interface design. 2)

The development of an interface layer with excellent nonfouling properties: above, we have pointed out the importance of preventing nonspecific adsorption in biosensor development. We will not only show

45

Chapter 2: Design of high quality interface layer

the design of novel interface layer for preventing nonspecific adsorption, but also investigate the mechanism of preventing nonspecific adsorption. The mechanism will in turn serves as guidance to optimize the performance of the interface layer. 3)

The development of a novel interface layer for the immunosensor: the specific antibody-antigen interaction renders immunosensor with potential for a variety of applications. We are aiming to develop a robust interface layer with well controllable surface properties. Improvement of the performance of immunosensor is expected by optimizing the antibody immobilization on the novel interface layer.

4)

The development of surface engineering strategy for electronic DNA sensor: DNA microarray is attractive due to its high throughout screening ability. Electronic DNA sensor based on III-V transistor provides the potential for highly sensitive and label-free detection of DNA hybridization. In order to utilize the characteristics of III-V transistor for DNA sensor development, an approach that enables efficient DNA probes immobilization has to be developed. The density and distribution of probe molecules and the properties of the interface layer are of significance in terms of the behavior of the resultant biosensor, because these parameters determine the type and degree of nearest-neighbor interactions amongst immobilized probes and interactions between the sensor surface and probes. Such interactions further affect the efficiency, rate and selectivity of hybridization [162].

5) The development of a surface engineering strategy for hybrid neuron-transistor: we have described above the merit of cells-based biosensors. The important physiology role of neurons makes the development of hybrid neuron-transistor especially interesting for versatile biological and medical applications. A major challenge is to attach the neuronal network on the sensor surface will well-controlled geometric features. Intimate and precise alignment of neuron network on the sensor surface with well-defined network as well as with long term stability is the critical step in developing hybrid neuron biosensor. We will demonstrate that the performance of biosensors can be improved through the new interface layer with well-controlled organization and properties. We will show that for different applications (immunosensor, DNA sensor, cells-base sensor), good interface layers will share some common

46

Chapter 2: Design of high quality interface layer

features as described above, while at the same time the structure and property of the interface layer has to be optimized individually depending on the various natures of different transducers and different biological systems.

47

Chapter 2: Design of high quality interface layer

Reference: [1] Chong, K.T.; Su, X.; Lee, E.J.D.; O’Shea, S.J. Langmuir 2002, 18, 9932-9936 [2] Sethi, R.S. Biosens. Bioelectron. 1994, 9, 243-264 [3] Adlkofer, K.; Eck, W.; Grunze, M.; Tanaka, M. J. Phys. Chem. B 2003, 107, 587-591 [4] Uchida, K.; Otsuka, H.; Kaneko, M.; Kataoka, K.; Nagasaki, Y. Anal. Chem. 2005, 77, 1075-1080 [5] Clare, T.L.; Clare, B.H.; Nichole, B.M.; Abbott, N.L.; Hamers, R.J. Langmuir 2005, 21, 6344-6355 [6] Gopel, W.; Heiduschka, P. Biosens. Bioelectron. 1995, 10, 853-883 [7] Rogers, K.R. Mol. Biotechnol. 2000, 14, 109-129 [8] Pal, P.S.; Sarker, P. J. Indian Chem. Soc. 2002, 79, 211-218 [9] Voros, J. Biophys. J. 2004, 87, 553-561 [10] Karube, I.; Nomura, Y. J. Mol. Catal. B-Enzym. 2000, 10, 177-181 [11] Phillips, K.S.; Dong, Y.; Carter, D.; Cheng, Q. Anal. Chem. 2005, 77, 2910-2915 [12] Akkoyun, A.; Bilitewski, U. Biosens. Bioelectron. 2002, 00, 1-10 [13] Disley, D.M.; Morrill, P.R.; Sproule, K.; Lowe, C.R. Biosens. Bioelectron. 1999, 14, 481-493 [14] Stenger, D.A.; Gross, G.W.; Keefer, E.W.; Shaffer, K.M.; Andreadis, J.D.; Ma, W.; Pancrazio, J.J. Trends Biotechnol. 2001, 19, 304-309 [15] Park, T.H.; Shuler, M.L. Biotechnol. Prog. 2003, 19, 243-253 [16] Jenkner, M.; Tartagni, M.; Hierlamann, A.; Thewes, R. IEEE J. Solid-State Circuits 2004, 39, 2431-2437 [17] Cosnier, S. Electroanalysis 2005, 19, 1701-1715 [18] Byfield, M.P.; Abuknesha, R.A. Biosens. Bioelectron. 1994, 9, 373-400 [19] Lim, D.V.; Simpson, J.M., Kearns, E.A.; Kramer, F. Clin. Microbiol. Rev. 2005, 18, 583-607 [20] Poot, A. A.; Beuleling, T. In Modern Aspects of Protein Adsorption on Biomaterials; Missilis, Y. F., Lemm, W., Eds.; Kluwer: The Netherlands, 1991, pp 29-38 [21] Pasche, S.; Textor, M.; Meagher, L.; Spencer, N.D.; Griesser, H.J. Langmuir 2005, 21, 6508-6520 [22] Andreescu, S.; Sadik, O.A. Pure Appl. Chem. 2004, 76, 861-878 [23] Delgado, A.D.S.; Leonard, M.; Dellacherie, E Langmuir 2001, 17, 4386-4391 [24] Absolom, D.R.; Zingg, W.; Neumann, A.W. J. Biomed. Mater. Res., 1987, 21, 161-167

48

Chapter 2: Design of high quality interface layer

[25] Horisberger, M.; Vauthey, M. Histochemistry 1984, 80, 13-21 [26] Matsamura, H.; Kleijn, J.M. Colloid Surface B: Biointerfaces 1993, 1, 277-285 [27] Abad, J.M.; Pariente, F.; Hernandez, L.; Lorenzo, E. Anal. Chim. Acta 1998, 368, 183-189 [28] Dave, B.C.; Dunn, B.; Valentine, J.S.; Zink, J.I. Anal. Chem. 1994, 66, 1120A-1127 [29] Avnir, D.; Braun, S.; Lev, O.; Ottolenghi, M. Chem. Mater. 1994, 6, 1605-1612 [30] Ellerby, L.M.; Nishida, C.R.; Nishida, F.; Yamanaka, S.A.; Dunn, B.S.; Valentine, J.S. Science 1992, 255, 1113-1117 [31] Dave, B.C.; Soyez, H.; Miller, J.M.; Dunn, B.; Valentine, J.S.; Zink, J.I. Chem. Mater. 1995, 7, 1431-1440 [32] Yamanaka, S.A.; Dunn, B.; Valentine, J.S.; Zink, J.I. J. Am. Chem. Soc. 1995, 117, 9095-9103 [33] Chung, K.E.; Lan, E.H.; Davidson, M.H.; Dunn, B.S.; Valentine, J.S.; Zink, J.I. Anal. Chem. 1995, 67, 1505-1514 [34] Schuhmann, W.; Kranz, C.; Wohlschlager, H.; Strohmeier, J. Biosens. Bioelectron. 1997, 12, 1157-1166 [35] Foulds, N.C.; Lowe, C.R. J. Chem. Soc. Faraday Trans. 1986, 82, 1259-1267 [36] Yoshioka, M.; Mukai, Y. J. Chromatogr. 1991, 566, 361-369 [37] Pfeiffer, N.E.; Wylie, D.E.; Schuster, S.M. J. Immunol. Methods 1987, 97, 1-9; [38] Howell, S.; Kenmore, M.; Kirkland, M.; Badler, R.A. J. Mol. Recognit. 1998, 11, 200-203 [39] Johnsson, B.; Lofas, S.; Lindquist, G.; Edstrom, A.; Muller Hillgren, R.M.; Hansson, A. J. Mol. Recognit. 1995, 8, 125-131 [40] Wessa, T.; Rapp, M.; Ache, H.J. Biosens. Bioelectron. 1999, 14, 93-98 [41] Kupsu, S; Sleytr, U.B.; Sara, M. J. Immunol. Methods 1996, 196, 73-84 [42] Spiker, J.O.; Kang, K.A. Biotechnol. Bioeng. 1999, 66, 158-163 [43] Uttenthaler, E.; Kosslinger, C.; Drost, S. Biosens. Bioelectron. 1998, 13, 1279-1286 [44] Konig, B.; Gratzel, M. Anal. Chem.1994, 66, 341-344 [45] Spitznagel, T.M.; Jacobs, J.W.; Clark, D.S. Enzyme Microb. Technol. 1993, 15, 916-921 [46] Williams, R.A.; Blanch, H.W. Biosens. Bioelectron. 1994, 9, 159-167 [47] Sundarababu, G.; Gao, H.; Sigrist, H. Photochem. Photobiol. 1995, 61, 540-544

49

Chapter 2: Design of high quality interface layer

[48] Ben David, A.; Firer, M.A. Biotechnol. Tech. 1996, 10, 799-802 [49] Honda, T.; Miwatani, T.; Yabushita, Y.; Koike, N.; Okada, K. Clin. Diagn. Lab Immunol. 1995, 2, 177-181 [50] Sundarababu, G.; Gao, H.; Sigrist, H. Photochem. Photobiol. 1995, 61, 644-649 [51] Shriver Lake, L.C.; Donner, B.; Edelstein, R.; Breslin, K.; Bhatia, S.K.; Ligler, F.S. Biosens. Bioelectron. 1997, 12, 1101-1106 [52] D’Souza, S.F. Bisens. Bioelectron. 2001, 16, 373-353 [53] Turkova, J. J. Chromatogr. B, 1999, 722, 11-31 [54] Willner, I.; Katz, E. Angew. Chem. Int. Ed. 2000, 39, 1180-1218 [55] Nisnevitch, M.; Firer, M.A. J. Biochem. Biophys. Methods 2001, 49, 467-480 [56] Werner, S.; Machleidt, W. Eur. J. Biochem. 1978, 90, 99-103 [57] Gersten, D.M.; Marchalonis, J.J. J. Immunol. Methods 1978, 24, 305-311 [58] Langone, J. J. Immunol. Methods 1982, 55, 277-283 [59] Janis, L.J.; Regnier, F.E. Anal. Chem. 1989, 61, 1901-1906; [60] Cai, J.; Henion, J. Anal. Chem. 1996, 68, 72-76 [61] Morais, S.; Gonzales-Martinez, M.A.; Abad, A.; Montoya, A.; Maquieira, A.; Puchades, R. J. Immunol. Methods 1997, 208, 75-80 [62] Allara, D.L. Biosens. Bioelectron. 1995, 10, 771-783 [63] Wink, T.; van Zuilen, S.J.; Bult, A.; van Bennekom, W.P. Analyst. 1997, 122, 43R-50R [64] Ferretti, S.; Paynter, S.; Russell, D.A.; Sapsford, K.E.Trends in Analytical Chemistry 2000, 19, 530-540 [65] Flink, S.; van Veggel, F.C.J.M.; Reinhoudt, D.N. Adv. Mater. 2000, 12, 1315-1328 [66] Chaki, N.K.; Vijayamohanan, K. Biosens. Bioelectron. 2002, 17, 1-12 [67] Patel, N.; Davies, M.C.; Hartshorne, M.; Heaton, R.J.; Roberts, C.J.; Tendler, S.J.B.; Williams, P.M. Langmuir 1997, 13, 6485-6490 [68] Lahiri, J.; Issacs, L.; Tien, J.; Whitesides, G.M. Anal. Chem. 1999, 71, 777-790 [69] Wagner, P.; Zugg, F.; Kernen, P.; Hegner, M.; Semenza, G. J. Vac. Sci. Technol. B 1996, 14, 1466-1471 [70] Hegner, M.; Dreier, M.; Wagner, P.; Semenza, G.; Guntherodt, H.J. J. Vac. Sci. Technol. B 1996, 14, 1418-1421 [71] Chirakul, P.; Perez-Luna, V.H.; Owen, H.; Lopez, G.P.; Hampton, P.D. Langmuir 2002, 18, 4324-4330 [72] Cotton, C.; Glidle, A.; Beamson, G.; Cooper, J.M. Langmuir 1998, 14, 5139-5146

50

Chapter 2: Design of high quality interface layer

[73] Horton, R.C.; Herne, T.M.; Myles, D.C. J. Am. Chem. Soc. 1997, 119, 12980-12981 [74] Boozer, C.; Yu, Q.; Chen, S.; Lee, C.; Homola, J.; Yee, S.; Jiang, S. Sens. Actuator B-Chem. 2003, 90, 22-30 [75] Nelson, K.E.; Gamble, L.; Jung, L.S.; Boeck, M.S.; Naeemi, E.; Golledge, S.L.; Sasaki, T.; Castner, D.G.; Campbell, C.T.; Stayton, P.S. Langmuir 2001, 17, 2807-2816 [76] Sun, X.; Stabler, C.L.; Cazalis, C.S.; Chaikof, E.L. Bioconjugate Chem. 2006, 17, 52-57 [77] Yam, C.; Pradier, C.; Salmain, M.; Fischer-durand, N.; Jaouen, G. J. Colloid Interface Sci. 2002, 245, 204-207 [78] Ladd, J.; Boozer, C.; Yu, Q.; Chen, S.; Homola, J.; Jiang, S. Langmuir 2004, 20, 8090-8095 [79] Housseman, B.T.; Gawalt, E.S.; Mrksich, M. Langmuir 2003, 19, 1522-1531 [80] Sigal, G.B.; Bamdad, C.; Barberis, A.; Strominger, J.; Whitesides, G.M. Anal. Chem. 1996, 68, 490-497 [81] Huang, S.; Caldwell, K.D.; Lin, J.; Wang, H.; Herron, J.N. Langmuir 1998, 12, 4292-4298 [82] Shriver-Lake, L.C. in: Immobilized Biomolecules in Analysis, Cass, A.E.G.; Ligler, F.A. Ed., Oxford University Press, Oxford, UK, 1998, 1-14 [83] Mrksich, M. Exp. Cell Res. 1997, 235, 305-313 [84] Tarlov, M.J. Langmuir 1992, 8, 1398-1405 [85] Schoenfisch, M.H. J. Am. chem. Soc., 1998, 120, 4502-4513 [86] Gothelf, K.V.; Larsen, A.G. J. Colloid Interface Sci. 2002, 255, 356–362 [87] Schlenoff, J.B.; Li, M.;Ly, H. J. Am. Chem. Soc. 1995, 117, 12528-12536 [88] Chambers, R.Carlisle.; Inman, C.E.; Hutchison, J. E. Langmuir 2005, 21, 4615-4621 [89] Brewer, N.J.; Leggett, G.J. Langmuir 2004, 20, 4109-4115 [90] Buseman-Willams, J.; Berg, J.C. Langmuir 2004, 20, 2026 [91] Disley, D.M.; Morrill, P.R.; Sproule, K.; Lowe, C.R. Biosens. Bioelectron. 1999, 14, 481-493 [92] Cooper, M.A. Anal. Bioanal. Chem. 2003, 377, 834-842 [93] Yan, M.; Cai, S.X.; Wybourne, M.N.; Keana, J.F.W. J. Am. Chem. Soc. 1993, 115, 814-816 [94] Hengsakul, M.; Cass, A.E.G. Bioconjugate Chem. 1996, 7, 249-254 [95] Konig, B.; Gratzel, M. Anal. Chim. Acta 1993, 276, 329-337 [96] Konig, B.; Gratzel, M. Anal. Lett. 1993, 26, 1567-1571 [97] Zhou, X.D.; Liu, L.J.; Hu, M.; Wang, L.L.; Hu, J.M. Pharmaceutical and

51

Chapter 2: Design of high quality interface layer

Biomedical Analysis 2002, 27, 341-345 [98] Kuijpens, J.M.H.; Kardaun, G.A.; Blezer, R.; Pijpers, A.P.; Koole, L.H. J. Am. Chem. Soc. 1995, 117, 8691-8697 [99] Chaudhury, M.K. Biosens. Bioelectron. 1995, 10, 785-788 [100] Ferguson, G.S.; Chaudhury, M.K.; Biebuyck, H.A.; Whitesides, G.M. Macromolecules 1993, 26, 5870-5875 [101] Kwok, C.S.; Mourad, P.D.; Crum, L.A.; Ratner, B.D. Biomacromolecules 2000, 1, 139-148 [102] Kok, F.N.; Bozoglu, F.; Hasirci, V. Biosens. Bioelectron. 2002, 17, 531-539 [103] Tsiafoulis, C.G.; Prodromidis, M.I.; Karayannis, M.I. Anal. Chem. 2002, 74, 132-139 [104] Schellerand, F.; Schmid, R.D. Biosensors: Fundamentals, Technologies and Applications; GBF Monograph, VCH: Weinheim, 1992, 17 [105] Doyl, J.M.; Wehmeyer, K.R.; Halsall, H.B. Heterogeneous Enzyme Immunoassays with Electrochemical Detection, In: Electrochemical Sensors in Immunological Analysis, Ngo TT, Ed.; Plenum Press: New York, 1987; p87; [106] Yasuda, H. In Thin Film Process, Vossen, J.L.; Kern, W., Ed.; Academic Press, New York, 1978, p314; [107] Yasuda, H. Plasma Polymerization, Academic Press, New York, 1985; [108] Nakanishi, K.; Muguruma, H.; Karube, I. Anal. Chem. 1996, 68, 1695-1700 [109] Wu, Z.; Yan, Y.; Shen, G.; Yu, R. Anal. Chim. Acta 2000, 412, 29-34 [110] Davis, J.; Glidle, A.; Cass, A.E.G.; Zhang, J.; Cooper, J.M. J. Am Chem Soc. 1999, 121, 4302-4303 [111] Amos, R.A.; Anderson, A.B.; Clapper, D.L.; Duquette, P.H.; Duran, L.W.; Hohle, S.G.; Sogard, D.J.; Swanson, M.J.; Guire, L.W. Biomaterial surface modification modification using photochemical coupling technology, In: Encyclopedic Handbook of Biomaterials and Bioengineering, Part A: Materials, 1995, 895-926 [112] Su, X.D. Biochem. Biophys. Res. Commun. 2002, 290, 962-966 [113] Sigrist, H.; Colliond, A.; Clemence, J.; Gao, H.; Sanger, M.; Sundarababu, G. Opt. Eng. 1995, 35, 2339-2348 [114] Espadas, T.C.; Meyerhoff, M.E. Anal. Chem. 1995, 67, 3108-3114 [115] Cook, A.D.; Hrkach, J.S.; Gao, N.N.; Johnson, I.M.; Pajvnai, U.B.; Cannizzaro, S.M. J. Biomed. Mater. Res. 1997, 35, 513-523 [116] Bergstrom, K.; Osterberg, E.; Holmerg, K.; Hoffman, A.S.; Schuman, T.P.; Kozlowski, A.; Harris, J.M. J. Biomater. Sci., Polym. Ed. 1994, 6,

52

Chapter 2: Design of high quality interface layer

123-129 [117] Llanos, G.R.; Sefton, M.V. J. Biomed. Mater. Res. 1993, 27, 1383-1396 [118] Brinkman, E.; Poot, A.; van der Does, L.; Bantjes, A. Biomaterials 1990, 11, 455-453 [119] Claesson, P.M.; Blomberg, E.; Paulson, O.; Malmsten, M. Colloids Surf., A 1996, 112, 131-139 [120] Malmsten, M.; Emoto, K.; Van Alstine, J.M. J. Colloid Interface Sci. 1998, 202, 507-517 [121] Brink, C.; Osterberg, E.; Holmberg, K.; Tiberg, F. Colloids Surf. 1992, 66, 149-156 [122] Goessl, A.; Garrison, M.D.; Lhoest, J.B., Hoffman, A.S. J. Biomater. Sci., Polymer Edn., 2002, 12, 721-738 [123] Nakamura, H.; Karube, I. Anal. Bioanal. Chem. 2003, 377, 446-468 [124] Herrwerth, S.; Rosendahl, T.; Feng, C.; Fick, J.; Eck, W.; Himmelhaus, M.; Dahint, R.; Grunze, M. Langmuir 2003, 19, 1880-1887 [125] Ruiz-Taylor, L.A; Martin, T.L.; Zaugg, F.G.; Witte, K.; Indermuhle, P.; Nock, S.; Wagner, P. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 852-857 [126] Plunkett, M.A.; Claesson, P.M.; Erntsson, M.; Rutland, M.W. Langmuir 2003, 19, 4673-4681 [127] Tully, D.C.; Frechet, J.M.J. Chem. Commun. 2001, 14, 1229-1239 [128] Fischer, M.; Vogtle, F. Angew. Chem. Int. Ed. 1999, 38, 884-905 [129] Inoue, K. Prog. Polym. Sci. 2000, 25, 453-571 [130] Sidorenko, A.; Zhai, X.W.; Greco, A.; Tsukruk, V.V. Langmuir 2002, 18, 3408-3412 [131] Yoon, H.C.; Hong, M.Y.; Kim, H.S. Langmuir 2001, 17, 1234-1239 [132] Yoon, H.C.; Hong, M.Y.; Kim, H.S. Anal. Biochem. 2000, 282, 121-128 [133] Yoon, H.C.; Kim, H.S. Anal. Chem. 2000, 72, 922-926 [134] Yoon, H.C.; Hong, M.Y.; Kim, H.S. Anal. Chem. 2000, 72, 4420-4427 [135] Britland, S.; Perez-Arnaud, E.; Clark, P.; McGinn, B.; Connoly, P.; Moores, G. Biotechnol. Prog.1992, 8, 155-160 [136] Fodor, S.P.A.; Read, J.L.; Pirrung, M.C.; Stryer, L.; Lu, A.T.; Solas, D. Science, 1991, 251, 767-773 [137] Trojanowicz, M. Membr. Sci. Technol. Ser. 2003, 7, 807-845 [138] Tien, H. T.; Ottova, A. L. J. Membr. Sci. 2001, 189, 83-117. [139] Oellerich, S.; Lecomte, S.; Paternostre, M.; Heimburg, T.; Hildebrandt, P. J. Phys. Chem. B 2004, 108, 3871-3878. [140] Snejdarkova, M.; Rehak, M.; Otto, M. Anal. Chem. 1993, 65, 665. [141] Nikolelis, D. P.; Petropoulou, S.-S. E. Biochim. Biophys. Acta 2002, 1558, 238-245

53

Chapter 2: Design of high quality interface layer

[142] Nikolelis, D. P.; Pantoulias, S. Anal. Chem. 2001, 73, 5945-5952. [143] Matsuno N.; Murawsky M.; Ridgeway J.; Cuppoletti J. Biochim. Biophys. Acta 2004, 1665, 184-190. [144] Ottova, A. L.; Tien, H. T. J. Surf. Sci. Technol. 2000, 16, 115-148. [145] Stine, R.; Pishko, M. V.; Schengrund, C.-L. Langmuir 2004, 20, 6501-6506. [146] Terrettaz, S.; Stora, T.; Duschl, C.; Vogel, H. Langmuir 1993, 9, 1361-1369. [147] Liley, M.; Bouvier, J.; Vogel, H. J. Colloid Interface Sci. 1997, 194, 53-58. [148] Stelzle, M.; Weissmuller, G.; Sackmann, E. J. Phys. Chem. 1993, 97, 2974-2981. [149] Fang, Y.; Frutos, A. G.; Lahiri, J. Langmuir 2003, 19, 1500-1505. [150] Groves, J. T.; Ulman, N.; Cremer, P. S.; Boxer, S. G. Langmuir 1998, 14, 3347-3350. [151] Holden, M. A.; Jung, S.-Y.; Yang, T.; Castellana, E. T.; Cremer, P. S. J. Am. Chem. Soc. 2004, 126, 6512-6513. [152] Johnston, D. S.; Sanghera, S.; Pons, M.; Chapman, D. Biochim. Biophys. Acta 1980, 602, 57-69. [153] Jenkins, A. T. A.; Boden, N.; Bushby, R. J.; Evans, S. D.; Knowles, P. F.; Miles, R. E.; Ogier, S. D.; Schonherr, H.; Vancso, G. J. J. Am. Chem. Soc. 1999, 121, 5274-5280. [154] Matthews, J. R.; Tencel, D.; Jacobs, R. M. J.; Bain, C. D.; Anderson, H. L. J. Am. Chem. Soc. 2003, 125, 6428-6433. [155] Malzak, K. A.; Ellar, D. J.; Gizeli, E. Langmuir 2004, 20, 1386-1392. [156] Zhang, L.; Longo, M. L.; Stroeve, P. Langmuir 2000, 16, 5093-5099. [157] Anrather, D.; Smetazko, M.; Saba, M.; Alguel, Y.; Schalkhammer, T. J. Nanosci. Nanotechnol. 2004, 4, 1-22. [158] Podbielska, H.; Ulatowska-Jarza, A. Bull. Pol. Acad. Sci.-Chem. 2005, 53, 261-271 [159] Dale, N.; Hatz, S.; Tian, F.; Llaudet, E. Trends Biotechnol. 2005, 23, 420-428 [160] Coradin, T.; Boissiere, M.; Livage, J. Curr. Med. Chem. 2006, 13, 99-108 [161] Doong, R.; Shih, H.; Lee, S. Sens. Actuator B-Chem. 2005, 111-112, 323-330 [162] Piunno, P.A.E.; Krull, U.J. Anal. Bioanal. Chem. 2005, 381, 1004-1011

54

Chapter 3: Protein Adsorption at Solid-liquid Interface 3.1 Introduction The adsorption of the proteins at solid-interface attracts extensively notices because of its importance and relevance in biosensor development [1]. Herein, a crucial step is attaching the biological sensing elements onto the transducer surface in a qualitatively (e.g. activity, affinity, orientation) and quantitatively (e.g. loading amount) controllable manner [2]. The quality and quantity of the bound biological sensing elements greatly depend on the their interactions with the interface, such as the kinetics of their binding, the degree of hydrophobic, electrostatic and chemical attraction/repellence, and the conformational changes [1,3]. The investigation of the adsorption of biological materials, e.g. proteins, at the solid-liquid interface will provide valuable information for achieving quantitative and qualitative control of the binding of biological sensing elements on the biosensor surface. In addition, reducing the nonspecific protein adsorption is a critical issue in improving the sensitivity of the biosensor. The understanding of the mechanism of protein adsorption will also help to figure out the efficient way to block the nonspecific protein adsorption. Protein adsorption is a complex process involving van der Waals forces, hydrophobic and electrostatic interactions, hydrogen bonding, and sometimes even covalent chemical bonding. Although, the surface-protein interactions are not well-understood, it has been shown that the interactions are influenced by a variety of factors, such as the size, the charge, the hydrophobicity of the protein, the chemical and topographical feature of the surface, the competitive adsorption/desorption between mixed proteins, and the water molecules, ions,

Chapter 3: Protein Adsorption at Solid-liquid Interface

and other small molecules in the vicinity of the surface and the proteins [4]. Despite that it will be greatly favored if one can include all these parameters in one pot of research, the intrinsic complexity of such systems will prevent the un-ambiguous interpretation of the results. Moreover, the multi-variants involved in such systems result in poor control over the reproducibility of the experiments. Herein, we organized our study in a simpler but more controllable manner, where single protein adsorption on surface with single dominant property (e.g. hydrophobic or hydrophilic) was investigated. We first thoroughly characterized the adsorption of Immunoglobulin G (IgG) protein with varied concentration on a typical hydrophobic surface, which enables us to conclude clearly not only the concentration-dependence of the protein adsorption but also surface hydrophobicity-dependence of the properties of the adsorbed protein (e.g. kinetics, conformation, water content, etc.)(in Section 3.2). Then we investigated the influence of protein properties, such size, charge, hydrophobicity, on the adsorption at typical hydrophobic surface (in Section 3.3). Finally, we studied the influence of surface properties (hydrophobic vs. hydrophilic) on the adsorption of the single type of protein (in Section 3.4). In addition, the understanding of the protein adsorption at the solid-liquid interface depends on quantitative and qualitative information that can be obtained through the study. Thus, approaches that can provide reliable and complementary measurements of protein adsorption will be greatly favored. In parallel with the study of IgG protein adsorption on the hydrophobic surface, we also developed an efficient combined methodology, which can provide reliable ad complementary experimental information for our study, as shown in Section 3.2. 3.2 Human Immunoglobulin G Adsorption on a typical hydrophobic surface Adsorption of immunoglobin G (IgG) on solid surfaces has attracted strong research interest because of its wide application in biotechnology, immunoassays and biosensors [5-8]. The detailed understanding of the mechanisms and physicochemical parameters governing the adsorption behavior is essential for the development of novel immunoassays and biosensors. Various techniques, based on different principles such as radiolabeling [9-11], optical adsorption [12-14], refractive index changes [15-22], electromechanical microbalance [8,23,24] and others [25-32] have been used to investigate IgG adsorption. However, the adsorption of IgG molecules at the liquid-solid interface is a

56

Chapter 3: Protein Adsorption at Solid-liquid Interface

sophisticated process [13,14,18,20,26,28,32-34]. The understanding of the mechanisms of the IgG adsorption should be improved by the simultaneous investigation of the protein film properties, such as surface coverage, thickness, and conformation changes. One should mention that, due to the diversity in the reported experimental parameters, such as pH [32], ionic concentration [7], and IgG type [18,19] used in previous studies, a direct comparison of experimental results obtained by different measurement methods seems unrealistic. The combination of different in situ measurement techniques would provide a new opportunity to obtain complementary information about the protein adsorption process occurring at the same measurement time and under the same experimental conditions. So far, only a few multi-technique studies have been reported [35,36-39]. In this study, the adsorption behavior of human IgG (hIgG) is investigated in a broad solution concentration range (from 100 ng/ml to 3 mg/ml) on a hydrophobic surface by means of (i) Quartz Crystal Microbalance-Dissipation (QCM-DTM), (ii) combined Surface Plasmon Resonance (SPR) and Love mode Surface Acoustic Wave (SAW), and (iii) combined QCM-DTM and Atomic Force Microscopy (AFM) techniques. Among these, both the QCM-DTM and SAW techniques are based on acoustic probing of the protein layer under investigation, providing information about mass uptake on the sensor. The QCM-DTM technique also provides a unique set of information about visco-elastic property of the protein layer as a result of complex modeling of the interaction of the acoustic wave with the protein layer and the solvent [23,24,40,41] , while the SAW signal is not sensitive to visco-elastic effects [42]. The visco-elastic variation of the adsorbed protein layer was quantified by comparison of the results from QCM-DTM and SAW. SPR is an optical method, which provides information about adsorbed protein amount and layer thickness [15,16,35] . In addition, information about the water content and thickness of the adsorbed protein layer was deduced from combined SPR/SAW investigation. The water content of the protein film was extracted as both techniques provide information on one common parameter, the thickness of the layer. In a simultaneous measurement, SAW gives information about the density parameter of the film, while SPR gives information about the optical index parameter of the layer. Upon reducing the number of parameters by assuming that both density and optical index scale linearly with the protein/water ratio in the protein layer, one could extract a unique pair of layer thickness and water content values from the simultaneous set of measurements [43]. Moreover, the ambiguous quantitative information deduced from the QCM-DTM measurement,

57

Chapter 3: Protein Adsorption at Solid-liquid Interface

in terms of adsorbed protein amount [44-47], was precised using the additional information obtained from the combined SPR/SAW measurements. Finally, AFM is one of the most useful techniques to characterize the organization of the adsorbed protein film scaling down to a molecular resolution [28,30,31], the combined QCM-DTM/AFM measurement provides substantial new results, such as adsorption kinetics, lateral film organization and time-resolved information about the conformation change at the interface [48-50]. The combination of different surface sensitive techniques allowed us to obtain both adsorption kinetics and structural information, which demonstrates that the results deduced from the various measurement method not only validate but also complement each other. 3.2.1 Experimental section 3.2.1.1 Materials and preparation of the sensor chip surfaces Human immunoglobulin G (hIgG) (chrompure) was purchased from Jacksson ImmunoResearch Inc. (USA). 1-octadecanethiol (ODT) (>97%) was obtained from Aldrich. Ultrapure absolute ethanol was purchased from Riedel-de Haën. The inorganic salts were of p.a. grade (Merck or Fluka). The buffer solution (PBS, pH=7.4) was prepared with NaCl (0.15 M) and 1x10-2 M Na2HPO4/KH2PO4 Glycine hydrochloride was from Sigma. The water was of an ultra pure grade for microelectronic purposes ( HSA TR, while the slopes of the D5 versus Cprotein dependency vary as FB>IgG HSA TR. The visco-elastic behavior of the IgG, TR, HSA and FB films can be characterized by the shear relaxiation time [82], , defined by τ =−

1 δ ( ∆D ) (3.4) 4π δ ( ∆f )

Table 5 shows the outcome of the fit at Cprotein=0.46mg/ml. The four proteins

85

Chapter 3: Protein Adsorption at Solid-liquid Interface

in a sequence of increasing protein film shear relaxation time is TR10 µm) PEG gel layer would be required to provide the topographic features needed to completely mechanically cage the neuron somas (Figure 10). As shown by Figure 10, neurites of neurons in neighboring PEGfree areas can sometimes bridge the PEG gel, and displace the somas by pulling during growth. Although this is not uncommon in topographic growth guidance, it indicates that the PEG gel should be made thicker as compared to the size of the hippocampal neurons. Growing neurons extend their neurites, and as such discover their surroundings, which is an essential characteristic of their growth. Therefore, creating interconnecting PEG-free channels between PEG-free regions, should also help to further cage the neurons, as it would provide growing neurons an opportunity to extend and connect their neurites via PEG-free areas instead of across the PEG gel.

Figure 10 DIC (differential interference contrast) image of neuron cells on PEG gel pattern; the circular areas are free of PEG gel; the bright field consists of neuron cells.

247

Chapter 7: Controlling the Organization of Neuron Cells on Sensor Surfaces-a Step towards a Hybrid Neuron-Transistor

7.4 Conclusions The hybrid neuro-electronic systems show great potential for biosensing and for studying neuron physiology. For both purposes, a well defined neuron network pattern is highly desired. We have developed a novel approach to gain control over neuronal organization. This novel approach is based on a PEG gel, which is formed via the UV-light induced polymerization of poly(ethylene glycol) diacrylate (PEG-DA) monomer. Patterning of this PEG gel on oxide surfaces can be easily achieved through conventional photolithography. A PEG-DA with molecular weight of 575 Da was used as a model system to investigate the potential of patterned PEG gel in controlling the organization of growing neurons. Both the content of photoinitiator and the exposure time were optimized: in order to obtain a reliable pattern, the content of photoinitiator should be higher than 5 g/100 ml and the exposure time should range between 5 to 15 sec. The stability of the PEG gel on the oxide substrate was improved by introducing an adhesion promoter layer, consisting of a silane with functional acrylate endgroups. The resulting dry PEG gel had a thickness value of 5 µm and a minimal feature size of 8 µm. A physico-chemical restrictive pattern was obtained by immersing substrates with a patterned PEG gel in a solution containing cell-adhesion proteins. Since the cell-adhesion proteins were repelled by the PEG, they adsorbed exclusively onto the areas free of PEG gel, thereby clearly defining cytophilic areas next to cytophobic PEG covered regions. Subsequently, such substrates were tested successfully to control the organization of hippocampal neuron cells, as they would grow preferentially in areas free of PEG. A big advantage of this PEG gel is that its thickness could be adjusted by changing the spin-coating conditions (speed and time) and the viscosity of its PEG monomer solution (which is influenced by the molecular weight of the PEG monomer and the concentration of its solution). In addition, the thickness of the PEG gel can further increase by it swelling when in contact with a good solvent. The degree of swelling is amongst others determined by the stiffness of the PEG gel, which can be tuned by the content of photoinitiator. However, increasing the amount of photoinitiator might reduce the cytophobic nature of the PEG gel and has to be investigated. In this study, we only performed preliminary test on the ability of a patterned PEG gel to control the organization of growing neurons. Although the results

248

Chapter 7: Controlling the Organization of Neuron Cells on Sensor Surfaces-a Step towards a Hybrid Neuron-Transistor

suggest that our approach is promising, there are still a few tests that remain before the patterned PEG gel could be applied in real applications. For our preliminary neuron growth tests, we used a patterned PEG gel with a thickness of 5 µm. We found that such thickness was not enough to provide topographic features that could completely control neuron outgrowth. This is in part attributed to a thickness of the gel that was still too low in comparison with the size of the neuron soma (around 10 µm). In the future, thicker PEG gels should be tested. Another factor that probably adds to incomplete growth control, are the patterns that were used in the current study. These patterns were just simple lines and circles, and our tests focused mainly controlling on the neuron soma location. However, for real world application, patterns that are able to control also the outgrowth of neurites and the connections between neighboring cells are required. Such more complex patterns should be tested in the future as well. Finally and most importantly, for real applications, a patterned neuron network should retain its native electrophysiological behavior. It has been reported that patterning neurons can sometimes alter their electrophysiological activity. Hence, an urgent task for the future is to investigate the electrophysiology of patterned neuronal networks on PEG gel by means of suitable techniques such as patch clamp and field-effect transistors.

249

Chapter 7: Controlling the Organization of Neuron Cells on Sensor Surfaces-a Step towards a Hybrid Neuron-Transistor

Reference [1] Durick, K.; Negulescu, P. Biosens. Bioelectron. 2001, 16, 587 [2] Park, T.H.; Shuler, M.L. Biotechnol. Prog. 2003, 19, 243 [3] Wang, P.; Xu, G.X.; Qin, L.F.; Xu, Y.; Li, Y.; Li, R. Sens. Acutators B 2005, 108, 576 [4] Morin, F.O.; Takamura, Y.; Tamiya, E.; J. Biosci. Bioeng. 2005, 100, 131 [5] Fromherz, P., In Nanoelectronics and Information Technology, Waser, R. Ed., Wiley-VCH, Berlin, 2003, 781 [6] Shaffer, K.M.; Lin, H.J.; Maric, D.; Pancrazio, J.J.; Stenger, D.A.; Barker, J.L.; Ma, W. Biosens. Bioelectron. 2001, 16, 481 [7] Keefer, E.W.; Gramowski, A.; Stenger, D.A.; Pancrazio, J.J.; Gross, G.W. Biosens. Bioelectron. 2001, 16, 513 [8] Peterman, M.C.; Mehenti, N.Z.; Bilbao, K.V.; Lee, C.J.; Leng, T.; Noolandi, J.; Bent, S.F.; Blumenkranz, M.S.; Fishman, H.A. Artif. Organs 2003, 27, 975 [9] Jung, D.R.; Kapur, R.; Adams, T.; Giuliano, K.A.; Mrksich, M.; Craighead, H.G.; Taylor, D.L. Crit. Rev. Biotech. 2001, 21, 111 [10] Lakard, S.; Herlem, G.; Valles-Villareal, N.; Michel, G.; Propper, A.; Gharbi, T.; Fahys, B. Biosens. Bioelectron. 2005, 20, 1946 [11] Vogt, A.K.; Wrobel, G.; Meyer, W.; Knoll, W.; Offenhäusser, A. Biomaterials 2005, 26, 2549 [12] Merz, M.; Fromherz, P. Adv. Mater. 2002, 14, 141 [13] Chang, J.C.; Brewer, G.J.; Wheeler, B.C. Biosens. Bioelectron. 2001, 16, 527 [14] Arnold, F.J.L.; Hofmann, F.; Bengtson, C.P.; Wittmann, M.; Vanhoutte, P.; Bading, H. J. Physiol. 2004, 564, 3 [15] Li, N.Z.; Tourovskaia, A.; Folcb, A. Crit. Rev. Biomed. Eng. 2003, 31, 423 [16] Hammarback, J.A.; Palm, S.L.; Furcht, L.T.; Letourneau, P.C. J. Neurosci. Res. 1985, 13, 213 [17] Kleinfeld, D.; Kahler, K.H.; Hockberger, P.E. J. Neurosci. 1988, 8, 4098 [18] Wheeler, B.C.; Corey, J.M.; Brewer, G.J.; Branch, D.W. J. Biomech. Eng. 1999, 121, 73 [19] Martinoia, S.; Bove, M.; Tedesco, M.; Margesin, B.; Grattarola, M. J. Neurosci. Methods 1999, 87, 35 [20] Fromherz, P.; Schaden, H.; Vetter, T. Neurosci. Lett. 1991, 129, 77 [21] Fromherz, P.; Schaden, H. Eur. J. Neurosci. 1994, 6, 1500 [22] Prinz, A.A.; Fromherz, P. Biol. Cybern. 2000, 82, L1 [23] Jenkner, M.; Müller B.; Fromherz, P. Biol. Ctbern. 2001, 84, 239

250

Chapter 7: Controlling the Organization of Neuron Cells on Sensor Surfaces-a Step towards a Hybrid Neuron-Transistor

[24] Zeck, G.; Fromherz, P. Proc. Natl. Acad. Sci. USA 2001, 98, 10457 [25] Lee, J.H.; Kopecek, J.; Andrade, J.D. J. Biomed. Mater. Res. 1989, 23, 351 [26] Bryant, S.J.; Anseth, K.S. Biomaterials 2001, 22, 619 [27] Harris, J.M.; Zalipsky, S. Poly(ethylene glycol): Chemistry and Biological Applications; American Chemical Society: Washington, DC, 1997 [28] Sorribas, H.; Braun, D.; Leder, L.; Sonderegger, P.; Tiefenauer, L. J. Neurosci. Methods 2001, 104, 133 [29] Corey, J.M.; Feldman, E.L. Exp. Neurol. 2003, 184, S89 [30] Cruise, G.M.; Scharp, D.S.; Hubbell, J.A. Biomaterials 1998, 19, 1287 [31] Vassanelli, S.; Fromherz, P.; J. Neurosci. 1999, 19, 6767 [32] Ghandhi, S.K. VLSI Fabrication Principles, 2nd ed; John Wiley & Sons: New York, 1994 [33] Bruice P.Y.; Organic Chemistry, Prentice-Hall International, New Jersey, 1995 [34] Elbert, D.L.; Pratt, A.B.; Lutolf, M. P.; Halstenberg, S.; Hubbell. J.A. J. Control. Release 2001, 76, 11

251

Chapter 7: Controlling the Organization of Neuron Cells on Sensor Surfaces-a Step towards a Hybrid Neuron-Transistor

252

Chapter 8: Conclusions and Outlook 8.1 Task and Purpose The performance of a biosensor is collectively influenced by the properties of its three main components, i.e. the transducer, the interface layer, and the biological sensing elements. Among them, the interface layer plays a crucial role, since it interacts with both the transducer and biological sensing elements and influences their properties, thus influencing the performance of the biosensors. The sensitivity and response time of the transducer are related to the thickness of the interface layer. As for the biological sensing elements, the interface layer is responsible to attach them onto the sensor surface with fully retained activity and stability, while, at the same time, in a quantitatively, spatially and geometrically controllable manner. In addition, the importance of the interface layer lies in its ability to prevent nonspecific adsorption. In summary, the interface layer will influence the sensitivity, stability, and reproducibility of the biosensors to a significant degree. . Despite of the various types of transducers and biological sensing elements exploited in biosensor construction, a good interface layer should cope with some universal properties, such as stability, controllability of the film structure at the molecular scale, ability immobilization biomolecules in a controllable way (amount, activity, etc.), and ability to reduce the non-specific adsorption. But on the other hand, the design of an optimized interface layer should also take into account the specificity of the individual transducer and the specificity of the individual biological sensing elements. For example, a transducer based

Chapter 8: Conclusion and Outlook

on the conductimetric principle requires a conducting interface layer, while a potentiometric transducer requests an interface layer with good insulating properties. Given the importance of the interface layers for the biosensor development, we are interested to explore novel methods for constructing interface layers with well controlled properties, and to incorporate them in high performance biosensor In this PhD work, several interface layers, useful for different applications (immunosensing, DNA sensing and cells-based sensors), were developed and optimized, and these interface layers were exploited to realize different types of biosensors. 8.2 Fundamental study of protein adsorption Investigation of the adsorption of biological materials, e.g. proteins, at the solid-liquid interface can provide invaluable information for achieving quantitative and qualitative control of the binding of biological sensing elements in a biosensors. In addition, reducing the nonspecific protein adsorption is a critical issue in improving the sensitivity of the biosensor. Understanding the mechanism of protein adsorption helps to determine the efficient way to block the adsorption of nonspecific proteins. Four kinds of human blood proteins, human immunoglobulin G (hIgG), human serum albumin (HSA), human transferrin (HT), and human fibrinogen, were used as model systems. Their physisorption on surfaces with well controlled properties (hydrophobic and hydrophilic) were investigated with emphasis on the kinetics, visco-elastic variation, interfacial hydration and structural details. A novel methodology, namely combined Quartz Crystal Microbalance with Dissipation (QCM-D) and Surface Plasmon Resonance (SPR), combined SPR and Surface Acoustic Wave (SAW), and combined QCM-D and AFM, were applied for such study. The adsorption kinetics of hIgG on a hydrophobic surface showed concentration dependence: the higher the concentration, the faster the initial adsorption stage was; also, the higher the concentration, the longer the time to reach a saturation of the signal at the solid/liquid interface. With the increasing of the proteins concentrations, the adsorbed hIgG formed a coverage varying from monolayer to multilayers. The hIgG formed a fully covered monolayer with presumed “end-on” orientation when the concentration was higher than 11.5 g/ml, but did not exceed 57.5 g/ml. A conformation change

254

Chapter 8: Conclusion and Outlook

occurring after the extended contact of the initial monolayer with the solid surface resulted in a compact monolayer with a thickness much less than 10 nm. The adsorbed hIgG layer barely contained water. Both the surface density and the visco-elastic property of the adsorbed protein layers varied depending on the type of the proteins. The properties of the proteins, such as size, polarity, and surface charge, are believed to play a role in the interaction between the protein and the solid interface, because these properties will ultimately determine the type and intensity of the interactions between the surface and the protein itself. For same type of proteins, the surface density and the visco-elastic property of the adsorbed protein layer depended greatly on the surface properties. Much lower surface density and less visco-elastic change were observed within protein layers on hydrophilic surface compared to that on hydrophobic surface. The adsorbed protein layer suffered from less conformation change during the adsorption onto hydrophilic surface. In the future, the chemical binding instead of simple physisorption of biomolecules should be investigated using the methodology presented above, since the biomolecules are covalent bound onto the sensor surface in most biosensor applications. Moreover, the further binding of target biomolecules on the immobilized biological sensing elements should also be studied with regard to the information including kinetic, visco-elastic variation, interfacial hydration and structural details. We believe all these information will provide invaluable insight in optimizing the performance of biosensors. For example, the test time is a crucial factor for real biosensor application. To shorten the test course, a fast kinetics of the protein attachment will be preferred. Basing on the kinetics info gained in current study, we will be able to choose the optimal parameters to secure a fast kinetics in the real biosensor applications. 8.3 Constructing novel interfaces for preventing nonspecific adsorption we have developed a Polymeric Monolayer (PM) - based on a water-soluble PEG-grafted polymer - that combines the simplicity of SAMs with the mechanical robustness of polymers. This PM was used to construct proteinresistance surface, which can prevent the surface from nonspecific protein adsorption. Comparing the PMs formed from water and toluene based copolymer

255

Chapter 8: Conclusion and Outlook

solutions, we have found that the solvent of the copolymer solution used for the monolayer fabrication critically determines the structural organization and hence the protein resistance of the resulting PMs. Indeed, we have shown that PMs formed from a water based solution had a higher surface coverage and denser packing, exhibited a more stratified structure with a PEG enriched outer layer - compared to the entangled structure of the toluene based PMs. And overall, the water based PMs had a higher surface density of PEG than their toluene based counterparts. As demonstrated by our experiments, PMs formed from a water based copolymer solution have superior performance compared to toluene based monolayers in resisting protein adsorption, even withstanding protein adsorption in non-diluted human serum. In addition, a chemically patterned surface could be easily created through a conventional lithographic technique, i.e. a lift-off method when water is used as a solvent. We further optimized the protein-resistance of the PMs through investigating the relationship between the interfacial architecture of PMs and the protein adsorption. Both in-situ and ex-situ surface characterization techniques revealed that the surface PEG density varied with the composition of the copolymer: the length of the PEG chain and the alkyl chain, and the grafted PEG ratio. Factors, which are crucial in determining the protein-resistance property of the PM, include PEG density and EG density (or the extent of packing (L/2RF)). Additional factors, such as the content of hydration water and surface hydrophilicity came also into play in determining protein resistance when the surface EG reached a threshold value (15unit/nm2 in current study). Under optimal condition, the PMs can efficiently reduce the non-specific adsorption of singe protein (e.g. human serum albumin and fibrinogen) at high concentration (2mg/ml). Moreover, the PMs can reduce the non-specifc adsorption up to 95% under extremely demanding test conditions (surface exposed to undiluted whole blood serum with >55mg/ml proteins). In the future, the feasibility of using self-assembled polymeric monolayer to modify oxide surfaces should be explored. Indeed, if we replaced the disulfide groups in the PMs described above with silane groups, the resulted PMs should be able to assemble onto oxide surface. As we proved in the above study, the chains order in PMs did not influence the quality, such as defect and coverage of the PMs, while for conventional alkane silane, the quality of the SAMs depend significantly on the order of molecule chains. Thus, the PMs might provide advantage of conventional alkane silane in oxide surface modification with regard that better quality can be achieved through PMs. However, a major challenge to achieve such application is the intensive and difficult material

256

Chapter 8: Conclusion and Outlook

preparation. The oxide-surface affinity groups are easy to lose activity during the synthesis, and the controlling of the compositions of the copolymer will require a lot of try-and-error test. 8.4 Immunosensor based on a novel functional PEG grafted polymeric monolayer We have designed a novel self-assembled polymeric monolayer, which consists of a polysiloxane backbone co-grafted with functional PEG chains and disulfide chains. The formed polymeric monolayer bears the self-assembly property of the alkanethiol system, while showing improved stability due to the multisite attachment via polysulfides (or disulfies) groups. This type of PEG grafted polymeric monolayers have readily controllable and tunable structures. Different functional groups can be incorporated into the polymeric monolayer though the end functionalized PEG chains. The content of the functional groups can be controlled through the relative graft ratio of PEG side chains. We have studied the potential of this self-assembled polymeric monolayers for immunosensing applications. A series of polymeric monolayers containing carboxylic functionalized PEG were prepared, and SPR sensors were modified by these polymeric monolayers. The SPR sensors based on the polymeric monolayers were applied in the immunosensing of human S100 protein and humane transferring (HT). The SPR sensor modified by the polymeric monolayer could efficiently immobilize the antibody. The amount of immobilized antibody depends on the content of carboxyl groups on the surfaces. The performance of the SPR immunosensor modified by the polymeric monolayers is influenced by the properties of these monolayers and the properties of antibody-antigen systems. For anti-S100 /S100 system, the polymeric monolayer modified SPR immunosensor shows a limit of detection (LOD) of 4.5 ng/ml in a direct assay format, and the LOD can be further improved to 0.45 ng/ml in a sandwich assay format. For anti-HT/HT system, the amount of antibody immobilization on such surface was similar to system, but the recognition level of HT on that for anti-S100 /S100 immobilized anti-HT was much lower compare to that of S100 on immobilized anti-S100 . It was attributed to the low affinity of anti-HT antibody. We have compared the performance of SPR sensors modified by different interface layers: i.e the polymeric monolayer, the mixed SAMs and the carboxymethylated dextran (CM5 chip). The nonspecific adsorption of

257

Chapter 8: Conclusion and Outlook

different proteins including human serum albumin, fibrinogen, and normal human serum, were tested on the SPR sensors modified by these interface layers. The polymeric monolayer modified SPR sensor showed the lowest degree of nonspecific adsorption, which was attributed to the PEG chains in the polymeric monolayer. The polymeric monolayer modified surface showed a 3 times higher antibody immobilization capacity than the SAM modified surface. Using anti-S100 /S100 as model systems, the SPR sensor with polymeric monolayer showed the best performance: i.e highest sensitivity and lowest LOD, compared to SPR sensors basing on the mixed SAMs and CM5 layer. We found the performance of the polymeric monolayer modified SPR sensor could be further improved via optimal antibody immobilization strategies. The biotin-streptavidin mediated antibody immobilization significantly improved the antigen binding efficiency. When Protein G was used for the antibody immobilization, both the amount of the antibodies immobilization and the degree of antigen recognition were significantly enhanced. In the future, different biomolecules binding groups, such as amine, aldehyde, epoxy, etc., other than carboxyl groups could be incorporated into the polymeric monolayer in order to optimize the biomolecules immobilization. In addition, oxide-affinity groups (e.g. silane) can be introduced into the polymeric monolayer in order to modify oxide sensor surfaces. Moreover, the feasibility of applying such polymeric monolayers for other immunosensor applications, such as SAW sensor, IDE sensor, and TPB sensors developed in IMEC, should be investigated. 8.5 Thermal spotting for microelectronic DNA chips A big challenge in constructing microelectronic DNA chips is the precise alignment of the DNA sample array on the predefined sensing areas on the chips. We have developed a simple and efficient approach to meet this challenge. The principle of our approach is based on the fact that the efficiency of surface Diels-Alder reactions is temperature dependent. By modifying the sensor surface with maleimide groups and labelling the DNA probes with diene groups, the immobilization of DNA on the surface was achieved through heating controlled Diels-Alder reaction. Under optimal reaction condition, the specific DNA probes were selectively immobilized on the desired sensor areas on the chip with good spatial resolution. Our results represent a proof-of-principle for the on chip thermal spotting. In

258

Chapter 8: Conclusion and Outlook

the future, the on-chip DNA hybridization should be investigated in order to fully realize microelectronic DNA chips. Because of the specific property of the electronic transducer, for instance not compatible with high salt concentration and sensitive to pH variation, a lot of conventional conditions, which has been known to be beneficial for optimal hybridization, may not be able to apply. Important factors, including DNA probes density, hybridization buffers, non-specific adsorption blocking agent, regeneration buffers, rinsing buffer, have to be optimized in viewing of the working mechanism of the electronic transducers. 8.6 Controlling the organization of neurons cells on field effect transistors using patterned PEG gels We developed an efficient approach to achieve the control of neuronal organization. The approach uses a UV curable PEG gel, formed via the polymerization of poly(ethylene glycol) diacrylate (PEG-DA) under UV light. The conventional photolithography is exploited to obtain patterns of PEG gels on the oxide surfaces. The physiochemical pattern, i.e. defined cytophobic areas and cytophilic areas, was achieved by incubating the substrate with PEG gels pattern in the solution containing cells-adhesion proteins. The PEG gels pattern provided not only a physiochemical cueing pattern, but also a topographical one, i.e. the neuron cells being mechanically caged by the surrounded thick layer of PEG. The thickness of the PEG gels can be varied from nanometer to micrometer by adapting the spin-coating conditions and the viscosity of the PEG monomer solution (the viscosity of the PEG monomer solution is influenced by the molecular weight of PEG monomer and the concentration of its solution). The neuron cultures on the patterned PEG gels demonstrated that the patterning of PEG gels could efficiently control the organization of neuron cells. In the future, the factors, including the molecular weigh of the poly(ethylene glycol) diacrylate, the spin-coating conditions, and the content of photoinitiator, should be studied systematically in related to the PEG properties and the cells adhesion and outgrowth. Indeed, we have met several challenge issues need to be address in the future. For example, the stability of the PEG gel on the surface is a big problem for the long-term neuron growth test. The stability of the PEG layer was collectively affected by the property of the adhesion promoter, the spin-coating conditions, and the exposure conditions. A poor quality of the adhesion promoter, an inhomogeneous spin-coating layer, and an

259

Chapter 8: Conclusion and Outlook

over exposure, all will lead to poor adhesion. However, the quality of the adhesion promoter was quite irreproducible duo to the complex condensation progress during the formation of the adhesion layer. The degree of the homogeneity of the resulted spin-coating layer was directly related to the quality of the adhesion promoter, the Mw of the PEG-DA monomer, and the desired thickness of the resulted layer. Normally, the higher mw of PEG-DA monomer and higher thickness always led to worse homogeneous. In addition, the pattern configurations able to control the outgrowth of neurites and potentially the synapse localization should be explored in the future. 8.7 Final conclusion and outlook As presented above, we established certain insight of the protein adsorption at solid-liquid interface via combined characterization technologies. This insight is expected to contribute to the future optimization in biosensor design. We also developed and optimized several surface modification strategies--selfassembled polymeric monolayer, temperature-active maleimde monolayer, and light-sensitive PEG gels--potential for versatile biosensor applications. We also proved that how the performance of a biosensor can be improved through controlling the property of the interface layers, choosing the suitable immobilization strategies, and tuning the bio-assay format. In summary, we have proved that how the sensitive, stability and reproducibility of a biosensor can be dedicated controlled through the combination of surface chemistry, organic chemistry, nanotechnology, biochemistry, and biology. However, as we suggested above, most research conducted in current study just “open” a new horizon for the future biosensor development. The feasibility and potential of our new “tools” for real biosensor application are proved in current study, but a lot of in-detail test and further optimization are still on the track. A clear indication is that a high performance and cost-effective biosensor can be realized through the convergence of the multi-disciplinary researches: chemistry, material science, nanotechnology, and biotechnology.

260

© 2006 Faculteit Wetenschappen, Geel Huis, Kasteelpark Arenberg 11, 3001 Heverlee (Leuven) Alle rechten voorbehouden. Niets uit deze uitgave mag worden vermenigvuldigd en/of openbaar gemaakt worden door middel van druk, fotokopie, microfilm, elektronisch of op welke andere wijze ook zonder voorafgaandelijke schriftelijke toestemming van de uitgever.

All rights reserved. No part of the publication may be reproduced in any form by print, photoprint, microfilm, electronic or any other means without written permission from the publisher. ISBN [90-8649-058-1] D/ [2006/10.705/55]