TURUN YLIOPISTO Turku 2008 TURUN YLIOPISTON JULKAISUJA ANNALES UNIVERSITATIS TURKUENSIS SARJA - SER. A I OSA - TOM. 383 ASTRONOMICA - CHEMICA - PHYSICA - MATHEMATICA Streptavidin - A Versatile Binding Protein for Solid-Phase Immunoassays by Lasse Välimaa From the Department of Biochemistry and Food Chemistry University of Turku Turku, Finland Supervised by Professor Kim Pettersson, PhD Department of Biotechnology University of Turku Turku, Finland and Professor Timo Lövgren, PhD Department of Biotechnology University of Turku Turku, Finland Reviewed by Petri Ihalainen, PhD Center of Excellence for Functional Materials (FUNMAT) Laboratory of Paper Coating and Converting Åbo Akademi University Turku, Finland and Adjunct Professor Markku Viander, MD, PhD Department of Medical Microbiology and Immunology University of Turku Turku, Finland Opponent Markus Linder, PhD, Docent Biotechnology VTT Technical Research Centre of Finland Espoo, Finland ISBN 978-951-29-3569-7 (PRINT) ISBN 978-951-29-3570-3 (PDF) ISSN 0082-7002 Painosalama Oy – Turku, Finland 2008 To my family Contents 4 CONTENTS LIST OF ORIGINAL PUBLICATIONS ....................................................................7 1 REVIEW OF THE LITERATURE ..................................................................10 1.1 Introduction..................................................................................................10 1.2 Streptavidin and biotin................................................................................11 1.2.1 Streptavidin..............................................................................................11 1.2.1.1 Discovery and early characterization ........................................11 1.2.1.2 Synthesis and post-translational modifications ..........................11 1.2.1.3 Structure .....................................................................................12 1.2.2 Biotin .......................................................................................................13 1.2.3 Streptavidin-biotin interaction .................................................................15 1.2.3.1 Affinity ........................................................................................15 1.2.3.2 Structural origin and stability ....................................................16 1.2.3.3 Effect of the mutations ................................................................18 1.2.3.4 Other methods to control binding...............................................20 1.2.3.5 Streptavidin-binding peptides.....................................................20 1.3 Solid-phase assays and the immobilization of antibodies.........................21 1.3.1 Solid-phase assays ...................................................................................21 1.3.1.1 Development of solid-phase assays ............................................21 1.3.1.2 Introduction of avidin- and streptavidin-coated surfaces...........23 1.3.2 Immobilization of antibodies...................................................................24 1.3.2.1 Adsorption and streptavidin-based immobilization....................24 1.3.2.2 Antigenic specificity and heterogeneity ......................................26 1.3.2.3 Other non-adsorptive immobilization techniques.......................27 1.3.2.4 Orientation..................................................................................28 1.3.3 Aspects of protein adsorption ..................................................................29 1.3.3.1 Mechanisms and conditions........................................................29 1.3.3.2 Adsorption isotherms..................................................................30 1.4 Advanced coating technologies ...................................................................33 1.4.1 Description of techniques ........................................................................33 1.4.2 The features of streptavidin relevant to adsorption .................................34 1.4.3 Covalent coupling....................................................................................35 1.4.3.1 Regular covalent chemistries......................................................35 1.4.3.2 Photochemistry ...........................................................................36 4 ABSTRACT ...................................................................................................................9 ABBREVIATIONS .......................................................................................................8 Contents 5 1.4.4 Biotinylated interface...............................................................................39 1.4.5 Various interface polymers and homogeneous conjugation ....................40 1.4.6 Recombinant approach and self-assembly...............................................42 1.4.7 Oligonucleotide-mediated immobilization ..............................................43 1.4.8 Peptide tags..............................................................................................45 2 AIMS OF THE STUDY .....................................................................................46 3 SUMMARY OF THE MATERIALS AND METHODS.................................47 3.1 Streptavidin and coating (I - IV) ................................................................47 3.1.1 Streptavidin..............................................................................................47 3.1.2 Solid substrates ........................................................................................47 3.1.3 Coating.....................................................................................................47 3.1.4 Spot surfaces (IV) ....................................................................................47 3.1.5 Modified streptavidin (I, III, IV) .............................................................48 3.1.6 Fractionation and size-exclusion chromatography (I) .............................48 3.2 Biotinylation and labeling of the antibodies and reporter molecules......48 3.2.1 Biotinylation and labeling with lanthanide chelates ................................48 3.2.2 Reporter molecules ..................................................................................49 3.3 Fluorescence measurement (I - IV) ............................................................50 3.4 Characterization of streptavidin-coated surfaces .....................................51 3.4.1 Binding capacities (I - IV) .......................................................................51 3.4.2 Protein assay (II)......................................................................................52 3.4.3 Desorption assay (III) ..............................................................................52 3.5 Immunoassays ..............................................................................................52 3.5.1 Time-resolved immunofluorometric assays (I, III, IV) ...........................52 3.5.2 Enzyme immunoassays (III) ....................................................................53 4 SUMMARY OF THE RESULTS......................................................................54 4.1 Characterization of streptavidin used (unpublished) ...............................54 4.2 Modeling and quantification of the adsorption (II) ..................................54 4.2.1 Theoretical modeling of the adsorbed layer ............................................54 4.2.2 Quantification of adsorbed streptavidin...................................................55 4.3 Binding capacities (I - IV) ...........................................................................56 4.3.1 Steric effects ............................................................................................56 4.3.2 Total capacities and binding site densities...............................................57 4.3.3 Spot surfaces (IV) ....................................................................................57 4.4 Desorption (III, and unpublished)..............................................................57 4.5 Modifications (I, III, IV, and unpublished) ...............................................59 4.5.1 Polymerized streptavidin (I, and unpublished data) ................................59 5 Contents 6 4.5.2 Thiolated streptavidin (III) ......................................................................60 4.6 Immunoassays (I, III, IV)............................................................................61 4.6.1 Linearity and the dose-response relationship...........................................61 4.6.2 Characteristics of various assay types .....................................................61 4.6.3 Non-specific binding and the sensitivity .................................................62 4.6.4 Kinetics (I, IV).........................................................................................63 4.7 Spot-assay (IV) .............................................................................................63 5 DISCUSSION......................................................................................................65 6 CONCLUSIONS.................................................................................................70 7 ACKNOWLEDGEMENTS ...............................................................................71 8 REFERENCES ...................................................................................................73 9 ORIGINAL PUBLICATIONS ..........................................................................91 6 List of original publications 7 LIST OF ORIGINAL PUBLICATIONS This thesis is based on the following original publications. They are referred to by Ro- man numerals in the text. I Lasse Välimaa, Kim Pettersson, Markus Vehniäinen, Matti Karp, and Timo Lövgren (2003) A high-capacity streptavidin-coated microtitration plate. Bio- conjugate Chem. 14: 103 - 111. II Lasse Välimaa, Kim Pettersson, Jaana Rosenberg, Matti Karp, and Timo Löv- gren (2004) Quantification of streptavidin adsorption in microtitration wells. Anal. Biochem. 331: 376 - 384. III Lasse Välimaa, and Katja Laurikainen (2006) Comparison study of streptavidin- coated microtitration plates. J. Immunol. Methods 308: 203 - 215. IV Lasse Välimaa, Johanna Ylikotila, Hannu Kojola, Tero Soukka, Harri Takalo, and Kim Pettersson (2008) Streptavidin-coated spot surfaces for sensitive im- munoassays using fluorescence surface readout. Anal. Bioanal. Chem. Published Online May 3, 2008. doi:10.1007/s00216-008-2120-y. In addition, unpublished data are included mainly in sections 4.1, 4.4 and 4.5. The original publications have been reproduced with the permission of the copyright holders. 7 Abbreviations 9 ABBREVIATIONS AFM atomic force microscopy BCA bicinchoninic acid Bio- refers to a biotinylated molecule BSA bovine serum albumin CA125 cancer associated antigen 125 DDI DNA-directed immobilization DELFIA dissociation-enhanced lanthanide fluorescence immunoassay EIA enzyme immunoassay ELISA enzyme-linked immunosorbent assay Eu europium Fab / Fab' fragment antigen binding; refers to the antigen-binding fragment of im- munoglobulin consisting of the variable and one constant domains F(ab')2 a fragment of immunoglobulin consisting of two linked Fab' fragments Fc fragment crystallizable; refers to the constant stem region of immu- noglobulin GA-SAv refers to pretreated, polymerized high-capacity streptavidin ΔG Gibbs energy change hCG human chorionic gonadotropin HRP horseradish peroxidase IgG immunoglobulin G IgM immunoglobulin M Ka affinity constant Kd dissociation constant koff dissociation rate constant kon association rate constant Mab monoclonal antibody NHS N-hydroxysuccinimide pI isoelectric point PSA prostate specific antigen QCM quartz crystal microbalance RCSB Research Collaboratory for Structural Bioinformatics RSA random sequential adsorption SAM self-assembled monolayer SATA N-succinimidyl-S-acetylthioacetate SAv refers to unmodified, native streptavidin SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis SEM scanning electron microscopy SPM scanning probe microscopy SPR surface plasmon resonance STM scanning tunneling microscopy Tb terbium TR-IFMA time-resolved immunofluorometric assay TSH thyroid-stimulating hormone 8 Abstract 8 ABSTRACT Streptavidin, a tetrameric protein secreted by Streptomyces avidinii, binds tightly to a small growth factor biotin. One of the numerous applications of this high-affinity sys- tem comprises the streptavidin-coated surfaces of bioanalytical assays which serve as universal binders for straightforward immobilization of any biotinylated molecule. Proteins can be immobilized with a lower risk of denaturation using streptavidin-biotin technology in contrast to direct passive adsorption. The purpose of this study was to characterize the properties and effects of streptavidin-coated binding surfaces on the performance of solid-phase immunoassays and to investigate the contributions of sur- face modifications. Various characterization tools and methods established in the study enabled the con- venient monitoring and binding capacity determination of streptavidin-coated surfaces. The schematic modeling of the monolayer surface and the quantification of adsorbed streptavidin disclosed the possibilities and the limits of passive adsorption. The defined yield of 250 ng/cm2 represented approximately 65 % coverage compared with a mod- eled complete monolayer, which is consistent with theoretical surface models. Modifi- cations such as polymerization and chemical activation of streptavidin resulted in a close to 10-fold increase in the biotin-binding densities of the surface compared with the regular streptavidin coating. In addition, the stability of the surface against leaching was improved by chemical modification. The increased binding densities and capaci- ties enabled wider high-end dynamic ranges in the solid-phase immunoassays, espe- cially when using the fragments of the capture antibodies instead of intact antibodies for the binding of the antigen. The binding capacity of the streptavidin surface was not, by definition, predictive of the low-end performance of the immunoassays nor the as- say sensitivity. Other features such as non-specific binding, variation and leaching turned out to be more relevant. The immunoassays that use a direct surface readout measurement of time-resolved fluorescence from a washed surface are dependent on the density of the labeled anti- bodies in a defined area on the surface. The binding surface was condensed into a spot by coating streptavidin in liquid droplets into special microtiter wells holding a small circular indentation at the bottom. The condensed binding area enabled a denser pack- ing of the labeled antibodies on the surface. This resulted in a 5 - 6-fold increase in the signal-to-background ratios and an equivalent improvement in the detection limits of the solid-phase immunoassays. This work proved that the properties of the streptavidin-coated surfaces can be modi- fied and that the defined properties of the streptavidin-based immunocapture surfaces contribute to the performance of heterogeneous immunoassays. 9 Review of the literature 10 1 REVIEW OF THE LITERATURE 1.1 Introduction Streptavidin is a tetrameric protein of approximately 60 kD in size secreted by bacte- rium Streptomyces avidinii. Streptavidin was discovered in the early 1960s when the fermentation filtrates of Streptomycetes were searched for antibiotic activity (Chaiet et al., 1963). The antibiotic activity of streptavidin is related to its ability to bind strongly a small growth factor, biotin, which is also known as vitamin H (György et al., 1940). Each of the four identical streptavidin subunits can bind one molecule of biotin. The equilibrium constant of the streptavidin-biotin interaction (affinity constant, Ka) equals to 2.5 × 1013 M−1 and it is among the strongest non-covalent bonds in nature along with avidin-biotin interaction. Avidin, a biotin-binding protein from hen egg white, was discovered in the early 1940s (Eakin et al., 1941) as a result of long studies involved in the biotin deficiencies and injuries found in animals fed with abundant egg white. Avidin and streptavidin both have similar tetrameric structures and biotin-binding fea- 33 % sequence identity with full-length avidin, which consists of 128 amino acids (DeLange and Huang, 1971; Argaraña et al., 1986; Green, 1990). Relevant differences between these proteins are related to their glycosylation and isoelectric point (pI). Avidin is a glycosylated and highly basic protein with a pI of 10.5 while streptavidin has no carbohydrates and is slightly acidic with a pI of 5 - 6. These proteins have pro- vided scientists with an interesting paradigm to the molecular origins of strong binding. Furthermore, (strept)avidin-biotin technology has been widely used in several biotech- nological applications. However, the positive charge and the presence of carbohydrates afford stronger non-specific binding features to avidin and therefore streptavidin is considered a more favorable alternative for several purposes. A number of principal reviews and entire volumes have been devoted either to the fundamental properties or the applications of avidin and streptavidin in recent decades (Green, 1975; Wilchek and Bayer, 1988; 1990; Diamandis and Christopoulos, 1991; Sano and Stayton, 1999). One of the technological applications benefiting from the streptavidin-biotin system includes bioanalytical assays such as the immunoassays used in clinical diagnostics. Streptavidin can serve either as a solid-phase binding matrix or as a principal compo- nent in the signal collection and amplification system (Suonpää et al., 1992; Schetters, 1999; Scorilas et al., 2000; Qin et al., 2001). Streptavidin-coated microtiter wells and sensor surfaces are used as universal surfaces for convenient binding of biotinylated molecules with well-retained activity. This thesis investigates the properties and appli- cations of streptavidin and above all focuses on its use as a binding surface in solid- phase immunoassays. 10 tures. Core streptavidin, consisting of 125 - 127 amino acids (subunit), shares an approximately Review of the literature 11 1.2 Streptavidin and biotin 1.2.1 Streptavidin 1.2.1.1 Discovery and early characterization Purification of the antibiotic activity in the fermentation filtrates of Streptomycetes led to the discovery of streptavidin in the early 1960s (Chaiet et al., 1963). The filtration broths contained two components of different sizes (termed 235L and 235S) which could be separated from each other by dialysis. None of the separated components alone showed antibiotic activity, except the smaller one when the bacteria were grown in a synthetic medium containing no biotin. This antibiotic effect was reversed by the addition of exogenous biotin, and thus it was postulated that the smaller component prevented the synthesis of intracellular biotin. Administration of either the larger com- ponent from the dialysis or alternatively avidin from the hen egg white compensated the effect of exogenous biotin and restored the antibiotic activity. Since this larger component showed biotin-binding properties similar to avidin it was named strepta- vidin. The characterization of the larger component showed that it was a tetrameric protein with a molecular weight of approximately 60 000 D, it could bind four mole- cules of biotin, it contained no carbohydrates and it was neutral or acidic in nature, in contrast to the basic protein avidin. (Chaiet et al., 1963; Chaiet and Wolf, 1964; Tausig and Wolf, 1964.) 1.2.1.2 Synthesis and post-translational modifications The cloning and the sequencing of the streptavidin gene from the genomic libraries of Streptomyces avidinii (Argaraña et al., 1986) showed that the gene codes a polypeptide chain of 183 amino acids in length (Fig. 1). The first 24 residues constitute a signal sequence for secretion. The mature protein subunit consists of 159 amino acids and has a molecular weight of 16 500 D. The characterization of various streptavidin prepara- tions had, however, shown smaller sizes of the protein as well as different amino acid sequences and compositions compared with the evidence from the gene sequence (Hofmann et al., 1980; Argaraña et al., 1986; Pähler et al., 1987). This indicated post- translational modifications of the mature protein into core streptavidin, the mechanisms of which were eventually characterized more comprehensively by Bayer and co- workers (Bayer et al., 1989). They showed that mature streptavidin undergoes postse- cretory proteolytic digestions both in the N- and C-termini by secreted extracellular proteases. Several studies, principally the above-mentioned, have characterized core streptavidin as follows. The core consists of 125 - 127 amino acids and the predomi- nant truncated constituents start either at Ala-13 or Glu-14. The molecular weight of core streptavidin is about 13 200 D and it is stable against further proteolytic degrada- tion. Core streptavidin retains full biotin-binding activity compared with the non- truncated mature form. The binding activity of biotinylated macromolecules to the core has been found to be even higher than to the non-truncated form. Core streptavidin is highly soluble in water in contrast to the full-length protein which tends to form aggre- gates. 11 Review of the literature 12 MRKIVVAAIA VSLTTVSITA SASADPSKDS KAQVSA IDAAKKA GVNNGNPLDA VQQ DPSKDS KAQVSA IDAAKKA GVNNGNPLDA VQQ AEAG ITGTWYNQLG STFIVTAGAD GALTGTYESA VGNAESRYVL TGRYDSAPAT DGSGTALGWT VAWKNNYRNA HSATTWSGQY VGGAEARINT QWLLTSGTTE ANAWKSTLVG HDTFTKVKPS AAS -24 1 13 139 159 A) B) C) Figure 1. The amino acid sequence of streptavidin. A) The gene codes a polypeptide of 183 amino acids in length. The signal sequence is indicated by the negative numbering. B) The full- length mature protein consists of 159 residues. C) The mature protein goes through N- and C- terminal proteolytic digestions to core streptavidin consisting of 125 - 127 amino acids. One possible core including the residues 13 - 139 is shown as the white box. The amino acid se- quence is from the Swiss-Prot, entry P22629. The purification method used for isolation is supposed to affect the nature of a particu- lar streptavidin preparation, whether it is predominantly the full-length protein or the truncated core form. Bayer et al. (1986) isolated a streptavidin preparation which ex- hibited the molecular weights of 18 000 D for the subunit and approximately 75 000 D for the tetramer in SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electropho- resis) analysis. The values implied a non-digested protein rather than core streptavidin. The authors used a modified iminobiotin-based matrix for the affinity purification of streptavidin. Iminobiotin, first synthesized by Hofmann and Axelrod (1950), is a biotin analog possessing a pH-dependent binding affinity to streptavidin. The modified puri- fication method avoided a concentrating step in the process in contrast to the previous affinity purification and ammonium sulfate precipitation methods. It was suggested that the concentrating steps used in the previous purification processes simultaneously concentrated the proteases resulting in a more rapid degradation of mature streptavidin. In addition to the proteases of S. avidinii, several other proteases are able to convert mature streptavidin into the core molecule or close to it. These include Proteinase K, papain, subtilisin, thermolysin and elastase. Trypsin introduces partial digestion whereas pepsin and chymotrypsinogen were found unable to hydrolyze the mature protein into core streptavidin (Bayer et al., 1989). According to Bayer et al. (1990) commercial streptavidin preparations usually consist of core streptavidin. 1.2.1.3 Structure The three-dimensional structure of streptavidin was resolved in the late 1980s (Pähler et al., 1987; Weber et al., 1987; Hendrickson et al., 1989; Weber et al., 1989). The subunit consists of eight adjacent antiparallel β-strands which are hydrogen-bonded to each other, forming a β-barrel structure. The β-strands are extended with hairpin struc- 12 Review of the literature 13 tures at the ends of the strands and the biotin-binding site is located in one end of the barrel interior. Subunit pairs are hydrogen-bonded to each other forming a symmetric dimer and a pair of dimers interface each other, constituting the naturally occurring tetrameric structure which is sometimes described as a dimer of dimers. The dimeric nature of the subunit pairs is clearly recognized in the three-dimensional structures, where the interface of the dimers appears as a narrowing in the hour-glass resembling structure (Fig. 2). Currently, the RCSB (Research Collaboratory for Structural Bioinformatics) Protein Data Bank (PDB) contains approximately 130 structures related to streptavidin, includ- ing the wild type protein and various mutated variants as apo-forms (without ligand) and with a number of bound ligands, ranging from biotin and its analogs to oligopep- tides. Figure 2. The three-dimensional structures of streptavidin. The space-filling model (left) shows the organization of the four identical subunits. The ribbon diagram (right) displays the predomi- nant β-barrel structure and the intercalation of the biotin molecules in the binding pockets. The images were generated by the program Discovery Studio Visualizer (Accelrys, Inc.) using the RCSB Protein Data Bank entries 1SWC and 1SWE which are based on the work by Freitag et al. (1997). 1.2.2 Biotin Biotin was first discovered as a growth factor in 1901 and afterwards it was purified from egg yolk and liver in the 1930s and early 1940s (reviewed by Streit and Entcheva, 2003). At this time the same molecule was studied in three contexts: vitamin H (a cura- tive agent of "egg white injury"), biotin (a yeast growth factor isolated from egg yolk), and coenzyme R (a growth factor from rhizobia). In 1940 György and co-workers (György et al., 1940) suggested that all these were identical or closely related com- pounds. The history of biotin is closely related to the discovery and early phases of 13 Review of the literature 14 avidin. Indications of the necessity of biotin for life was obtained when rats that were abundantly fed with egg white generated specific symptoms such as dermatitis and baldness (Boas, 1924; 1927). In 1940 Eakin and co-workers (Eakin et al., 1940a) re- ported that the recognized condition, termed "egg white injury", was related to the de- ficiency of biotin in the tissues. The egg white which was included in the diet of the chicks was considered responsible for either destroying biotin or making it unavailable to the tissues. A constituent of the egg white was found to form a stable complex with biotin (Eakin et al., 1940b) and this "avidalbumin" was considered responsible of the biotin deficiency (György et al., 1941). This constituent was subsequently purified further and discovered as avidin (Eakin et al., 1941). Biotin is an essential compound to all organisms. The natural function of biotin is to carry carboxyl groups and function as a cofactor for carboxylation reactions catalyzed by a group of enzymes, biotin-dependent carboxylases. Most prokaryotes and plants are able to synthesize biotin themselves while animals and humans need biotin in their diet. Comprehensive reviews of the biosynthesis and enzymatic mechanisms of biotin have been written for example by Knowles (1989) as well as Streit and Entcheva (2003). The structure of biotin was presented for the first time as early as 1942 (du Vigneaud, 1942; Melville et al., 1942) and the first chemical synthesis, starting from l-cystine, chloroacetic acid and glutaric acid was reported in 1944 (Harris et al., 1944). Biotin group in the end (Fig. 3). The molecular weight of the compound is 244 D. In addition to the biotechnological purposes, biotin is used in bulk quantities as an additive in crobial production is supposed to reduce the environmental burden caused by the chemical synthesis, but the yields of the biological production have not yet been eco- nomically profitable. (Streit and Entcheva, 2003.) Numerous modifications of biotin have been synthesized and implemented for research purposes and biotechnological applications (Fig. 3). One group of the derivatives com- prises those which have lower, environmentally sensitive binding affinity to avidin and streptavidin. These include, for example, 2-iminobiotin (Hofmann and Axelrod, 1950; Hofmann et al., 1980) and desthiobiotin (Hirsch et al., 2002). These biotin derivatives exhibit reversible binding and they are useful in the purification processes which rely on streptavidin-biotin technology by enabling mild elution conditions. The molecular mainly responsible for the tight binding. Other main variants of biotin enable covalent coupling of biotin to target molecules (biotinylation) through chemically reactive groups introduced to the end of the side chain. These include, for example, the N- hydroxysuccinimide (NHS) ester derivatives (Becker et al., 1971; Heitzmann and 14 10 - 30 tons. Commercially available biotin is mainly prepared by chemical synthesis. Mi- consists of two fused rings and an aliphatic valeric acid side chain with a carboxyl food, feed and cosmetology. Its annual production is estimated at approximately alterations have been typically targeted to the double-ring structure of biotin which is Richards, 1974; Bayer and Wilchek, 1990) and biotin-isothiocyanate (Mukkala et al., 1993) Review of the literature 15 for coupling to thiol-groups. Furthermore, some biotin derivatives possess specific functions in their side chain such as cleavable phenyl esters (Mouton et al., 1982), 1985; see section 1.4.3.2 for further use of photochemistry). In addition, various long- chain dendritic structures of biotin have been synthesized containing multiple branches and coupling sites (Wilbur et al., 1999). These are mainly intended for therapeutic pretargeting purposes. A B C OH NHNH S HH O O OH NHNH S HH NH O O N O O N H NHNH S HH O O O Figure 3. The structures of biotin and some derivatives. A) Intact biotin, drawn according to du Vigneaud et al. (1942). B) 2-iminobiotin has lower affinity than intact biotin to streptavidin and avidin. Drawn according to Hofmann and Axelrod (1950). C) Biotinamidohexanoic acid N- hydroxysuccinimide ester has an extended side chain with a reactive N-hydroxysuccinimide (NHS) ester functionality enabling the covalent coupling of biotin to other molecules through their amine groups. Drawn according to Costello et al. (1979). 1.2.3 Streptavidin-biotin interaction 1.2.3.1 Affinity The binding affinity between biotin and streptavidin (and avidin) is obviously high, but there are few studies which actually report the equilibrium constants between intact biotin and wild type streptavidin. Two review publications by Green (1975; 1990) are the "golden standards" when referring to the strong interaction and high binding affin- ity. The latter publication reports a dissociation constant (Kd) of 4 × 10−14 M for the streptavidin-biotin interaction which equals to the widely accepted value of 2.5 × 1013 M−1 when expressed in terms of the affinity constant (Ka). The interaction between biotin and streptavidin is so strong that regular biochemical methods, for ex- ample, those which rely on an equilibrium binding accompanied by a separation of bound biotin from non-bound either by dialysis or ultrafiltration, do not evidently pro- vide adequate separation efficiency and sensitivity for measuring the minute quantities 15 for coupling to primary amines as well as maleimide- biotin (Bayer et al., 1985) disulfide bridges (Soukup et al., 1995) and photochemical reactivity (Forster et al., Review of the literature 16 of unbound biotin. These assays are, therefore, conflicted by an upper limit in the measurable equilibrium constants. It has been stated that the equilibrium constant be- tween (strept)avidin and biotin can only be estimated from the ratio of the rate con- stants for binding and dissociation (Green, 1963; 1990). Dissociation rate constants (koff) have been determined in several studies. For example, Piran and Riordan (1990) reported a value of 2.4 × 10−6 s−1 measured at +25 °C whereas Chilkoti and Stayton (1995) reported a value of 5.4 × 10−6 s−1 at +25 °C and Klumb et al. (1998) defined a value of 4.3 × 10−5 s−1 at +37 °C for the dissociation of biotin from wild type strepta- vidin. The determination of the association rate constant (kon) is obviously more com- plicated due to the rapid, high-affinity reaction. An association rate constant of 7.5 × 107 M−1s−1 was recently measured by means of stopped-flow fluorescence tech- nology relying on the decrease of the intrinsic fluorescence of streptavidin upon the binding of biotin (Hyre et al., 2006). Based on the measured association rate constant, the study reported an affinity constant of 1.9 × 1013 M−1 for the streptavidin-biotin in- teraction. Mutated streptavidin and biotin analogs frequently exhibit lower binding affinities which are measurable by regular chemical methods. The measurement methods used in this lower affinity region include, for example, equilibrium dialysis in a microdialyzer (Sano and Cantor, 1995), equilibrium binding followed by ultrafiltration (Chilkoti et al., 1995b; Sano et al., 1997), radiometric competition assay (Klumb et al., 1998; Freitag et al., 1999a; Howarth et al., 2006) and a special modification of the common enzyme-linked immunosorbent assay (ELISA) (Chilkoti and Stayton, 1995). Further- more, the energies of streptavidin-biotin interaction have been measured using iso- thermal titration calorimetry; these studies have reported Gibbs energy changes (ΔG) of approximately –18 kcal/mol (≈ –75 kJ/mol) for the binding of biotin to wild type streptavidin (Chilkoti and Stayton, 1995; Chu et al., 1998). In addition, the binding forces between streptavidin and biotin have been measured using atomic force micros- copy (AFM) and a special surface force measurement apparatus (Chilkoti et al., 1995a; Allen et al., 1996; Wong et al., 1999). These studies have reported the binding forces in the range of 90 - 400 pN for the binding of biotin to various mutated and wild type streptavidin. 1.2.3.2 Structural origin and stability The structural basis and the network of the bonds between streptavidin and biotin were defined in the structural studies published in 1989 (Hendrickson et al., 1989; Weber et al., 1989). The binding site is solvent-accessible when biotin is not bound. The binding of biotin displaces the water molecules and biotin becomes deeply buried in the bind- ing site. The tight interaction is essentially enabled by three special binding motifs. One motif is the binding loop (residues 45 - 52) which is disordered when biotin is not bound but becomes ordered upon binding and closes the binding site (Freitag et al., 1997). The second motif comprises the hydrophobic interactions which are provided by Trp-79, Trp-92 and Trp-108 from the given subunit and Trp-120 from the adjacent subunit across the dimeric interface. The third binding motif is an extensive network of 16 Review of the literature 17 the hydrogen bonds between the heteroatoms of biotin and the amino acid residues of streptavidin (eight bonds). In addition to the direct hydrogen bonds to biotin, the hy- drogen bond network extends between several residues in the vicinity of the binding site, thus stabilizing the environment. A clear summary of the interactions between biotin and the amino acids of streptavidin has been written by Freitag and co-workers (Freitag et al., 1999b). Crystallographic studies have shown that the quaternary structure of streptavidin adopts a stabilized, more compact shape upon the binding of biotin so that the subunit barrels become slightly flattened and more tightly bound to each other (Weber et al., 1989; Katz, 1997). These observations have suggested some co-operation or communi- cation between the subunits. Evidence of the subunit communication was first obtained with avidin in a study by Green and Toms (1973) who reported lower affinity with an isolated subunit in comparison with the intact tetrameric molecule. The lower affinity of the monomeric form has also been confirmed with streptavidin (Qureshi and Wong, 2002; Wu and Wong, 2005). In streptavidin, the Trp-120 residue from the adjacent subunit plays a major role in the high biotin-binding affinity and binding-induced sta- bility. The replacement of Trp-120 with phenylalanine reduces the binding affinity to the 108 M−1 level and weakens the tetrameric stability (Sano and Cantor, 1995). These two determinants are closely related to each other; tetrameric integrity is essential for complete biotin-binding activity and on the other hand, the binding of biotin increases the stability of the tetrameric composition. The streptavidin-biotin bond and the tetrameric integrity tolerates harsh conditions such as 6 M urea (Sano and Cantor, 1990; Kurzban et al., 1991), a wide pH range and an excess of free biotin (Sano and Cantor, 1995). The tetrameric state of the protein with bound biotin is at least partially maintained upon heating in the presence of 8 M urea and 6 M guanidine hydrochloride (González et al., 1997; 1999). In contrast, the same studies showed that streptavidin without bound biotin denatured at significantly lower temperatures. In another study (Sano et al., 1995), streptavidin tolerated harsh conditions to some extent without bound biotin. It retained approximately 80 % of its biotin-binding activity after expo- sure to 6 M guanidine hydrochloride at pH 7.4. However, at pH 1.5, the remaining capacity decreased to 20 %. The disruption of the streptavidin-biotin bond requires harsh conditions but some methods have been reported for the release of bound biotin or biotinylated molecules in somewhat milder conditions. For example, a mixture of formamide (95 %) and either ethylenediaminetetraacetic acid (EDTA) or sodium acetate combined with heating up to +90 °C has been used to elute biotinylated DNA from streptavidin-coated magnetic beads, resulting in yields of 95 - 98 % (Tong and Smith, 1992). Biotinylated oligonu- cleotides have been also recovered from streptavidin-based matrices in the presence of either 2.0 M 2-mercaptoethanol or 2.0 M dithiotreitol (Jenne and Famulok, 1999). At room temperature the release was partial but it was greatly increased upon heating to +95 °C. The release of biotinylated macromolecules appeared to be easier than the release of intact biotin, since biotin remained completely bound at room temperature. 17 Review of the literature 18 Even upon heating, no more than 3 % was released. The conditions reported in the study enabled the recovery of the biotinylated target molecules in an active state as well as subsequent recycling of the streptavidin matrix. Another example of fairly mild conditions being able to release bound biotin from streptavidin was shown in a recent magnetic beads only by brief exposure to elevated temperatures (+70 °C) in nonionic water solution (Holmberg et al., 2005). Owing to the conformational changes and the involvement of the tryptophan moieties in the binding interaction, the intrinsic fluorescence emission of streptavidin is de- creased upon the binding of biotin (Kurzban et al., 1990). The emission peak of strep- tavidin without bound biotin appears at 333 nm whereas the binding of biotin shifts the peak to 329 nm (blue-shift) and decreases the intensity. The total fluorescence intensity is reduced by 39 % and the peak intensity by 33 %. This feature, the quenching of the intrinsic fluorescence upon binding of biotin, has been used for quantitative determina- tion of the biotin-binding capacities of streptavidin layers (Moll et al., 2002; Huber et al., 2006) and the determination of the association rates for binding using the stopped- flow measurement technology (Hyre et al., 2006). 1.2.3.3 Effect of the mutations The streptavidin-biotin system has been an attractive research target when discerning the origins of the strong interaction between biomolecules. Analyses of mutated strep- tavidin and their interactions with biotin have provided further insight into the binding mechanisms and the importance of various amino acid residues for high binding affin- ity. As discussed above, Trp-120 from the adjacent subunit contributes significantly to the strong binding as well as to the stabilizing interaction between the subunits. A mu- tation of Trp-120 to phenylalanine reduced the affinity constant to the 108 M−1 level in one study (Sano and Cantor, 1995) and increased the dissociation rate constant (koff) by 70-fold compared with wild type streptavidin in another study (Chilkoti and Stayton, 1995). Mutations of Trp-79-Phe and Trp-108-Phe increased the dissociation rate con- stant by 5.5 - 17-fold in comparison with wild type streptavidin (Chilkoti and Stayton, 1995). The effect of the tryptophan mutations in the other positions except 120 was less significant to the binding affinity which confirmed the overall importance of inter- subunit contact in the strong interaction between streptavidin and biotin. Tryptophans in positions 79, 108 and 120 have also been replaced with alanine (Chilkoti et al., 1995b). The affinity constants of the Trp-79-Ala and Trp-120-Ala mutants were de- creased down to the 107 M−1 level. The mutations to the binding site have not usually induced significant changes in the three-dimensional folding of tetrameric streptavidin. For example, X-ray crystallographic structure analysis of the mutants Trp-79-Phe, Trp- 108-Phe, Trp-120-Phe and Trp-120-Ala showed only slight changes in the structure compared with the wild type protein, which implies conservation in the folding proper- ties of streptavidin (Freitag et al., 1998). 18 study where biotinylated DNA was reversibly recovered from streptavidin-coated Review of the literature 19 In addition to the binding motif of the hydrophobic interactions, the hydrogen bond network has been subjected to mutations. For example, elimination of one hydrogen bond by mutating Asp-128 to alanine resulted in a 1000-fold increase in the dissocia- study showed again no major changes in the overall folding, but instead, a 0.5 - 1.0 Å shift was observed in the position of the bound biotin which was regarded as a kind of snapshot from the biotin dissociation pathway consisting of successively breaking hy- drogen bonds. The oxygen atom in the double-ring of biotin is coordinated by three hydrogen bonds, and the contribution of these principal bonds to the binding affinity was elucidated by mutating the residues Asn-23, Ser-27 and Tyr-43 one by one (Klumb et al., 1998). Furthermore, Ser-45 and Asp-128 were replaced with alanine individually or together in combination to produce a double mutant (Hyre et al., 2006). While the single mutants exhibited moderate decreases in the affinities, the biotin-binding affinity of the double mutant was decreased to 8.2 × 105 M−1. The third binding motif, the loop structure which adopts a closed conformation upon the binding of biotin, also has an important role in the high binding affinity. Deletion of the loop, as studied by circularly permuted streptavidin (Chu et al., 1998), decreased the affinity constant down to ap- proximately 2.3 × 107 M−1. Though a sequence of four amino acids was removed from the structure, the X-ray analysis did not show remarkable alterations in the overall fold- ing of this mutant either. A recent report introduced an interesting monovalent strepta- vidin exhibiting a biotin-binding affinity almost indistinguishable from wild type strep- tavidin (Howarth et al., 2006). This chimeric streptavidin was constructed by refolding subunits bearing three mutations with subunits of a wild type binding affinity at a 3:1 ratio. The mutated subunits had negligible biotin-binding affinity but they were capable of maintaining tetrameric integrity. This monovalent streptavidin enabled cellular im- aging and the tracking of biotinylated cell-surface proteins by labeled streptavidin without adverse cross-linking, a property which was typically exhibited by tetravalent streptavidin. In addition to the biotin-binding site, other mutations have been targeted to the subunit interfaces and to the overall structure of the protein. An interesting example is a mutant possessing increased stability of the tetrameric structure. Covalent bonds through the weaker dimeric interface were introduced by replacing His-127 either with cysteine or lysine, while the biotin-binding ability of streptavidin remained intact (Reznik et al., 1996). An opposite objective has been to weaken the subunit interactions by mutations to produce monomeric streptavidin variants which have gained interest due to the re- versible biotin-binding properties (Qureshi and Wong, 2002; Wu and Wong, 2005). A recently published comprehensive review (Laitinen et al., 2006) summarizes the numerous mutations made to streptavidin and avidin as well as describes their essential features. According to the review, none of the streptavidin mutants exhibit increased binding affinity compared with the wild type protein. A few mutants retain, or almost retain, wild type affinity. Another review from the same group surveys the possibilities 19 tion constant (Freitag et al., 1999a). The crystallographic data obtained in the Review of the literature 20 and applications of modified streptavidin and avidin in modern bioscience, medicine and nanotechnology (Laitinen et al., 2007). 1.2.3.4 Other methods to control binding In addition to the biotin derivatives of lower affinity, monomerization and the binding- site mutations as discussed above, some other strategies are available to control bind- ing affinity. These are chemical modifications which result in environmentally sensi- tive binding ability. One approach is to use nitrated streptavidin (Morag et al., 1996a; 1996b). Treatment of streptavidin (or avidin as well) with tetranitromethane results in the nitration of the tyrosine moieties. Some of them are involved in the binding of bio- tin and, therefore, the binding becomes dependent on the pH so that biotin is bound at low pH (4 - 5) and released either after increased pH or by the introduction of excess biotin. Another interesting technology to control biotin-binding activity is to use syn- thetic polymers attached close to the binding site (Ding et al., 1999; Bulmus et al., 2000; Ding et al., 2001; Stayton et al., 2004). Polymers of poly(N- isopropylacrylamide) and poly(N,N-diethylacrylamide) reversibly change their state between soluble linear and collapsed globular forms in response to changes in tempera- ture, pH or light. These polymers can shield or expose the binding site upon mild envi- ronmental changes when coupled to the proximity of the binding site, as shown by site- specific coupling to a mutated streptavidin (Glu-116-Cys). Thus, these modifications were called "smart" streptavidin conjugates. 1.2.3.5 Streptavidin-binding peptides In addition to biotin and its derivatives, various peptide motifs bind to streptavidin although with substantially lower affinity than biotin (Ka ≈ 103 - 108 M−1) (Weber et al., 1992; Giebel et al., 1995; Keefe et al., 2001; Lamla and Erdmann, 2004). The well- defined peptides which have been applied in biotechnology are Strep-tag (Schmidt and Skerra, 1993) and Strep-tag II (Schmidt et al., 1996), nine and eight amino acid pep- tides containing the common His-Pro-Gln sequence found in most streptavidin-binding peptides. Strep-tag II exhibits increased affinity to an engineered recombinant strepta- vidin ("Strep-Tactin") (Voss and Skerra, 1997), and this has further improved the ver- satility and utility of the Strep-tag affinity system in biotechnology. The Strep-tag sys- tem has been reviewed by Skerra and Schmidt (Skerra and Schmidt, 1999; Schmidt and Skerra, 2007). The main applications of the streptavidin-binding peptides are involved in affinity purification. The relatively low binding affinities of the peptides enable re- versible release of the peptide-tagged fusion proteins from the streptavidin-based ma- trices by the addition of biotin or an appropriate analog. Furthermore, other reported uses of the peptides include peptide-tagged detection antibodies to serve the binding of an enzyme-labeled streptavidin for immuno-detection in membranes or in ELISA-type assays. In addition, a recent application used a nano-tag (a nine amino acid peptide) and a SBP-tag (a 38-amino acid peptide) to immobilize binding molecules to strepta- vidin-coated surface plasmon resonance (SPR) chips (Li et al., 2006). Reversible bind- ing enabled the release of the bound molecules and thereby convenient re-use of the expensive chips. 20 Review of the literature 21 1.3 Solid-phase assays and the immobilization of antibodies 1.3.1 Solid-phase assays 1.3.1.1 Development of solid-phase assays Immunoassays are widely used for recognizing and quantifying minute quantities of proteins and small molecules from various biological fluids. At present, the detection limits of some of the most sensitive assays reach the sub-femtomolar concentration range, for example a prostate specific antigen (PSA) assay using europium (Eu) -filled nanoparticle labels and time-resolved fluorescence detection (Soukka et al., 2001). Active development in immunoassays originates in the 1950s. A recent review by Wide (2005) surveys the development of the immunoassays in those days. According to the author, one immunoanalysis method used at that time was Boyden's passive haemagglutination inhibition method which used antibodies and antigen-coated eryth- rocytes. Depending on the absence or presence of the antigen in the sample followed by the binding of the antibodies to the erythrocytes or inhibition of the binding by the antigen from the sample, the coated cells sedimented to the bottom of the test tubes in a specific manner. This technique was used in the first industrially manufactured diag- nostic test kit launched in 1962 as a pregnancy test and was based on the detection of human chorionic gonadotropin (hCG) in urine. Another important branch in the development of the immunoassays in the 1950s was involved in the studies of insulin metabolism which eventually led to the measurement of insulin by means of competitive binding between radioactively labeled and non- labeled insulin (Berson et al., 1956; Berson and Yalow, 1958). This technique was first used to assay animal insulin (beef, pork) employing antiserums from human subjects. As expected, it was not applicable for the measurement of human insulin due to weak binding until Yalow and Berson (1959) discovered that antiserum from guinea-pigs immunized with beef insulin cross-reacted strongly with human insulin. This provided the basis for the quantitative measurement of human insulin by means of competitive binding between endogenous insulin and 131I-labeled insulin (Yalow and Berson, 1960). Independently of the above work, Ekins (1960) determined the concentrations of thyroxine (T4) in human plasma by measuring the distribution of 131I-labeled thyrox- ine between albumin and a thyroxine-binding protein in a saturation assay. Though not using antiserum or antibodies, and thus not literally an immunoassay, the work by Ekins is also regarded as a milestone publication in the context of the immunoassays, or in wider sense, in measuring the concentrations of substances by means of a com- petitive ligand binding accompanied by a labeled component. Heterogeneous immunoassays require separation of the bound analyte from the non- bound before measuring the signal from a labeled component. In the first assays, sepa- ration was enabled by chromatographic or electrophoretic means. In subsequent assays, solid materials were used to facilitate separation. For example, charcoal was capable of binding free insulin but not antibody-bound insulin and, therefore, it was used for sepa- ration in the immunoassays of insulin (Herbert et al., 1965; Keane et al., 1968). Solid 21 Review of the literature 22 phases with immobilized antibodies were likewise reported at that time. According to Wide (2005), competitive binding assays of several hormones were evaluated using antibody-coated formalinized sheep erythrocytes and 125I-labeled hormones in 1965 at the University Hospital in Uppsala. Subsequently, solid phases made from polymeric materials were introduced. For example, Catt et al. (1966) reported the use of an anti- body-coated graft polymer of styrene and polytetrafluoroethylene in an assay of human growth hormone. In another work, Wide and Porath (1966) used antibodies coupled to Sephadex-isothiocyanate for the assays of hCG and luteinizing hormone. The solid materials in the first solid-phase assays were provided in the format of particles or powder in suspension, and separation by centrifugation constituted an advantage over the earlier chromatographic or electrophoretic methods. A shift to a more solid format was introduced by Catt and co-workers, first with solid discs prepared from synthetic polymers (Catt et al., 1967) and then with tubes made from polystyrene or polypropyl- ene (Catt and Tregear, 1967). In the latter work, the antibodies were immobilized to the inner surfaces of the plastic tubes by passive adsorption and the tubes were success- fully used for the assays of human placental lactogen and human growth hormone. The tube format enabled convenient separation and washing under tap water. Earlier studies had essentially used reactive chemistries to couple antibodies to the solid phases, whereas the current study relied on unsubstituted polymers where the antibodies ad- sorbed spontaneously. The spontaneous adsorption of the proteins to the solid phases had already been recognized and reviewed earlier; for example, according to Neurath and Bull (1938) some findings of protein adsorption to solid phases had been made in 1905. Afterwards, Bull (1956) reported and modeled the adsorption of albumin to glass. Furthermore, the search for blood-compatible, non-thrombogenic, plastics had revealed the adsorption of blood proteins to unmodified plastic surfaces (Leininger et al., 1966). The work by Catt and Tregear (1967), however, introduced the adsorption of antibodies to be utilized among an emerging solid-phase assay application. Further significant advances in immunoassays were introduced by the end of the 1960s. One was the principle of an immunoradiometric reagent-excess assay using 125I- labeled antibodies (Miles and Hales, 1968) which yielded an ascending linear relation- ship between the amount of the added antigen and the measured radioactivity, instead of the descending relation in the previous competitive binding assays. Another devel- opment was the introduction of a non-competitive two-site assay by Wide and co- workers (Wide et al., 1967). This was first employed to detect allergen antibodies us- ing a set-up where a solid-phase coupled allergen bound specific antibodies from se- rum followed by detection with a 125I-labeled antibody against the allergen antibody. Soon after, the principle of using a solid-phase coupled catching antibody in the two- site assay was presented and the term "sandwich assay" was launched (Wide, 1969); the assay format that currently is probably the most common non-competitive immu- noassay type. 22 Review of the literature 23 1.3.1.2 Introduction of avidin- and streptavidin-coated surfaces Immobilized avidin and streptavidin were first used in separation and purification ma- trices before their entry into solid-phase assays. For example, sepharose and agarose matrices with covalently coupled streptavidin were used to immobilize biotinylated capture molecules such as growth hormones, monoclonal antibodies or lectins and their soluble targets to facilitate separation and purification (Haeuptle et al., 1983; Updyke and Nicolson, 1984; Buckie and Cook, 1986). Avidin-biotin technology was intro- duced in the context of the enzyme immunoassays in the late 1970s. In the work by Guesdon et al. (1979) avidin was used to mediate the binding of both biotinylated and covalently coupled reporter enzymes to the immunocomplex in bridging sandwich assays. Avidin-coated solid phases of the immunoassays were introduced in the mid 1980s. Sutton and co-workers (Sutton et al., 1985) used avidin-coated microtiter plates to immobilize biotinylated bacterial polysaccharides in an ELISA-assay. Odell and colleagues (Odell et al., 1986) used avidin-coated polystyrene beads in a rapid radio- immunoassay of human thyrotropin (hTSH). The beads were first coated with bioti- nylated bovine serum albumin (BSA) to immobilize avidin, which further bound the biotinylated capture antibody through the remaining unoccupied biotin-binding sites. The same principle was subsequently used in the assays of hCG (Griffin and Odell, 1987), lutropin (Odell and Griffin, 1987) and corticotropin (Zahradnik et al., 1989). Streptavidin-coated solid phases of the immunoassays were likewise reported in the mid 1980s by Suter and Butler (1986). In fact, rather than an actual immunoassay, the authors essentially studied the amounts and the activities of the monoclonal antibodies which were either directly adsorbed to a plastic surface or immobilized by a novel "protein avidin biotin capture" (PABC) system. In this concept, streptavidin was im- mobilized via biotinylated interface polymers (proteins or gelatin) to the surface of the immunoassay well and the remaining biotin-binding sites of streptavidin served for the binding of a biotinylated monoclonal antibody. Though the direct adsorption resulted in larger amounts of the immobilized antibodies, the PABC system was superior in terms of the antigen binding activity of the immobilized antibody. Subsequently, the system was comprehensively studied and optimized in stoichiometric terms (Suter et al., 1989). A covalent coupling of streptavidin to immunoassay surfaces was soon re- ported using bromoacetyl-activated polystyrene beads (Peterman et al., 1988). The antigen-binding activities of the antibodies immobilized to the streptavidin surface were again excellent in comparison with the activities of the passively adsorbed or directly coupled antibodies. Covalent chemistry was, furthermore, used to couple strep- tavidin to the surfaces of magnetic beads to facilitate the separation of biotinylated human chromosomes from human-hamster hybrid cell lines (Dudin et al., 1988) and to enable magnetic bead based solid-phase DNA-sequencing (Hultman et al., 1989). These studies thus extended the use of the streptavidin-coated solid phases to the area of nucleic acid research. 23 Review of the literature 24 1.3.2 Immobilization of antibodies 1.3.2.1 Adsorption and streptavidin-based immobilization Immobilized proteins continuously constitute an important part of solid-phase immu- noassays and it is reasonable to review some particular aspects involved in the immobi- lization of the antibodies. The biologically active binding layer and its quality are es- sential for the optimal performance of the assay and it may become a challenge to in- troduce a binding surface of correct and adequate functionality. Owing to its simplicity and cost-efficiency, direct coating of the capture antibodies by means of passive ad- sorption (also called physisorption) has been widely used to functionalize solid phases. Despite the well-characterized disruptive effects of the adsorption on protein confor- mation and activity, some scientists have until recently recommended it as the first choice before proceeding to more complicated methods (see a review by Butler, 2000). In addition to providing a universal surface for binding of any biotinylated molecule, the non-adsorptive immobilization of the antibodies via streptavidin constitutes an advantage with respect to the activity of the immobilized antibody, as will be reviewed below. Adsorbed proteins tend to occur in clusters of active proteins instead of forming an evenly distributed layer. The cluster formation has been demonstrated with monoclonal antibodies on polystyrene using various imaging methods such as scanning electron microscopy (SEM), scanning tunneling microscopy (STM) and atomic force micros- copy (AFM) (Butler et al., 1992; Davies et al., 1994a; 1994b). Furthermore, clustering has been observed with the adsorption of ferritin on various substrates (polycarbonate, carbon, quartz) as visualized by electron microscopic studies (Feder and Giaever, 1980; Nygren, 1988). Biotinylated antibodies immobilized to the streptavidin-coated surfaces have appeared in uniform distribution instead of clustering when visualized by STM and AFM (Davies et al., 1994a; 1994b). Studies have suggested that the interactions between the binding molecules are stronger on the solid-liquid interface than in the solution (Lehtonen, 1981; Nygren et al., 1985). These observations are usually explained by the formation of the binding clusters where the increase in local densities results in secondary bonds which increase overall binding avidity. This model originates from the well-known lateral diffusion of the receptor molecules in the lipid layer of the cell membranes when binding their ligands. Likewise, lateral diffusion of the adsorbed proteins has been shown (Michaeli et al., 1980). The clustering of the antibodies on the surface may be, nevertheless, an advantage in immunoassays since possible cross-linking accompanied by slower disso- ciation rates should guarantee better preservation of the bound analyte on the surface during the repeated washing and incubation steps. Adsorbed antibodies (and proteins in general) typically exhibit decreased activity and antigenicity. According to Butler (2000), conformational changes of the adsorbed pro- teins have been recognized at least since the 1950s. More comprehensive and quantita- tive studies of the adsorption-induced changes to the activity of immunoassay antibod- 24 Review of the literature 25 ies were commenced by Butler and co-workers in the 1980s. A common finding has been the better survival of the polyclonal antibodies compared with the corresponding monoclonal antibodies in adsorption (Butler et al., 1986; Suter et al., 1989; Butler et al., 1992; Joshi et al., 1992; Butler et al., 1993). One explanation for the better survival of the polyclonal antibodies is that among the range of the antibodies of multiple speci- ficities there are always some variants which tolerate adsorption better than others. Another suggestion is that the multiple specificities of the antibody preparations result in a more efficient cross-linking of the bound antigen followed by a pronounced avid- ity effect. Some quantitative values and estimates of the magnitude of the lost or re- maining activities have also been reported. For example, Joshi et al. (1992) reported that 5 - 7 % of polyclonal antibodies while less than 1 % of the monoclonal antibodies retained their activity when adsorbed on polystyrene. Fluorescein-specific polyclonal antibodies were reported to retain the activities of 5 - 10 % and monoclonal antibodies less than 3 % (Butler et al., 1992). In another similar study the respective values were 22 % and 2 - 6 % (Butler et al., 1993). In addition, Davies et al. (1994a; 1994b) re- ported that 5 % of the adsorbed monoclonal antibodies were functional for binding ferritin, which was used as an antigen when studying the surfaces by means of atomic force microscopy and scanning tunneling microscopy. The preservation of binding activity is perhaps one of the strongest arguments in favor of streptavidin surfaces. For example, Suter and co-workers found 5 - 400-fold (Suter and Butler, 1986) and 5 - 6-fold (Suter et al., 1989) higher antigen-binding activities with antibodies immobilized to the streptavidin-coated surface in comparison with the antibodies adsorbed directly to polystyrene. Davies and co-workers (Davies et al., 1994a; 1994b) reported that 60 % of the antibodies remained functional on the strepta- vidin-coated surface compared with 5 % for the adsorbed antibodies. A panel of fluo- rescein-specific monoclonal antibodies showed a functionality of less than 3 % when adsorbed, whereas immobilization by a streptavidin-coated surface preserved the func- tionality at around the 20 % level (Butler et al., 1992). Another study from the same group reported an approximately 5-fold improvement in the activity of the capture antibodies when immobilized via streptavidin instead of direct adsorption (Butler et al., 1993). Monoclonal antibodies have benefited more from streptavidin-based immobili- zation than polyclonal antibodies (Suter et al., 1989; Butler et al., 1993). This is obvi- ous considering that typically a lower fraction of the monoclonal antibodies survive passive adsorption and, thereby, the non-destructive immobilization by streptavidin certainly has a greater effect. In addition to the antibodies, immobilization by strepta- vidin has also been shown to be beneficial for other proteins, as exemplified by a bridging ELISA-assay of the antibodies against recombinant erythropoietin (Gross et al., 2006). The immobilization of the antigen (erythropoietin) to the streptavidin-coated surface resulted in a 100 - 300-fold more sensitive assay than an identical assay using adsorbed erythropoietin. Again, the lower activity of the adsorbed antigen was ex- plained by conformational changes and the loss of epitopes induced by adsorption. 25 Review of the literature 26 Spitznagel and co-workers (Spitznagel and Clark, 1993; Spitznagel et al., 1993) re- ported decreased specific activities of antibodies (meaning active binding sites per antibody) on surfaces of high antibody densities; an especially dramatic decrease was observed above certain surface densities. Furthermore, Xu et al. (2006; 2007) studied the adsorption and the binding capacities of a monoclonal antibody against hCG on silicon oxide substrate by means of neutron reflection and ellipsometry. Specific activi- ties of 0.1 - 0.3 were found with the antibodies that were adsorbed at pH 4 - 7, whereas the ratio was 0.8 at pH 8.0. In the higher pH, the amount of the adsorbed antibody was also found to be lower. The reduced specific activities of the antibodies on the dense surfaces can largely be explained by steric hindrances and conformational issues rather than by the denaturation of the antibodies. After increased packing, the accessibility of the binding sites evidently becomes hindered at least for large antigens. In the extreme, the high antibody density on the surface may result in the reduction of the total binding capacity of the surface, as shown for example by Bae et al. (2005). 1.3.2.2 Antigenic specificity and heterogeneity Adsorbed proteins may exhibit altered antigenic specificities. This is potentially a more severe problem than sole activity loss in immunoassays, since the lost activity can be partially compensated by relatively large quantities of the adsorbed antibodies. The new antigenic determinants introduced by adsorption are usually the regions of the proteins which are buried in the native conformation but become exposed upon adsorp- tion-induced changes. Evidence of the altered antigenicity of the adsorbed proteins was found already in the 1960s (Kochwa et al., 1967) when altered specificities were re- ported in rabbit antisera which were raised by immunizing the animals with particles bearing adsorbed antigens. In a later study (Djavadi-Ohaniance et al., 1984), mono- clonal antibodies of eight epitope specificities against Escherichia coli tryptophan syn- thase were monitored with respect to their antigenic specificities. Five of the eight an- tibodies reacted very slowly with a native antigen in solution, while they reacted rap- idly with antigens adsorbed to a microtiter plate. In a subsequent study (Friguet et al., 1984), the same antibodies were tested in solution both against the native antigen and against the antigen that was first subjected to denaturation by N-ethylmaleimide. Those five antibodies that reacted rapidly with the adsorbed antigen in the previous study recognized the denatured antigen in the solution reasonably well. This suggested ad- sorption-induced changes which resemble those that are introduced by chemical dena- turation of protein structure. Butler et al. (1997b) showed that intradomain allotypes (supposedly) of immunoglobulin G2a (IgG2a) were more readily recognized by respec- tive monoclonal antibodies when the antibody was passively adsorbed instead of being immobilized on the streptavidin-coated surface (although streptavidin-based immobili- zation almost invariably constituted an advantage in terms of activity). This was ex- plained by the exposure of the buried domains of the antibody upon adsorption. In addition, the study showed that the antibodies which were adsorbed after denaturation by 6 M guanidine hydrochloride did not show remarkable differences in binding activi- ties compared with the antibodies that were adsorbed without prior denaturation. On 26 Review of the literature 27 the streptavidin surface, however, the difference in activity between the chemically denatured antibody and the one not denatured was remarkable. Adsorbed antibodies present heterogeneity in their antigen binding affinities (Rabbany et al., 1997; Vijayendran and Leckband, 2001). The unique feature of the monoclonal antibodies, the single homogeneous antigen binding affinity value in solution, is lost upon adsorption and the antibody surface exhibits multiple heterogeneous affinities. This probably originates both from the different survival of the individual antibody molecules upon adsorption and from the varying orientation and the accessibility of the binding sites. Streptavidin surfaces may alleviate the problem of non-optimal orienta- tion to some extent and recover the homogeneity of the affinities as shown, for exam- ple, by Vijayendran and Leckband (2001). The study found that the antibodies that were biotinylated at their carbohydrate moieties and immobilized to an oriented strep- tavidin surface showed the highest and most homogeneous affinities in comparison with alternative methods. The orientation of the streptavidin surface was enabled by a biotinylated interface consisting of biotin that was covalently bound to the solid sub- strate (more about this type of surface will be reviewed in section 1.4.4). The alterna- tive surfaces in the study consisted of either covalently bound antibody, covalently bound streptavidin or protein G surfaces. The covalent coupling of the antibody and streptavidin was mediated through amines. This is a somewhat random process and, therefore, the surfaces were unable to render the homogeneity of the bound proteins. 1.3.2.3 Other non-adsorptive immobilization techniques Typically, the antibodies that are immobilized to a superficial second layer preserve their activity significantly better compared with antibodies that are in direct contact with solid substrate (Butler et al., 1997b). In addition to streptavidin surfaces, an alter- native method to separate the antibodies from the inorganic substrate is provided by antibody-binding antibodies which are employed as the underlying first layer. This approach retains the activity of the immobilized antibodies fairly well, even better than immobilization via streptavidin. At best, complete preservation of the activity is re- tained (Butler et al., 1993). However, as the underlying antibody layer is vulnerable to adsorption-induced losses in binding activity, the total amount of the immobilized anti- body may become quite low. This may eventually decrease the actual number of func- tional binding sites to a lower level than obtained by direct adsorption (Butler et al., 1992). Other means of immobilizing the antibodies to the second layer is to use Fc-binding proteins such as protein A and protein G. Forsgren and Sjöquist (1966) found that pro- tein A from the cell wall of Staphylococcus aureus binds to the Fc part (fragment crys- tallizable, the constant region of the antibodies) of the antibody and this principle was afterwards exploited in the immunoaffinity columns using the protein A-Sepharose matrix (Moseley et al., 1977; Gersten and Marchalonis, 1978). Protein G, a streptococ- cal cell wall protein, constitutes a more versatile binder since it has a wider specificity range to different IgG isotypes in comparison with somewhat limited specificities of 27 Review of the literature 28 protein A (Langone, 1978; Björck and Kronvall, 1984; Reis et al., 1984). A more uni- versal binder, employing the specificities of both of these, was provided by a recombi- nant chimeric fusion of protein A and protein G (Eliasson et al., 1988; 1989). Exam- ples of the use of protein A and protein G in the bioassays include the capturing of antibodies on magnetic beads (Widjojoatmodjo et al., 1993), the capturing of pre- equilibrated antigen-antibody complexes to a microtiter plate (Protein A antibody cap- ture ELISA, PACE) (Ngai et al., 1993), the immobilization of capture antibodies to fiber optic biosensors (Anderson et al., 1997) and the capturing of antigen-antibody complexes to a protein G column in an enzyme flow immunoassay of herbicides (Bjarnason et al., 2001). 1.3.2.4 Orientation Fc-binding proteins enable the immobilization of antibodies in a correctly oriented manner. In addition to protein A and protein G, Fc-specific antibodies can be used to facilitate correct orientation, as shown by immobilizing enzymes on the surface via an oriented antibody bridge (Gunaratna and Wilson, 1990; 1992). The streptavidin sur- faces can facilitate correct orientation which is accomplished by combining the strep- tavidin-based immobilization with site-specifically biotinylated antibodies. Another feasible means of orientation relies on the carbohydrate moiety present in the Fc do- main of the antibodies. The reactive aldehyde groups which emerge through oxidation with mild sodium periodate (O'Shannessy and Quarles, 1985) can be biotinylated with the hydrazide derivatives of biotin (O'Shannessy et al., 1984) or they can be covalently coupled to hydrazide-activated surfaces (Hoffman and O'Shannessy, 1988). Further- more, carbohydrate moieties can be used for oriented immobilization by lectins. Lectins comprise a diverse group of carbohydrate-binding proteins (Lis and Sharon, 1986) which have been successfully used to immobilize antibodies, for example, to sensor surfaces (Starodub et al., 2005). Furthermore, one popular method of orientation relies on the disulphide bridges present in the hinge region of the antibody. The thiols generated upon mild reduction can be used either for covalent coupling or for targets of biotinylation by thiol-reactive biotin derivatives (Domen et al., 1990; Cho et al., 2007). This thiol-based modification and immobilization has proven applicable to intact anti- bodies, but above all it is useful for the immobilization of the Fab, Fab' and F(ab')2 fragments as was demonstrated by Prisyazhnoy et al. (1988) and subsequently charac- terized in other studies (Lu et al., 1995; Bonroy et al., 2006). The oriented, site-specific immobilization of the antibodies has typically resulted in favorable effects with respect to surface homogeneity, binding affinities and capacities (Domen et al., 1990; Shmanai et al., 2001; Vijayendran and Leckband, 2001; Peluso et al., 2003; Bonroy et al., 2006; Kang et al., 2007). Comprehensive reviews on the methods and effects of the oriented protein immobilization have been written, for example, by Lu et al. (1996), Rao et al. (1998) and Turková (1999). 28 Review of the literature 29 1.3.3 Aspects of protein adsorption 1.3.3.1 Mechanisms and conditions Adsorption is in many situations the prevailing mechanism in the adhesion of proteins to solid substrates, even though covalent chemistries or some other advanced methods are employed. The presence of passive adsorption must always be taken into account when a protein and a solid substrate interact, unless adsorption is prevented by using detergents or very hydrophilic surfaces which confer little adsorption. The multitude of protein features with respect to the size, shape, isoelectric point, hydrophobicity and rigidity combined with a range of the solid substrates of various surface hydrophobic- ity, charge and format constitute a range of conditions whereby it has proved difficult to find useful, well-accepted models of physisorption (Dent and Aslam, 1998). Some rules, however, prevail which will be briefly discussed. In order to occur spontaneously, adsorption must be an exothermic reaction, meaning that Gibbs energy (ΔG) must be less than zero. Gibbs energy is defined by the follow- ing equation STHG Δ−Δ=Δ where H refers to the enthalpy of the system, T is the absolute temperature and S is the entropy. While the change in enthalpy during the adsorption process is small, the posi- tive gain in entropy dominates, resulting in overall negative Gibbs energy. Therefore, adsorption is said to be an entropically driven process. The positive gain in entropy is supposed to originate from the dehydration of the substrate and the protein surfaces. Typically, the constituents are first surrounded by well-organized water molecules (solvated). The hydrophobic regions of the substrate and the protein tend to interact minimally with water. This is achieved upon contacting the hydrophobic regions of the protein and the substrate while dislodging water from these sites. The released water molecules then increase the entropy of the system. This explains why the proteins ad- sorb preferentially to hydrophobic surfaces contributed by hydrophobic interactions. The driving force of dehydration is not so strong on hydrophilic surfaces and passive adsorption is weaker. The theory of entropically driven adsorption was postulated by Norde and Lyklema (1979) and, thereafter, it has been reviewed in several studies and reports (Lyklema, 1984; Andrade and Hlady, 1986; Norde, 1986; Dent and Aslam, 1998). Adsorption is preferred to hydrophobic surfaces, but extreme hydrophobicity is not favorable either (Lee and Ruckenstein, 1988). Rather, surfaces with both non-polar and polar properties have been optimal for passive adsorption. This is because of the other interactions such as hydrogen bonds, electrostatic effects and van der Waals forces (transient polarities) contributing to adsorption (Esser, 1997b). The electrostatic inter- actions are, however, considered less dominant, though not negligible, on the surfaces where the hydrophobic interactions predominantly take place. For example, IgG ad- sorbs to negatively charged surfaces in the pH values above its isoelectric point (pI) 29 Review of the literature 30 where its net charge is negative and thus repulsive forces are expected (Bagchi and Birnbaum, 1981; Buijs et al., 1995). Several studies have confirmed that the optimal pH for adsorption is usually close to the pI of the protein (see Dent and Aslam, 1998). At the isoelectric point, where the net charge of the protein equals zero, the protein adopts the most compact structure and exhibits the least repulsion between the neighboring protein molecules. Outside the isoelectric point the repulsion between the adjacent molecules and the more extended conformation obstructs denser packing of the molecules into a layer. The adsorption peak at the pI has also been observed on hydrophilic surfaces, where the electrostatic interactions should be more favorable. For example, the decrease in the pH of the solution and consequent increase in the positive charge of the protein should facilitate more binding to a negatively charged hydrophilic surface due to the pronounced opposite charges. This is not always true, probably be- cause of the increased repelling forces emerging between the adjacent molecules (Buijs et al., 1995; Xu et al., 2006). Increased ionic strength of the coating solution decreases the electrostatic interactions and the repulsion between the adjacent molecules, and it indeed facilitates adsorption (Lee and Ruckenstein, 1988). The reduced electrostatic effects by the increased ionic strength should also apply to the interactions between the protein and the substrate, so that under repulsive electrostatic conditions adsorption is enhanced while under attractive conditions binding is decreased. Therefore, the effect of the ionic strength is quite strongly linked to the pH of the solution. Studies have shown, however, that the relations between the particular protein and its pI, solution pH and the ionic strength of the solution are not that straightforward (Buijs et al., 1995; Butler et al., 1997a). Adsorption is generally considered an irreversible, non-equilibrium reaction (for re- views see Haynes and Norde (1994) and Nakanishi et al. (2001)). Dilution of the solu- tion usually results in a hysteresis in the adsorption isotherms, meaning unequal as- cending and descending branches of the isotherm. This is indicative of an irreversible process (more about isotherms will be reviewed below). On the other hand, some de- sorption and replacement of the adsorbed protein is well documented, especially under rinsing or in the presence of replacing proteins (Vroman and Adams, 1969; 1986; Wahlgren et al., 1993; Butler et al., 1997a; Xu et al., 2006). Desorption is obvious from the surfaces which have been coated using saturating protein quantities (Lee and Ruckenstein, 1988), since beyond the monolayer-forming concentrations the emer- gence of the loosely bound secondary layers is evident. When the desorption of the adsorbed protein is monitored, it is essential to recognize whether the question is about the breakage of the protein-solid substrate or the weaker protein-protein interactions before making further conclusions on the stability of the adsorbed layer and the inter- actions of the particular protein and the solid phase. 1.3.3.2 Adsorption isotherms Adsorption isotherms can be used to link the amount of the adsorbed protein to the amount of the non-adsorbed protein at the equilibrium state (Fig. 4). The simplest physically plausible isotherm is the Langmuir isotherm which was originally developed 30 Review of the literature 31 to model the adsorption of gas molecules to solid phases (Langmuir, 1916) and it has been further applied to the adsorption of proteins. It defines the surface coverage, or the fraction of the occupied binding sites as follows: [ ] [ ]Pb Pb += 1θ In the equation, θ refers to the surface coverage, [P] to the protein concentration in the solution and b is a constant, analogous to the equilibrium constant of a ligand-receptor binding equilibrium, here indicating the affinity (or preferably avidity) of the protein to the surface. Due to the several prerequisites of the Langmuir isotherm such as reversi- ble binding, monolayer requirement (no secondary upper layers), homogeneous bind- ing sites and solutes as well as the absence of lateral interaction between the molecules, the isotherm is frequently criticized for its unsuitability to protein adsorption. Indeed, scientists have found their adsorption data to fit variably in the Langmuir isotherm (Wilkins Stevens and Kelso, 1995; Li et al., 2005; Zhang et al., 2005). It nevertheless provides a simple model for an empirical description of adsorption behavior. 0.20 0.40 0.60 0.80 1.00 High b Low b C ov er ag e Protein concentration Figure 4. Langmuir adsorption isotherms with low (broken line) and high (solid line) values of the constant b. The coverage on the y-axis refers to θ in the Langmuir equation. Another well-established adsorption isotherm is an empirical Freundlich isotherm which defines surface coverage as follows: [ ] nPa /1=θ In the equation a and n are constants and [P] refers to protein concentration. The Freundlich isotherm remains convex through the whole range, in contrast to the Lang- muir isotherm which tends to linearity near the origin and approaches saturation in the high concentrations. The Freundlich isotherm is supposed to consider the issues of 31 Review of the literature 32 lateral interactions between molecules and surface heterogeneity better than the Lang- muir isotherm. (Dent and Aslam, 1998; Yang, 1998.) Another useful means to describe adsorption behavior is the percentage bound plot (Fig. 5) which is defined as the ratio of the adsorbed protein to the added amount of the protein. In several adsorption situations these plots have shown a constant value in the dilute region (linear binding region) followed by a clear decrease in the percentage bound value (Cantarero et al., 1980; Butler et al., 1992; Joshi et al., 1992; Butler et al., 1993; 1997a). The percentage value of the linear binding region (or the region of inde- pendence) describes the protein's "avidity" to the surface and the end of the linear re- gion corresponds to the formation of a protein monolayer. Cantarero et al. (1980) showed that the percentage value of the linear region corresponded mainly to the size of the protein. The values ranged from less than 20 % for α-lactalbumin (14 kD) to more than 80 % for immunoglobulin M (IgM, molecular weight close to 1000 kD) which showed that the larger proteins had higher overall adsorption capability. The fact that the larger size enhances adsorption is explained by a larger number of individual binding sites. Proteins are supposed to rearrange on the surface after initial adsorption contacts and under this process small units may lose all their footholds while larger polymers more probably have some sites attached all the time (Dent and Aslam, 1998). 20 40 60 80 100 Small protein, low avidity Large protein, high avidity Ad so rb ed % Added protein amount Figure 5. A schematic presentation of the percentage bound plots, showing the be- havior of the proteins with "high avidity" (solid line) and "low avidity" (broken line) to the surface. Several studies have reported the total quantities of proteins adsorbed to various sur- faces and some of these values have been collected by Dent and Aslam (1998) and Nakanishi et al. (2001). The actual maximum quantities of the adsorbed proteins range from a few dozen to several hundred nanograms per square centimeter depending on the type of protein and surface. 32 Review of the literature 33 A substantial number of adsorption studies have been carried out using immunoglobu- lins as the model proteins, which is understandable considering their wide use in the solid-phase assays (Underwood and Steele, 1991; Buijs et al., 1995; Wilkins Stevens et al., 1995; Wilkins Stevens and Kelso, 1995; Butler et al., 1997a; Vermeer et al., 2001). In addition, the use and applications of streptavidin in a variety of binding surfaces have been widely reported. However, the fundamental adsorption characteristics of intact streptavidin on different substrates in various physical and chemical conditions have not been comprehensively reported. The structural and quantitative features of adsorbed streptavidin layers have been stud- ied using, for example, scanning probe microscopy (SPM) techniques and quartz crys- tal microbalance (QCM, brief descriptions of the various techniques are reported in section 5). Cooper et al. (1994) imaged individual streptavidin molecules adsorbed on a graphite surface using scanning tunneling microscopy (STM). The images revealed some denaturation of the protein and the organization of four adsorbed subunits on the surface in a somewhat separated manner. Atomic force microscopy (AFM) images have been acquired from streptavidin layers covalently bound to mica and gold sub- strates (Kim J. et al., 2004) and to biotin-functionalized Langmuir-Schaefer layers chemisorbed on gold (Ihalainen and Peltonen, 2004). In the latter study, the tip-sample convolution typically recognized with the AFM (smoothing of the lateral resolution due to a non-optimal curvature of the tip head) was corrected by means of software which enabled the acquisition of sharp, high-resolution images of the streptavidin lay- ers. The study revealed pronounced clustering and high local densities of adsorbed streptavidin after increasing protein bulk concentration and the amount of adsorbed protein. A surface coverage of 49 % was measured at the highest level. In the same study, comparative measurements using QCM revealed a maximum quantity of ad- sorbed streptavidin of 250 ng/cm2 on a dried surface. This represented a surface cover- age of 63 %. A typical property, the clustering of the adsorbed protein was thus recog- nized with streptavidin, although elsewhere streptavidin surfaces have enabled a smooth surface distribution of biotinylated capture antibodies compared with passively adsorbed antibodies (see section 1.3.2.1). 1.4 Advanced coating technologies 1.4.1 Description of techniques Adhesion mechanisms frequently make use of several simultaneous physical and chemical phenomena. Therefore, it may be inappropriate to strictly categorize which particular mechanism is employed in a given application. However, some suggestive classification of the techniques is useful in order to elucidate the principles of various mechanisms. One clear definition is whether the particular protein is coated either di- rectly to the solid substrate or through an interface layer consisting of a kind of "car- rier" molecule. Another determinant is whether adhesion is based on either plain phy- sisorption or whether other interactions such as covalent chemistry are used to facilitate binding. Furthermore, the target protein and the carrier molecule can be introduced either in monomeric or polymerized forms. An additional, unique feature involved in 33 Review of the literature 34 the immobilization of streptavidin evolves from its multiple biotin-binding properties. A biotinylated interface layer can be used to immobilize streptavidin while still retain- ing multiple unoccupied binding sites to serve the immobilization of the desired bioti- nylated molecules. 1.4.2 The features of streptavidin relevant to adsorption As shown in several studies (I, II, Davies et al., 1994b; Huang et al., 1994; Allen et al., 1996) streptavidin adsorbs spontaneously to the polystyrene surfaces in optimized con- ditions. The adsorption property of streptavidin is frequently explained by its hydro- phobic nature. Some chromatographic data supports the hydrophobicity of streptavidin (Reh et al., 1986; Schwidop et al., 1990; Tischer et al., 1991) but on the other hand, Server (http://au.expasy.org, based on the algorithm by Kyte and Doolittle (1982), using entry P22629) returns the GRAVY-value of –0.331 for core streptavidin, the negative value thus classifying the protein as hydrophilic rather than hydrophobic. In gions of greater hydrophobicity in the subunit and these hydrophobic patches are likely to be responsible for the hydrophobic character recognized for streptavidin. Though directly adsorbed streptavidin provides functional biotin-binding surfaces, there is some ongoing discussion on the adequacy of the surface density, the binding capacity and the stability of the adsorbed protein layer. Surface-bound streptavidin has also shown decreased binding affinities for biotinylated macromolecules. Fujita and Silver (1993) reported an affinity constant (Ka) in the order of 1012 M−1 for the binding of biotinylated DNA to the streptavidin-coated magnetic beads whereas Huang et al. (1996) reported affinity constants in the range of 0.6 × 108 - 1.1 × 1010 M–1 for the binding of biotinylated DNA to the streptavidin-coated polystyrene particles. The large number of the publications and reports generated in the context of the protein coatings may reflect the fact that in the absence of well-defined adsorption models scientists have been obliged to find the appropriate conditions for their application experimentally. Another explanation for the active research and development in the field relates to the prominent commercial share of the applications using immobilized proteins, of which streptavidin-coated microtiter wells and sensor surfaces are repre- sentative examples. Due to commercial interests, manufacturers do not by definition disclose the coating techniques in the scientific literature, but informative technical details can be found in the patent files instead. Some advanced protein coating and immobilization technologies will be reviewed below using interesting examples of streptavidin and avidin coatings reported in the literature. 34 cipitate. The hydrophobicity plot (computed at the ExPASy) shows two distinct re- average of hydropathicity (GRAVY) -value computed at the ExPASy Proteomics 30 - 50 mg/ml (our experimental results) and it remains soluble without tendency to pre- hydrophilic features can also be recognized for streptavidin. For example, the grand addition, core streptavidin is readily dissolved in water in concentrations as high as Review of the literature 35 1.4.3 Covalent coupling 1.4.3.1 Regular covalent chemistries Covalent coupling is a useful alternative to improve the immobilization efficiency of proteins since a variety of chemistries enabling straightforward coupling are available. Perhaps the most common coupling approach is based on the reactivity of the ε-amines of the lysine residues and the N-terminal amines of the polypeptide chains. Widely used amine-reactive chemical groups include N-hydroxysuccinimide (NHS) esters, iso(thio)cyanates and aldehydes. Though easy to perform, amine-mediated coupling may result in quite random modification owing to the abundance and diverse distribu- tion of lysines in proteins. Other modification targets of the proteins include the car- boxyl groups of aspartate and glutamate. These react with amine-containing modifica- tion reagents in carbodiimide-mediated reactions. More controlled and site-specific modification relies on the carbohydrate residues of proteins which are reactive with amines and hydrazides after oxidation by mild sodium periodate and consequent for- mation of the aldehyde groups. Furthermore, one of the most important protein modifi- cation methods utilizes reactive sulfhydryl- (or thiol) groups of cysteines which repre- sent efficient modification targets at exact locations. Thiol-groups can be generated by a mild reduction of the cystine disulfide bridges to two cysteines. Common sulfhydryl- reactive chemistries include maleimides, haloacetyls and alkyl halides which form stable thioether bonds with the sulfhydryls. Furthermore, the sulfhydryls can be used to create disulfide bridges in the oxidation reactions and thiol-disulfide exchange reac- tions. Examples of the latter include the reactions of the thiols with pyridyl disulfides and with the reagents containing the 5-thio-2-nitrobenzoic acid (TNB) structure. The disulfide bonds can be cleaved quite easily by mild reduction enabling specific break- age of the linker which is a useful property in certain applications. (Hermanson, 1996a; 1996c.) When inherent cysteines are absent or scarce, additional introduction of sulf- hydryls can be used to render reactivity. Appropriate thiolating agents include, for ex- ample, 2-iminothiolane (Traut's reagent, Traut et al., 1973) and heterobifunctional agents such as N-succinimidyl-S-acetylthioacetate (SATA, Duncan et al., 1983). Sup- plemental thiolation renders, however, a more random distribution since thiols are usu- ally introduced through amines. One of two main strategies can be used when coupling proteins to other proteins and to the solid substrates. First, homobifunctional cross-linkers such as dialdehydes and bi- functional NHS-esters enable simultaneous coupling of proteins in a single reaction. Homobifunctional coupling is easy to perform and it is best suited for a homogeneous "self-conjugation" of proteins. The polymerization degree is hard to control and the result is a mixture of polymers containing the initial proteins in various ratios. In con- trast, heterobifunctional cross-linkers allow more controlled reactions in two- or three- step procedures. A representative example of their use is a conjugation process where maleimide functionality is applied in one partner and sulfhydryl functionality (inher- ently present or chemically added) in the other partner to serve the covalent conjuga- tion of the molecules. Some special heterobifunctional reagents (for example SATA) 35 Review of the literature 36 provide chemically protected groups which are non-reactive until exposed by certain de-protective agents. (Hermanson, 1996b.) A feasible strategy for the covalent coupling of the proteins to the solid phases is to activate the substrate first with reactive chemical functionalities and then introduce the protein of interest containing a specific reactivity to the activated support (Niveleau et al., 1993; Shriver-Lake et al., 1997; Suzuki et al., 1997; Page et al., 1998; Wang and Jin, 2004; Pyun et al., 2005). When compared with the passive coating methods in parallel, the covalent approaches have regularly shown higher amounts of bound pro- tein. Covalent modification and immobilization techniques have also been applied for streptavidin, though due to the absence of cysteines and carbohydrates only amine- or carboxyl-reactive chemistries work. The covalent binding of streptavidin to solid phases was reported already in the early studies quite soon after the first reports of using streptavidin-based solid phases for the assays. Peterman et al. (1988) bound streptavidin to the bromoacetyl-activated polystyrene spheres through nucleophilic groups such as amines and found the covalent binding to increase binding capacity and decrease dissociation compared with passively adsorbed protein. Dudin et al. (1988) coupled streptavidin to magnetic beads using sulfonyl chloride chemistry (amine- reactive) to be used in chromosome separation. Amine surfaces and isothiocyanate- based coupling chemistries were also used in a recent magnetic particle application (Zhang et al., 2007). Carbodiimide chemistry has been used to coat streptavidin to carboxyl-terminated Eu-nanoparticles to be used as reporters in highly sensitive immu- noassays (Härmä et al., 2001). Further, Schiestel et al. (2004) used carbodiimide chem- istry for the covalent coupling of streptavidin to amine- and carboxyl-terminated silica nanoparticles. The coupling of streptavidin to the amine surface through the protein's carboxyls proved ineffective. This was explained by the higher reactivity of lysines of streptavidin compared with the amines on the surface. In contrast, the binding of strep- tavidin to carboxyl nanoparticles through the protein amines was successful resulting in (according to the authors) the highest binding densities reported on the nanoparticle surfaces by that time (1.1 streptavidin molecules per 100 nm2). Furthermore, Douglas and Monteith (1994) reported a 96-well plate that bound proteins covalently through amines when introduced in a buffer with the pH above the pI of the protein. These plates were used for the covalent binding of streptavidin and the resulting surfaces showed significantly higher uniformity and increased biotin-binding capacities com- pared with the passively coated reference plates. 1.4.3.2 Photochemistry A particular technique in covalent binding is photochemistry, which makes use of the chemical reactivities that are induced by the exposure of special chemical structures to light (usually ultraviolet light). The most popular photosensitive groups are aryl azides which form reactive nitrene intermediates upon photolysis. These intermediates go through either an addition reaction to the double bonds or an insertion reaction into the C–H and N–H bonds. They can also react with nucleophiles such as primary amines. Other photosensitive functionalities include benzophenones, diazo compounds and 36 Review of the literature 37 diazirine derivatives. Upon photoactivation, the principal binding targets of all these are the C–H and N–H bonds, hence the site-specificity of the coupling is rather weak. However, photochemistry enables the binding of ligands to the targets where the regu- lar chemical functionalities are lacking or scarce. Various heterobifunctional cross- linkers have been synthesized, which contain both a photoreactive end and a regular thermochemical end. This may be amine-reactive, sulfhydryl-reactive or carboxyl- reactive. The photoreactive groups are typically inert until exposed by a light pulse and, therefore, these reagents enable controlled and timed initiation of the cross- linking. (Hermanson, 1996d; 1996e.) In addition to the regular photosensitive groups mentioned above, common fluorescent reporter dyes such as fluorescein and Alexa 594 are capable of forming bonds when extensively photobleached by prolonged exposure to light (Holden and Cremer, 2003). These are activated by longer wavelengths (470 nm and 560 nm) and thereby some of the potential drawbacks of ultraviolet light are avoided. One approach to immobilize proteins is to couple a photoreactive substance to an inert solid support by photoactivation while maintaining the other, thermochemical func- tionality of the reagent available for the covalent coupling of the protein. For example, an aryl azide derivative, 1-fluoro-2-nitro-4-azidobenzene (FNAB) has been used to activate polystyrene and polycarbonate surfaces (Nahar et al., 2001; Bora et al., 2006) and these surfaces have been shown to rapidly bind high quantities of proteins. Fur- thermore, they have introduced higher sensitivities and heat-stability in ELISA-assays in comparison with non-treated surfaces (Bora et al., 2002; 2004). In another study, the thermoreactive end of FNAB was first reacted with a carrier polymer while photoreac- tive functionality was used for protein coupling (Naqvi and Nahar, 2004). Alterna- tively, quinones can be used to activate surfaces for photochemical coupling (Jacobsen and Koch, 2000). An alternative to the activation of the solid phase is to activate the protein of interest with a photoreactive ligand. Nakajima and co-workers (Nakajima et al., 2006) bound a photoreactive ligand to glucose oxidase and horseradish peroxidase, after which the activated enzymes were patterned inside microfluidic channels by means of a laser beam. Regioselective patterning is conveniently enabled by photo- chemistry since the target proteins can be bound from a bulk solution to exact locations by means of a laser beam or exposure of light through special photomasks. Streptavidin and avidin are typically linked to photochemistry through photoactivatable derivatives of biotin (Fig. 6). A common strategy has been to couple an aryl azide de- rivative of biotin (photobiotin, Forster et al., 1985) to the solid substrate by a photo- chemical reaction and consequently use the biotinylated surfaces for the immobiliza- tion of either streptavidin or avidin (Hengsakul and Cass, 1996; Dontha et al., 1997; Sabanayagam et al., 2000; Wilde et al., 2001; Choi et al., 2005). Some of these studies were confined to visualizing the patterning of biotin by streptavidin-conjugated en- zymes and fluorophores, whereas some studies further proved the availability of the unoccupied biotin-binding sites by immobilizing biotinylated antibodies and oligonu- 37 Review of the literature 38 cleotides to the surface. The concept is also applicable in reverse order. Pritchard et al. (1995) immobilized photobiotin to an avidin-coated surface through the biotin-end and the remaining photochemical functionality was used for the coupling of an antibody. Another photosensitive derivative of biotin, biotin-4-fluorescein, has also been used to immobilize streptavidin-conjugated enzymes inside microchannels (Holden et al., 2004). Photochemical biotinylation, as photochemical reactions in general, can be fo- cused to confined locations enabling subsequent introduction of streptavidin-enzyme conjugates and biotin-binding surfaces to defined regions. N H NHNH S HH O O N CH3 N H N3 NO2 O N O O O N H NHN S HH O O O ONO2 O O A B Figure 6. A) Photobiotin, N-(4-azido-2-nitrophenyl)-N'-(N-d-biotinyl-3-aminopropyl)-N'- methyl-1,3-propanediamine, contains a photoactivatable terminal group to enable a photo- chemical coupling of biotin to the solid substrates. Drawn according to Forster et al. (1985). B) Methyl α-nitropiperonyl-oxycarbonyl biotin - aminocaproic-NHS ester (MeNPOC-biotin- AC–NHS), or "caged" biotin, contains a photolabile protective group which obstructs the bind- ing of biotin to avidin and streptavidin. Drawn according to Pirrung and Huang (1996). In (strept)avidin-biotin technology, the use of photochemistry is not only limited to the photochemically mediated coupling of biotin through the reactive group of the ali- phatic side chain. In addition, the double ring structure of biotin can be protected with a photolabile group which can be removed by light exposure and thus recover the bind- ing activity of biotin to avidin and streptavidin (Sundberg et al., 1995). The chemistry of this "caged" biotin relies on a photoremovable nitrobenzyl group which is bound to one nitrogen in the double ring of biotin and this biotin derivative is named as methyl α-nitropiperonyl-oxycarbonyl biotin (MeNPOC-biotin). These biotin derivatives have been coupled to solid substrates either by direct covalent coupling (Sundberg et al., 1995), using BSA as an interface protein (Pirrung and Huang, 1996; Blawas et al., 1998) and by self-assembled alkanethiol monolayers on gold (Yang et al., 2000). The immobilization of streptavidin by this approach is probably not very efficient; for ex- 38 Review of the literature 39 ample, Blawas et al. (1998) yielded densities of 9 × 109 streptavidin molecules per square millimeter (≈ 0.015 pmol) and an efficiency of 20 % for the immobilization of streptavidin to BSA-biotin surfaces. In addition to the photochemically activatable biotin derivatives, an electrochemically activatable variant, a hydroquinone-caged bio- tin has been reported (Kim K. et al., 2004). This biotin derivative was likewise used for the patterning of streptavidin to the solid phase using an electrode array instead of a photomask. The examples discussed above concerned the use of photochemistry for creating biotinylated surfaces, but an opposite approach was presented by Koyano et al. (1996). Surface-bound biotins were ruptured by ultraviolet light exposed through a mask while the unexposed sites retained their ability to bind streptavidin. 1.4.4 Biotinylated interface Biotin-functionalized surfaces can also be introduced either by regular covalent chem- istries or by self-assembled monolayers on gold surfaces in addition to photochemistry (more about the self-assembly in section 1.4.6). Covalent methods include, for exam- ple, the coupling of NHS-biotin to amine-terminated microtiter well surfaces (Bugari et al., 1990; Rasmussen, 1990; Bicknell et al., 1996), to aminosilanized TiO2 surfaces (Ye et al., 2007) and to hydrazide functionalized surfaces (Wang et al., 2005). Other functionalities are also feasible, such as the binding of maleimide-biotin to a thiolated surface (Vijayendran and Leckband, 2001) and the binding of unmodified biotin to a titania layer through the inherent carboxyl group of biotin (Huang et al., 2006). The strong binding of sulfur to gold is exploited in the self-assembly of the thiol-terminated biotin derivatives into ordered monolayers (Pérez-Luna et al., 1999; Pradier et al., 2002a; Xia et al., 2004; Jung et al., 2006). Another way to prepare the biotinylated monolayer is to use a thiolated, self-assembling interface molecule containing a reac- tive terminal group to serve subsequent covalent binding of biotin (Yam et al., 2001; Pradier et al., 2002b; Liu and Amiridis, 2005). A biotinylated polymer in the interface between the solid substrate and the streptavidin layer frequently constitutes a further advantage compared with the surfaces of directly coupled biotin. The interface layer better prevents direct contact of streptavidin with inorganic solid material and introduces a three-dimensional structure enabling higher quantities of immobilized streptavidin. Furthermore, it can facilitate the orientation of streptavidin to the uppermost layer (Fig. 7) when a more readily adsorbing polymer is used, or when the surface is constructed successively in a layer by layer manner. The principle of a biotinylated interface polymer was already used in early reports of strep- tavidin-coated surfaces. Suter and Butler (1986) used biotinylated keyhole limpet hemocyanin (KLH), gelatine and rabbit gamma globulin as the primary layer for the immobilization of streptavidin. In their study, the antibody-binding capacity of the adsorbed streptavidin surface was only 1 - 10 % compared with the streptavidin sur- faces immobilized via the biotinylated interface. Berger et al. (1993) described the use of heat-treated, denatured and polymerized BSA (thermo-BSA, see the next section) as the biotinylated interface polymer. This, in combination with chemically polymerized streptavidin, exhibited high biotin-binding capacity and good desorption stability 39 Review of the literature 40 against detergents as well as providing greater response in an assay of thyroid stimulat- ing hormone (TSH) compared with control surfaces. Furthermore, the same patent issue described the use of biotinylated aminodextran and biotinylated polyaminoacid (a copolymer of lysine and phenylalanine) as the interface polymers for the immobiliza- tion of streptavidin. Biotinylated BSA has also been used in a co-coating approach to create a solid phase to serve the simultaneous detection of TSH and thyroxin (T4) (Wu et al., 2003). In the first step, a monoclonal antibody against TSH and extensively biotinylated BSA were adsorbed to the surface. The latter provided subsequent binding of streptavidin and the immobilization of a biotinylated monoclonal antibody against T4. Passive adsorption of both antibodies had resulted in the reduction of activities and inadequate assay performance, but in contrast, this co-coating approach proved useful in this assay combining the principles of the sandwich-assay and the competitive assay in one vessel. Other interface proteins are also useful, for example, Strohner and co- workers (Strohner, 2000; Strohner and Immer, 2001) yielded surfaces of high biotin- binding capacity when using biotinylated, cross-linked IgG as the interface layer for the immobilization of a mixture of avidin and streptavidin to the second layer. Figure 7. A schematic presentation of the immobilization of streptavidin using an interface layer. 1.4.5 Various interface polymers and homogeneous conjugation Instead of a biotin link, streptavidin can be coupled to appropriate polymers by means of chemical conjugation. For example, Tischer et al. (1991) coupled streptavidin either to BSA or β-galactosidase and obtained significantly steeper calibration curve slopes in a TSH-assay compared with those obtained using surfaces coated with unmodified native streptavidin. An even more substantial effect was observed when using thermo- BSA as the conjugation partner (Tischer et al., 1991; Berger et al., 1993; Schmitt et al., 1994). In these experiments, BSA was denatured and polymerized by heating to a polymer of the molecular weight above 700 kD. This polymer was activated chemi- cally to provide reactive thiols and streptavidin was modified to offer maleimide moie- ties. Upon combination, these were efficiently conjugated into a heteropolymer and the polymer was coated into the microtiter wells. In addition to excellent assay perform- ance, the thermo-BSA-streptavidin surfaces showed minimal desorption of the ad- sorbed polymer by detergents. 40 Review of the literature 41 In addition to proteins, carbohydrates can be used as carrier polymers in chemical con- jugation and in biotin-mediated immobilization. A patent publication by Dapron et al. (2005) described a method where benzophenone and thiol moieties were first coupled to a dextran polymer. Subsequently, photochemical irradiation was used to couple the polymer to the solid substrate through the benzophenone groups while the thiols en- abled the coupling of maleimide-activated streptavidin to the polymer. Streptavidin quantities of 3.3 - 6.67 µg/cm2 were bound in the wells and the immobilization capaci- ties for biotinylated albumin were 1.13 - 1.76 µg/cm2. In another dextran-based ap- proach, functional streptavidin surfaces were constructed on gold-layered piezoelectric crystals (Chen and Lin, 2007). The first layer was fabricated with amine-terminated alkanethiols to which periodate-oxidized dextran was coupled through a reaction be- tween amines and dextran aldehyde moieties. Streptavidin was immobilized either covalently by reductive amination to the dextran aldehyde groups or through biotin linkers coupled to dextran. Both these hydrogel-based streptavidin-coatings showed further ability to bind biotinylated BSA. Polyamidoamine (PAMAM) dendrimers are synthetic copolymers of repeated mono- meric units such as ethylenediamine and methylacrylate. These structures have a de- fined number of either reactive amine-, hydroxyl- or carboxyl-groups on their surface (Singh et al., 1994; Singh, 1998; Nourse et al., 2000). The well-characterized structure and functionality of the dendrimers favor their use as the building blocks in biological structures such as the interface layers for the immobilization of proteins. Amine- terminated dendrimers have been used as the carrier polymers for the immobilization of streptavidin and avidin to aminosilylated glass chips (Benters et al., 2001), to mag- netic particles (Gao et al., 2005) and to gold substrates for the analysis of the surface by various means (Hong et al., 2003; Liu and Amiridis, 2004; Mark et al., 2004; Hong et al., 2005; Kim et al., 2006). Both covalent coupling and biotinylated dendrimers have been used for the immobilization of streptavidin. Various surface research meth- ods in all these studies have shown either increased binding capacities or improved homogeneity of the surfaces with dendrimer-based coatings in comparison with respec- tive control surfaces without the dendrimer interface. Polylysine, a synthetic polymer of amino acid lysine, is widely used to render nega- tively charged solid substrates, for example glass and metal oxides, to positively charged surfaces and, furthermore, to provide primary amines for covalent coupling (Eisen and Brown, 1999; Haab et al., 2001; Kusnezow et al., 2003). Polyethylene gly- col (PEG) and polyethylene oxide (PEO, referring to longer polymers) are known for their ability to resist non-specific binding of proteins when layered on surfaces (re- viewed by Kingshott and Griesser, 1999). A co-polymer of polylysine and PEG, poly(L-lysine)-grafted poly(ethylene glycol) (PLL-g-PEG), has been used as an inter- face layer for the immobilization of streptavidin (Ruiz-Taylor et al., 2001a; 2001b). Defined ratios of polylysine, PEG and biotin-terminated PEG strands reacted to pro- duce a desired co-polymer. When layered on the titanium oxide surface, the primary amines of polylysine provided binding of the polymer to the negatively charged sub- 41 Review of the literature 42 strate whereas the PEG-strands, some of them terminated with biotin, protruded out of the surface. The grafting and the biotinylation degrees and thus the properties of the surface could be modified by adjusting the stoichiometric ratios of the reagents. Low non-specific binding of proteins to this surface was observed and streptavidin was effi- ciently immobilized by the biotinylated polymer up to monolayer densities of 6.3 pmol/cm2. This type of streptavidin surface was subsequently used by Peluso et al. (2003) in a study which compared the performance of random and oriented immobili- zation of biotinylated antibodies and Fab' fragments. The interface polymer is not a prerequisite for enhanced coating performance. Homo- geneous polymerization of streptavidin without other macromolecules may also be utilized, as shown, for example, in one of the study (I). In another example, Desai et al. (2003) used maleimide- and thiol-chemistry to polymerize streptavidin, resulting in a mixture of products consisting of monomeric to tetrameric conjugates. When coated on polystyrene, this surface showed a 25-fold greater biotin-binding capacity and a 70 % higher binding capacity for biotinylated peptide compared with the surface manufac- tured using non-polymerized streptavidin. Other studies have also reported the en- hanced performance of surfaces made from homogeneous streptavidin polymers (Tischer et al., 1991; Berger et al., 1993). 1.4.6 Recombinant approach and self-assembly Some articles have reported that recombinant and fusion protein technologies improve the immobilization properties of streptavidin, albeit most of the numerous mutations of streptavidin have been targeted at the biotin-binding site and at the subunit interfaces mainly for scientific interest (see section 1.2.3.3). One genetically modified variant (Reznik et al., 2001) possessed a six-residue immobilization sequence (Gly-Gly-Ser- Gly-Cys-Pro) fused to the C-terminus of the streptavidin subunit. This recombinant streptavidin was covalently coupled to a maleimide-functionalized microtiter plate through the unpaired thiol groups of the cysteine residues. The biotin-binding proper- ties of this variant were minimally affected when bound on the surface and the accessi- bility of the binding sites for biotinylated macromolecules was improved in compari- son with natural core streptavidin due to the linkers that separated the protein from the surface. The fusion protein derivatives and the genetic variants of streptavidin have been fre- quently used in combination with self-assembled monolayers. The self-assembled monolayers are highly ordered single-molecular layers which are generated spontane- ously by the inherent properties of the organizing molecules. A classical example of this is the organization of alkanethiols on gold surfaces (Nuzzo and Allara, 1983; Por- ter et al., 1987). In these structures, aliphatic alkyl molecules with a thiol-group in one end bind tightly to the gold surface due to the strong interaction between gold and sul- fur. Meanwhile the protruding alkyl chains are organized in highly ordered manner. Self-assembling properties have also been recognized with some proteins. A well- characterized protein which self-assembles in vitro and has potential in nanotechnology 42 Review of the literature 43 is the S-layer protein from the surface of gram-positive bacteria (Jaenicke et al., 1985; Pum and Sleytr, 1999; Sleytr and Beveridge, 1999; Sleytr et al., 1999; 2007). In one application, Moll et al. (2002) fused an S-layer protein SbsB from Geobacillus stearothermophilus to a streptavidin subunit and constructed four N-terminal and two C-terminal fusions. The fusion proteins were produced in E. coli and then refolded in a ratio of 1:3 with native subunits of core streptavidin in order to construct a hetero- tetrameric streptavidin containing one subunit with the S-layer protein extension. The transmission electron microscopy images showed that the tetramers with the N- terminal, but not the C-terminal, streptavidin fusions formed highly ordered layers. When layered on liposomes, the tetramers retained their capability to bind biotinylated proteins. Huber et al. (2006) introduced a principally similar construct; streptavidin self-assembling property of this S-layer protein is dependent on the calcium ions which enables a controllable assembly in vitro. Other self-assembling proteins, applicable in biotechnology, are hydrophobins from filamentous fungi (Wösten, 2001; Hektor and Scholtmeijer, 2005). Hydrophobins are involved in the growth and extension of the aerial hyphae of fungi and they self- assemble into amphipathic membranes at interfaces. Hydrophobins HFBI and HFBII from the mold Trichoderma reesei have been extensively studied with respect to their surface organization and feasibility for the immobilization of useful proteins in bio- technology (Linder et al., 2001; 2002; Torkkeli et al., 2002). Szilvay and co-workers (Szilvay et al., 2006; 2007) constructed genetically engineered variants of HFBI con- taining either N- or C-terminal short amino-acid linkers with an unpaired cysteine moi- ety serving for biotinylation with maleimide-biotin. The biotinylated HFBI-variants were used to immobilize avidin to the surface in a highly ordered manner yielding amounts of 8.1 - 9.8 pmol/cm2 which represented 140 - 170 % surface coverage com- pared with a complete monolayer density. 1.4.7 Oligonucleotide-mediated immobilization The principle of oligonucleotide-mediated, or DNA-directed immobilization (DDI), is to immobilize a protein-oligonucleotide conjugate by hybridization to a complemen- tary oligonucleotide on the solid surface (Fig. 8). The streptavidin-oligonucleotide conjugates have been constructed either through an affinity-based binding of bioti- nylated oligonucleotides to streptavidin or by the covalent coupling of chemically acti- vated streptavidin and DNA. Scouten and Konecny (1992) constructed a complex con- sisting of biotinylated single-stranded polyadenylic acid (poly[dA]), streptavidin and biotinylated antibody for immobilization and separation on thymidine-coated (poly[dT]) magnetic beads. The reversible binding of the complex to the particles was shown since 80 % of the bound radioactivity was eluted in mild conditions. In another application, Niemeyer et al. (1994) covalently coupled a 5'-thiolated oligonucleotide to streptavidin which was functionalized with maleimide groups by means of a heterobis- pecific cross-linker. A 1:1 complex of streptavidin and DNA was observed after the purification and characterization. The covalent DNA-streptavidin conjugates were 43 was fused to the C-terminus of an S-layer protein SbpA from Bacillus sphaericus. The Review of the literature 44 combined with biotinylated capture antibodies and subsequently the complexes were hybridized to complementary oligonucleotide-surfaces in microtiter wells. The immu- noassays performed in the wells demonstrated the transformation of an oligonucleotide surface into a functional protein surface. Solid substrate Complementary oligonucleotide pair Streptavidin Biotinylated antibody Linker Figure 8. The principle of oligonucleotide- mediated, or DNA-directed, immobiliza- tion of streptavidin and a biotinylated antibody. The linking between the oli- gonucleotide and streptavidin can be either biotin-mediated binding or based on cova- lent coupling of the protein to DNA. A comparative immobilization of biotinylated reporter enzymes (horseradish peroxi- dase, β-galactosidase and alkaline phosphatase) either by the DDI-process or by direct binding to the streptavidin-coated microtiter plates showed 1.7 - 2.7-fold higher inten- sities for the DDI-method (Niemeyer et al., 1999). Similar improvements, compared either with direct physisorption or with immobilization to an adsorbed streptavidin surface, were observed for the immobilization of antibodies in the same study. The results showed the general efficiency and gentleness of the DDI-process. Niemeyer et al. (2003) have, furthermore, reported a combination of the DDI of the capture anti- body with immuno-PCR detection technology (Sano et al., 1992), yielding an ex- tremely sensitive assay concept. The DDI of the capture antibody has been further used in an array-based microscale fluorescence immunoassay (DDI-µFIA) for simultaneous detection of four protein antigens using a Cy5 label (Wacker and Niemeyer, 2004), resulting in reasonable sensitivities and reproducibility. Ladd et al. (2004) described the functionalization of the biosensor surfaces by combining the self-assembled monolayers with DDI. The surface consisted of successive layers of biotin monolayer, streptavidin, biotinylated oligonucleotide and an oligonucleotide-coupled antibody to provide the capture of hCG and subsequent detection by surface plasmon resonance. Wacker et al. (2004) reported a comparison of three different capture concepts in an array-based antibody-assay. A capture antibody was immobilized either by DDI, direct adsorption or by streptavidin-coated surface for the binding and determination of rabbit IgG. Equal detection limits (150 pg/ml) were observed with each surface, but the DDI- method provided the most homogeneous spots when visualized by imaging and exhib- 44 Review of the literature 45 ited the least variation between the repeated tests. Furthermore, DDI enabled equal performance with the use of 100 times less antibody compared with direct spotting. 1.4.8 Peptide tags Various peptide tags fused to desired target proteins have recently become popular in biotechnology, essentially enabling reversible immobilization of recombinant fusion proteins to affinity purification matrices (reviewed by Terpe, 2003). In addition to puri- fication purposes, peptide tags have been used to some extent to immobilize target proteins to surfaces in a specified orientation through the tag. For example, a positively charged arginine-tag has been used to facilitate the immobilization of proteins to nega- tively charged mica (Nock et al., 1997) and the FLAG-tag (a well-defined common octapeptide) has been used to enable an oriented immobilization of subtilisin to a sur- face coated with anti-FLAG monoclonal antibodies (Wang et al., 2001). Perhaps the most common peptide-tag mediated immobilization is based on the polyhistidine tag which makes use of the interaction of histidine with certain metal ions such as nickel (Ni2+). This technology has been used, for example, in surface plasmon resonance sen- sor chips (Gershon and Khilko, 1995; Nieba et al., 1997), for the immobilization of enzymes (Carlsson et al., 1996) and for the oriented immobilization of Fab fragments (Vallina-García et al., 2007). Immobilization using peptide tags requires a functional primary layer such as antibod- ies against the tags or a metal-activated surface such as a nitrilotriacetic acid (NTA) matrix or other nickel chelators (Johnson and Martin, 2005). The additional layers may be sources of instability and they can introduce some inconvenience to the preparation of the surface. Thus, peptides exhibiting direct affinity to a solid substrate, for example to polystyrene, would be of interest for direct immobilization. Loomans and co- workers (Loomans et al., 1997; 1998a; 1998b) showed that certain N-terminal exten- sions (oligo-lysine moieties) enhanced the adsorption of desired epitope peptides on polystyrene which enhanced their immunoreactivity in ELISA-assays. Sakiyama et al. (2004) and Kumada et al. (2006) have screened dodecapeptides exhibiting affinity to polystyrene from an E. coli random peptide display library using the enzymatic activity of glutathione-S-transferase as a reporter of the adsorption efficiency. One of their screened peptides was chemically linked to either streptavidin or monoclonal antibody (Kumada et al., 2007). These proteins showed higher binding to the hydrophilic poly- styrene surface and also enabled improved performance over conventional technology in ELISA-assays. In addition to the polystyrene-specific peptides, polypeptides which have direct affinity to various inorganic surfaces have been designed and synthesized, facilitating the assembly of nanostructures to sensor surfaces (reviewed by Sarikaya et al., 2003). 45 Aims of the study 46 2 AIMS OF THE STUDY The aim of the study was to discover the effects of specified streptavidin surface prop- erties on the performance of solid-phase immunoassays and examine the potential to improve assay performance by modifications of the binding surface. The specific aims were to: 1. Model the surface and investigate the adsorption capabilities of streptavidin. 2. Prepare various reporter molecules and implement straightforward assays for the characterization and monitoring of the surfaces. 3. Develop surface performance by selected, straightforward modifications to the protein and coating process. 4. Investigate various streptavidin-coating chemistries in use and determine the influence of the defined surface properties on the efficiency of various immu- noassay types through a comprehensive study of the surfaces exhibiting dis- tinct features. 5. Investigate the potential of preparing condensed, high density capture surfaces and study their influence on immunoassays which are based on the surface readout measurement of time-resolved fluorescence. 46 Summary of the materials and methods 47 3 SUMMARY OF THE MATERIALS AND METHODS An overview of the materials and methods is given below. Detailed descriptions are presented in the original publications. 3.1 Streptavidin and coating (I - IV) 3.1.1 Streptavidin Streptavidin was mainly from Biospa (Milan, Italy). It was delivered as lyophilized powder, then dissolved in pure water (usually 10 mg/ml) and stored in aliquots at −20 °C. The product was characterized at the Department of Biotechnology using N- terminal sequencing and SDS-PAGE analysis. 3.1.2 Solid substrates Microtiter wells either in the C12 strip format, in separate single wells or as special spot wells were used as the solid substrates for the coatings. The wells were fabricated from low-fluorescent polystyrene and were irradiated to enable maximal protein ad- sorption capability. The spot wells (IV) had an indentation of approximately 0.2 mm in depth and either 2.5 mm, 3.5 mm or 4.5 mm in diameter at the bottom of the single- well format microtiter well. The spot was introduced in the well within the injection molding. All wells and strips were from Nunc A/S (Roskilde, Denmark). 3.1.3 Coating Native, unmodified streptavidin or some of the modified variants were diluted in an appropriate coating buffer which typically consisted of 100 mM Na2HPO4 and 50 mM citric acid adjusted to pH 5.0. The coating volume was 200 µl per well. The plates were incubated at +35 ºC and the evaporation of the solution was minimized by closing the plates in a humid container. The plates were washed after coating using a wash solu- tion containing Tween 20 (0.05 % v/v) as a detergent to provide efficient washing. The streptavidin-coated surfaces were subsequently incubated with protective and blocking agents (BSA, D-sorbitol). The blocking solution was aspirated after overnight incuba- tion and the plates were dried, packed with desiccant and stored at +4 ºC. Typically, the coated plates retained their initial biotin-binding activities for several years. 3.1.4 Spot surfaces (IV) The purpose of the spot approach was to introduce the binding surface to coincide with the excitation beam and condense the capturing of the labeled antibodies to increase the signal-to-background ratios of the assays that utilize the direct surface readout measurement. Modified single microtiter wells holding a circular indentation at the bottom were used as templates for the droplets of the coating solution. The droplet volume ranged from 4 to 15 µl depending on the spot diameter. Excluding the volumes and concentrations used, the stages of the coating process were principally similar to the regular protocol including the overnight incubation and the blocking steps. Avoid- 47 Summary of the materials and methods 48 ance of evaporation was considered especially critical while incubating the droplets at the elevated temperature. 3.1.5 Modified streptavidin (I, III, IV) Streptavidin was subjected to amine reactive glutaraldehyde prior to coating (I, IV) in a reaction containing approximately 100 mM (1 % v/v) glutaraldehyde and 35 µM strep- tavidin. The reaction mixture was purified by gel filtration or successive desalting col- umns. The pretreatment resulted in a modified, polymerized protein (referred to as GA- SAv) with enhanced adsorption capability. The reaction conditions in terms of the temperature, time and concentrations were studied in order to obtain an optimal po- lymerization degree and adsorption capability. Alternatively, streptavidin was modified using thiol-chemistry (III). Streptavidin was first reacted with a 40-fold molar excess (2500 µM) of amine reactive N-succinimidyl S-acetylthioacetate (SATA, Pierce Biotechnology, originally described by Duncan et al., 1983). The incorporated heterobifunctional linkers were deacetylated using hydro- xylamine to reveal reactive thiols which served for polymerization and efficient bind- ing to the surface. The number of the reactive thiols incorporated to streptavidin was determined by Ellman’s reaction using cysteine as a calibrator of the sulfhydryl quan- tity. 3.1.6 Fractionation and size-exclusion chromatography (I) Streptavidin and the polymerized variant (GA-SAv) were analyzed and fractionated by size-exclusion chromatography on Superose 12 HR10/30 column, run by the ÄKTA explorer system (both from GE Healthcare, Uppsala, Sweden). Alkaline buffers, for example, Tris-Cl or borate buffers at pH 8.4 - 8.8, without NaCl, were found suitable for the elution of streptavidin and GA-SAv. 3.2 Biotinylation and labeling of the antibodies and reporter molecules 3.2.1 Biotinylation and labeling with lanthanide chelates Modified biotin containing a reactive isothiocyanate group in the end of the extended side chain (Mukkala et al., 1993) was used for the biotinylation of proteins (I, III, IV). The pH of the biotinylation reaction was adjusted to 9.8 to provide optimal conditions for the reaction between the amines and the isothiocyanate group. Typically, an 80-fold molar excess of biotin over the protein was used and the protein concentrations in the reactions were approximately 6 - 8 µM for the antibodies and 70 - 80 µM for myoglo- bin. The biotinylated proteins were separated from non-reacted biotin on NAP- or PD- 10 desalting columns (GE Healthcare). The biotinylated antibodies were characterized using an in-house test where a sample of the antibody was immobilized to a strepta- vidin-coated well and subsequently an aliquot of the liquid was transferred to an anti- mouse IgG-coated plate to assess the fraction not bound to the streptavidin surface. The test indicated the proportion of the antibody molecules bearing functional biotin moie- ties and the result was typically 90 - 100 %. 48 Summary of the materials and methods 49 The principles of the lanthanide chelate labels and their measurement will be described below in section 3.3. Proteins were labeled with inherently fluorescent seven-dentate (Takalo et al., 1994) or nine-dentate (von Lode et al., 2003) Eu-chelates (I, III, IV). Alternatively, a terbium- (Tb) chelate was used (Mukkala et al., 1989) for the labeling of streptavidin (II). Amine reactive isothiocyanate chemistry was used for the coupling. The protein concentrations in the labeling reactions were approximately 4 - 10 µM for the antibodies, 40 µM for myoglobin and 40 µM for streptavidin. The chelates were used in 5 - 60-fold molar excess over the protein depending on the desired purpose; the biotinylated reporter molecules were labeled with a lower amount whereas the detec- tion antibodies for the immunoassays were labeled with larger excess to guarantee a higher degree of labeling. The labeled proteins were purified by size-exclusion chro- matography on a Superdex 200 HR10/30 column (GE Healthcare). The concentrations of the proteins, after labeling or any other process, were typically assayed by measuring absorbance at 280 nm or by a Bradford protein assay (Bradford, 1976). Quantification by absorbance was based on the protein-specific absorbance values which were 1.34 AU (absorbance units) for the antibodies and 2.8 AU for strep- tavidin when the concentration of the given protein is 1 mg/ml. The value for strepta- vidin deviated a little from that reported in the literature (3.1 - 3.4 AU), but the value used was experimentally confirmed with several dissolved batches. The Bradford pro- tein assay frequently resulted in a weak response with streptavidin. An enhanced method based on heating (Sharma and Tihon, 1988) increased the absorbance values but complicated the assay. Therefore, the Bradford assay was not routinely used for assaying the concentrations of streptavidin. An alternative color-forming assay, the Micro BCA assay (Pierce Biotechnology) was used instead. The labeled proteins were further characterized with respect to their lanthanide ion contents. The concentrations of the ions were measured against respective calibrators using DELFIA® measurement (Dissociation-Enhanced Lanthanide Fluorescent Immu- noassay, PerkinElmer Life Sciences and Analytical Sciences - Wallac Oy, Turku, Finland). The lanthanide concentration defined was divided by the protein concentra- tion to calculate the degree of labeling (lanthanide chelates/protein). 3.2.2 Reporter molecules Biotinylated and labeled reporter molecules provided a straightforward, direct means for studying streptavidin-coated surfaces (I - IV). Biotinylation was first introduced to a part of the amines and subsequently a part of the remaining amines was used for the coupling of the chelates. Monoclonal antibody H117 (Lövgren et al., 1995; Piironen et al., 1998) was used (I, III) for the determination of the binding capacities for a large molecule (MW ≈ 160 000 D) whereas myoglobin (I) represented a smaller macromole- cule (MW ≈ 17 000 D). The labeling degrees of the reporter molecules were adjusted to be quite low (approximately 1 - 2) to avoid too high signal intensities in the assays where abundantly bound reporters were measured. 49 Summary of the materials and methods 50 The smallest reporter molecule, Eu-biotin, was synthesized from biotin and the seven- dentate Eu-chelate in one of the studies of this thesis (II). The product was purified using reverse phase high performance liquid chromatography (HPLC) and its molecular weight was assayed 1036 D by means of mass spectroscopy. Eu-biotin was intended to simulate the binding of intact biotin and report the total binding capacity of the surface without steric constraints. The degree of labeling for Eu-biotin was exactly 1.0. 3.3 Fluorescence measurement (I - IV) The fluorescence detection technology utilized in this study was based on lanthanide chelate labels. Lanthanide chelate labels consist of lanthanide ions (inner transition metals) encompassed by strong chelators (ligands). Typically, trivalent europium (Eu) and terbium (Tb) ions are used, but samarium (Sm) and dysprosium (Dy) are also use- ful. While bare ions are poor absorbers, the ligand absorbs the excitation energy and transfers the energy from the excited S1 state to the triplet state T1 (intersystem cross- ing) and further through an intermolecular energy transfer to the bound lanthanide ion which then emits specific fluorescence. The lanthanide labels exhibit narrow emission peaks and large Stokes' shift (the excitation and the emission wavelengths are sepa- rated by hundreds of nanometers) which prevents the scattering of the excitation light into the measurement window. The main advantage, however, is the long-lasting fluo- rescence enabling time-resolved recording after the decay of prompt background fluo- rescence typically generated by biological material. The characteristics of lanthanide measurement are high sensitivity and wide dynamic range. For example, the dissocia- tion-enhanced measurement (see below) of the Eu ions exhibits a detection capability of approximately 10–14 - 10–13 M concentrations and a linearity range of six decades. (Siitari et al., 1983; Hemmilä et al., 1984; Soini and Lövgren, 1987.) Time-resolved fluorescence of the lanthanide labels was measured using Victor 1420 Multilabel counters (Wallac Oy). A 340 nm excitation filter was used and the emission filters were 615 nm and 545 nm for Eu and Tb, respectively. The Eu measurement used 1000 µs cycles (time between successive flashes) with 400 µs delay and 400 µs re- cording times. The respective values for the Tb measurement were 2000 µs (cycle), 500 µs (delay) and 1400 µs (recording window). Two principally different measuring methods were employed (Fig. 9). In dissociation-enhanced measurement, the lanthanide ion is dissociated from a cou- pling chelate (a chelate where the ion is coupled to an other molecule, for example to a protein) into a solution by lowering the pH. In the solution, the ion is chelated by a second ligand structure (the enhancement chelate) forming a fluorescent complex. So- lution measurement was performed using DELFIA Enhancement solution (Wallac Oy). This measurement method represented an integrated signal from the entire area of the captured reporter molecule and enabled the straightforward quantification of the bound molecules by measuring the appropriate lanthanide calibrators. Since the labeled re- porter molecules were directly and abundantly bound to the surface, the instrumental 50 Summary of the materials and methods 51 response had to be decreased by reducing the excitation intensity and replacing a smaller emission slot compared with regular Eu measurement. Figure 9. The principles of surface readout measurement (left) and solution measurement (right). Development in lanthanide label technology has led to inherently fluorescent chelates (Takalo et al., 1994; von Lode et al., 2003) where the coupling and sensitizer function- alities are introduced in the same structure. These labels can be measured directly from a washed surface without an additional development step. This provided the basis for the other measurement method used in the studies, the surface readout measurement, where the excitation beam is focused on the bottom of the well and the response repre- sents a fraction of the bound label. The signal reflects the density of the labeled mole- cule under the beam projection rather than the total capacity of the well. The wells were regularly dried under warm air to eliminate the excess moisture before being sub- jected to surface measurement. Of the chelate types used, the seven-dentate Eu-chelate (Eu ion coordinated by seven bonds in the chelate) was utilized both in the surface readout and dissociation-enhanced measurements, the nine-dentate Eu-chelate was used mainly in surface readout measurement and Tb-chelate in dissociation-enhanced measurement only. 3.4 Characterization of streptavidin-coated surfaces 3.4.1 Binding capacities (I - IV) The biotinylated and labeled reporter molecules were prepared in the assay buffer (In- notrac buffer solution red, Innotrac Diagnostics Oy, Turku, Finland) and subsequently bound into the streptavidin-coated wells, typically by shaking for one hour at room temperature. After washing and drying, the plates were subjected to surface readout and solution measurement. The molar quantification of the immobilized reporter mole- cules was based on the determination of the lanthanide ion concentration in solution measurement against the respective lanthanide ion calibrators. The lanthanide concen- 51 Summary of the materials and methods 52 tration measured was divided by the degree of labeling defined for the reporter mole- cule to calculate the quantity of the bound molecule. 3.4.2 Protein assay (II) A sensitive Micro BCA protein assay was used to quantify adsorbed streptavidin. The assay is based on the reduction of the Cu2+ ion to Cu1+ ion by the protein. Conse- quently, the Cu1+ ion forms a complex with bicinchoninic acid (BCA) which exhibits strong absorbance at 562 nm (Smith et al., 1985). Adsorbed streptavidin was quantified against streptavidin calibrators dried in the wells. The drying of the calibrators was intended to mimic the conditions of the adsorbed protein. The absorbance was read with the Victor device using a 545 nm filter which was closest to the specified optimal 562 nm wavelength. The sensitivity of the assay was, however, well preserved in the slightly deviating wavelength. 3.4.3 Desorption assay (III) Desorption, or leaching, of adsorbed streptavidin from the surface was studied under regular assay conditions. Assay buffer (Innotrac Diagnostics Oy) was shaken in the streptavidin-coated wells after which an aliquot of the buffer was transferred to a bioti- nylated surface. The biotinylated surface consisted of a biotinylated IgM antibody coated into the 96-well plate. The plates were prepared and optimized in-house. De- sorbed streptavidin, which bound to the Bio-IgM surface, was detected using Eu-biotin. The quantification was enabled by employing a series of defined streptavidin concen- trations in parallel Bio-IgM-coated wells. 3.5 Immunoassays 3.5.1 Time-resolved immunofluorometric assays (I, III, IV) Several immunoassays were used as models to study various aspects of the strepta- vidin-coated surfaces to immunoassay performance. The time-resolved im- munofluorometric (TR-IFMA) assays consisted of separate immobilization of the biotinylated capture antibody to the streptavidin-coated surface and subsequent incuba- tion of the antigen and the labeled tracer antibody. The TR-IFMA of PSA (I) utilized biotinylated capture antibody H117 or alternatively a recombinant Fab fragment of it. The Fab fragment was cloned, produced and bioti- nylated site-specifically at the Department of Biotechnology (Eriksson et al., 2000). The input quantities of the antibodies used in the assays were 400 ng/well of the intact antibody and 600 ng/well of the Fab fragment. The detection antibody was Eu-labeled 5F7 (Nurmikko et al., 2000), which was used at 200 ng or 600 ng in the assay. The hCG-assay (I) utilized a biotinylated antibody E27 as the immobilized capturing antibody (400 ng/well) and an Eu-labeled antibody 8D10 as a tracer (300 ng/well). The antibodies were from Wallac Oy. 52 Summary of the materials and methods 53 The TSH-assays (III, IV) were carried out using various antibody combinations. One set consisted of biotinylated 5409 and Eu-labeled 5405 and the other comprised bioti- nylated 5404 and Eu-labeled 5409. The quantities of the capture antibodies used for immobilization varied from 100 ng to 600 ng in a regular assay and 32 - 160 ng in spot-assays. The Eu-labeled detection antibodies were used at 50 ng per reaction. All antibodies of the TSH-assays were from Medix Biochemica Oy (Kauniainen, Finland). 3.5.2 Enzyme immunoassays (III) The enzyme immunoassay (EIA) kits were gifts from Fujirebio Diagnostics AB (for- merly CanAg Diagnostics AB, Gothenburg, Sweden). The PSA assay was intended for a quantitative determination of total PSA. The assay comprised a simultaneous incuba- tion of the biotinylated antibody, antigen and the horseradish peroxidase (HRP) conju- gated tracer antibody in one mixture. The PSA-assay was a sandwich type assay in principle, but it deviated from the above-mentioned TR-IFMA assays which used sepa- rate incubation steps for the biotinylated capture antibodies. The EIA-assay of the cancer associated antigen 125 (CA125) started with a simultane- ous incubation of the biotinylated capture antibody (Ov197) and the antigen. The de- tection antibody (Ov185) was added after washing. The detection technologies of both EIA-assays were based on the HRP-conjugated tracer antibodies and the photometric measurement of the color formed upon the substrate conversion at 620 nm. 53 Summary of the results 54 4 SUMMARY OF THE RESULTS Detailed descriptions of the results are reported in the original publications. Here is a summary of the results accompanied with some previously unpublished data. 4.1 Characterization of streptavidin used (unpublished) One batch of streptavidin from the supplier was subjected to N-terminal sequencing. The sequencing showed a main N-terminal sequence of Ala-Glu-Ala-Gly-Ile-Thr-Gly, which confirmed the preparation as core streptavidin. In the SDS-PAGE analysis the protein appeared as a pure single band at approximately 14 kD when temperatures of +80 - 100 °C were used for the preparation of the samples, which further verified the presence of core streptavidin. 4.2 Modeling and quantification of the adsorption (II) 4.2.1 Theoretical modeling of the adsorbed layer The three-dimensional structure and the dimensions of tetrameric streptavidin were disclosed in diffraction studies in the late 1980s (Hendrickson et al., 1989; Weber et al., 1989). The dimensions of a single molecule are approximately 54 × 58 × 48 Å (5.4 × 5.8 × 4.8 nm). These values were employed to calculate the theoretical maximum monolayer coverage of adsorbed streptavidin considering two quadratic patterns (end- on and side-on orientations) and a globular adsorption pattern (Fig. 10). Figure 10. Two possible models for the organization of the streptavidin monolayers. A) A quadratic pattern with an end-on orientation yields a coverage of 6.41 pmol/cm2 (380 ng). B) A globular pattern with an estimated molecular diameter of 5.2 nm yields a maximum monolayer coverage of 7.09 pmol/cm2 (430 ng). The patterns were used for the modeling of the surfaces in original publication II. Since the actual adsorption pattern is unknown, an average of the models conferring to the highest (globular) and the lowest (side-on with 5.4 × 5.8 nm side to the surface) adsorption capacities was employed. The calculated value, 6.20 pmol/cm2 (370 ng) was subsequently used as an estimate of the maximal monolayer adsorption capacity for streptavidin. Consequently, the maximum capacity of a microtiter well coated with a typical 200 µl volume utilizing a binding area of 1.54 cm2 equals to 9.55 pmol (570 ng). 54 Summary of the results 55 0 1000 2000 3000 0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 B Eu -b io tin b ou nd (p m ol /w el l) SAv input in coating (ng/well) 0 1000 2000 3000 0 100 200 300 400 500 A Ad so rb ed (n g/ w el l) SAv input in coating (ng/well) 4.2.2 Quantification of adsorbed streptavidin Adsorbed protein was quantified using two independent methods; either by Tb-labeled streptavidin or by measuring adsorbed streptavidin using a sensitive protein assay. The Micro BCA assay is one of the most sensitive colorimetric protein assays providing sufficient sensitivity for measuring adsorbed streptavidin directly from the well. The results of the fluorescent assay and the protein assay were consistent with respect to maximum adsorption, the 350 ng/well with the Tb-labeled streptavidin and the 380 ng/well with the protein assay (Fig. 11 A). The quantities represented 61 - 67 % of the maximum monolayer coverage. The monitoring of adsorbed streptavidin using Eu- biotin (Fig. 11 B) complemented the direct measurement of adsorbed protein and gave estimates of the average biotin-binding sites available per adsorbed streptavidin mole- cule. Table 1 summarizes the calculated monolayer capacities and experimentally ob- served results. Figure 11. A) The quantification of streptavidin adsorbed in the microtiter well by means of a sensitive protein assay (Micro BCA). B) Monitoring of adsorbed streptavidin through the im- mobilization of Eu-biotin to the coated wells. The graphs are modified from original publica- tion II. Table 1. The theoretical monolayer adsorption capacities of streptavidin and the actual ad- sorbed quantities measured from the microtiter wells. Based on the data from original publica- tion II. pmol ng pmol ng Maximum monolayer 6.2 370 9.6 570 Measured 4.1 250 6.3 380 a Well of C-geometry with a coated area of 1.54 cm2 Per cm2 In the well a 55 Summary of the results 56 The adsorbed quantities of streptavidin observed in the microtiter wells using labeled streptavidin and the protein assay for the monitoring were consistent with the previous results of maximal quantities (250 ng/cm2, 49 - 63 % coverage) observed using scan- ning probe microscopy and quartz crystal microbalance (Ihalainen and Peltonen, 2004). 4.3 Binding capacities (I - IV) 4.3.1 Steric effects Size-dependent steric hindrance to the binding of the reporter molecules was clearly recognized when comparing the binding of the three reporter molecules representing different molecular sizes. For example, a regular streptavidin surface bound approxi- mately 8 - 11 pmol of Eu-biotin, 2.8 pmol of biotinylated myoglobin and 2.2 pmol of biotinylated antibody. In addition, the masking effect was observed when the reporter molecules were immobilized to the streptavidin-coated surfaces of different binding capacities. For example, the ratios of the bound quantities between the GA-SAv- surfaces and the regular streptavidin surfaces were 1.6 with antibody, 1.8 with my- oglobin and 2.4 with Eu-biotin (Fig. 12). Furthermore, the comparison of the strepta- vidin-coated plates from different manufacturers (III) showed the binding-capacity ratios of more than 30-fold with Eu-biotin but only 5.3-fold with the antibody between the plates of the highest and the lowest binding capacities. The results indicate that a larger proportion of the biotin-binding sites of the streptavidin-coated surface are util- ized when immobilizing small molecules and the increased binding capacities and den- sities provided by the streptavidin layer are not completely benefited when immobiliz- ing large molecules. 0 20 40 60 80 100 0 50 % 100 % 150 % 200 % 250 % C Eu-biotin input (pmol/well) 0 2 4 6 8 10 12 14 0 25 % 50 % 75 % 100 % 125 % 150 % 175 % A R at io Bio-Mab input (pmol/well) 0 2 4 6 8 10 12 14 0 25 % 50 % 75 % 100 % 125 % 150 % 175 % 200 % B Bio-myoglobin input (pmol/well) Figure 12. The relative binding capacities of a regular streptavidin-coated surface (●) in com- parison with the high-capacity surface of polymerized streptavidin (GA-SAv) (▲) with respect to the size of the reporter molecule. A) Biotinylated antibody (MW = 160 000). B) Biotinylated myoglobin (MW = 17 000). C) Eu-biotin (MW = 1 036). Graphs A and B are modified from original publication I, graph C contains unpublished data. 56 Summary of the results 57 4.3.2 Total capacities and binding site densities The total binding capacities of the streptavidin-coated microtiter wells (the 96-well format) defined in the studies of this thesis ranged from 4.4 to more than 150 pmol/well for Eu-biotin and from 1.2 to 6.4 pmol/well for the biotinylated anti- body. The total binding capacity of the well is dependent on the coated surface area, and it is not actually a good indicator of the performance of the coating chemistries used. Therefore, it was necessary to employ another parameter, the area-corrected spe- cific capacity, or the binding density, to assess a universal value for convenient com- parison of various coating chemistries and conditions (III, IV). The binding site density was calculated by dividing the defined total capacity (pmol) by the coated surface area (mm2). The 96-well plates were coated using volumes of either 100 µl, 200 µl or 300 µl per well and the respective surface areas covered in the wells were approxi- mately 90 mm2, 150 mm2 and 210 mm2. The binding areas were based on the specifica- tions by Esser (1997a) and experimental measurements of the filled wells. The calcu- lated binding densities varied from 0.040 to more than 1.00 pmol/mm2 for Eu-biotin and 0.011 to 0.061 pmol/mm2 for the antibody. The plates representing the highest binding capacities and densities were found among the commercial plates when comparing the features of the commercial and in-house 96-well plates (III). The coating chemistries used for the plates were determined as extensively as could be found in the scientific literature, patent files and manufacturer's technical notes. Some of these data are reviewed in section 1.4 of this thesis, though without indications to any particular plate. 4.3.3 Spot surfaces (IV) Regular streptavidin (referred as SAv in the study, IV) and polymerized streptavidin (GA-SAv) were successfully coated in the droplet format yielding functional and evenly distributed spot surfaces. The binding capacities of the SAv-coated and GA- SAv-coated spots ranged from 0.5 to 5.5 pmol for Eu-biotin depending on the spot size and the coating type. The respective binding site densities of the spots were 0.076 - 0.47 pmol/mm2 which were at least equal to those obtained with the ordinary method using larger volumes. The GA-SAv-surface further provided a close to 6-fold increase in binding site densities compared with the surface prepared from unmodified strepta- vidin (SAv). 4.4 Desorption (III, and unpublished) The desorption (or leaching) of streptavidin from the streptavidin-coated wells was quantified by subjecting the surface to a mobile buffer solution followed by the transfer of an aliquot of the solution to a biotinylated surface (Bio-IgM) to serve the binding of desorbed streptavidin and detection using Eu-biotin. The dose-response curve of the streptavidin calibrators on the Bio-IgM surface appeared slightly sigmoidal in shape and the assay allowed the quantification of 0.1 - 80 ng streptavidin when bound to the Bio-IgM surface (Fig. 13). The concave shape of the curve in the low region evidently 57 Summary of the results 58 0 20 40 60 80 100 0 1 000 000 2 000 000 3 000 000 4 000 000 5 000 000 6 000 000 cp s Streptavidin (ng/well) 0 2 4 6 8 10 0 200 000 400 000 600 000 800 000 originates from the proportionally higher occupation of streptavidin's four biotin- binding sites when using small streptavidin quantities. The biotin-binding sites are readily occupied by the biotin residues from the Bio-IgM surface while a smaller pro- portion remains detectable by Eu-biotin compared with the more crowded situation of the higher streptavidin quantities. The occupation of the biotin-binding sites was ex- perimentally confirmed since the low-end lag region was significantly extended after increased amounts of the coated Bio-IgM. Therefore, the construction of the assay required careful optimization to compromise between the sensitivity and adequate dy- namic range of the assay. Desorbed streptavidin was calculated using an appropriate fitting of the sigmoidal curve or alternatively by dividing the graph into smaller fragments and applying quad- ratic fitting to the sections of the curve. The desorbed quantities from various strepta- vidin-coated microtiter wells (commercial and in-house plates) ranged from 0.5 to 76 ng/well. These represented approximately 0.003 - 3.4 % of the total biotin- binding sites of the wells. The washing of the streptavidin-coated wells before the de- sorption assay typically resulted in decreased leaching. Some surfaces exhibited rather low desorption even from unwashed wells proving their inherently stable characteris- tics. Figure 13. The calibration curve of strep- tavidin on the Bio-IgM surface, the assay that was used for the quantification of streptavidin desorbed from the strepta- vidin-coated wells. The inset graph shows the dose-response relation with low strep- tavidin quantities (0.5 - 10 ng). Unpub- lished figure, drawn according to data from original publication III. A prerequisite of the desorption test used is that the dissociating units consist of at least subunit dimers of streptavidin. The test is not responsive for single subunits (unless the subunits are refolded in the solution). One of the two dimeric interfaces of streptavidin is weaker than the other, at least without bound biotin (see sections 1.2.1 and 1.2.3). In addition, the superficial protein layers are typically dissolved more readily than those which are in direct contact with the solid substrate (section 1.3.3). Therefore, it may indeed be true that the dissociating units are the subunit dimers. The data could not, however, distinguish between the dimeric and intact tetrameric structures. 58 Summary of the results 59 4 6 8 10 12 14 16 18 0 200 400 600 800 1000 1200 1400 V = 7.89 ml 3 2 1 V = 10.28 ml V = 8.43 ml A 2 80 (m AU ) Elution volume (ml) 4.5 Modifications (I, III, IV, and unpublished) 4.5.1 Polymerized streptavidin (I, and unpublished data) Streptavidin was successfully polymerized (GA-SAv) using chemical treatment as indicated by size-exclusion chromatography analysis (Fig. 14). The polymerization degree could be adjusted depending on the conditions used. For example, the increased protein concentration resulted both in larger sizes of the polymer and a higher propor- tion of the polymerized protein. The largest conjugates represented molecular weights above 200 kD. Before the analysis and the fractionation of streptavidin and GA-SAv by size-exclusion chromatography, efforts were made to find appropriate conditions for the elution. The regular starting conditions frequently used for proteins, such as a buffer of pH 7.0 - 7.5 containing 150 mM NaCl, were found totally inappropriate for streptavidin. The elution profile appeared as a flat and broad peak and several column volumes of the solution were required to run the protein completely out from the col- umn. The non-specific adhesion of the protein to the matrix was found to originate from hydrophobic forces rather than ionic interactions, since the removal of NaCl from the elution buffer and the raising the pH to 8.4 - 8.8 resulted in a more efficient resolu- tion. Figure 14. The elution profiles of polym- erized streptavidin (GA-SAv) and BSA on a Superose 12 size-exclusion column. GA- SAv was produced using protein concen- trations of 2 mg/ml (36 µM, solid line) and 4 mg/ml (72 µM, dashed line) in the po- lymerization reaction. BSA was run as a molecular weight control (dotted line). The three fractions used for the coating from the 2 mg/ml reaction are shown with bars. The figure is modified from original publi- cation I. The highest binding capacities of the coated surfaces were obtained when a non- fractionated mixture containing varying molecular sizes was used for the coating. Nei- ther larger polymers nor any fraction alone from the separation exceeded the capacity of the surface made from the mixture when monitored using the reporter antibody and myoglobin. In the time when the work with the polymerized streptavidin was made (I) the Eu- biotin reporter molecule was not available and the surfaces were monitored using pro- tein reporters (antibody and myoglobin) which were conflicted by steric restrictions. Subsequent characterization of the GA-SAv-surface using Eu-biotin revealed that a higher input quantity of polymerized streptavidin was required in the coating in order to reach maximum binding capacity. The surfaces coated with polymerized strepta- 59 Summary of the results 60 0 40 80 120 160 200 0 500 000 1 000 000 1 500 000 2 000 000 2 500 000 A cp s Eu-biotin input (pmol/well) 0 40 80 120 160 200 0 500 000 1 000 000 1 500 000 2 000 000 2 500 000 B cp s Eu-biotin input (pmol/well) vidin showed close to a 10-fold increase in the signal of the surface readout measure- ment (Fig. 15 A) compared with the regular surface when assayed using Eu-biotin. The increase in the total binding capacity was close to 5-fold as determined by the solution measurement (Fig. 15 B). The total binding capacity of the GA-SAv-coated well for Eu-biotin yielded 40 pmol/well. Figure 15. The binding capacity comparison of the streptavidin-coated surfaces produced from polymerized streptavidin (GA-SAv) (▲) and unmodified core streptavidin (●). Eu-biotin was used as the reporter molecule. A) Surface readout measurement. B) Solution measurement. Unpublished results. A special feature observed with the surface readout and the solution measurements using each reporter molecule was the larger signal ratio of the surface readout meas- urement between the surfaces made from polymerized and regular streptavidin. This implied the higher density of binding sites on the bottom surface which is beneficial for applications using surface measurement. Another specific feature was the slight non- linearity in the binding isotherms of the high capacity surfaces (both in the regular wells and the spots) with surface measurement. Solution measurement, which inte- grated the signal from the whole area, showed equal binding and linearity within the sub-saturated region. The apparent deviation in binding behavior originated essentially from the uneven distribution of the bound reporter molecules outside the measurement beam in surface measurement. The liquid motion induced by shaking resulted in more efficient binding of the biotinylated reporter molecules outside the illuminated central region. This feature was clearly pronounced below the saturating reporter molecule quantities. 4.5.2 Thiolated streptavidin (III) An alternative means to modify streptavidin for improved adsorption capability and better surface stability was tentatively introduced in one of the studies (III). Reactive thiols were introduced to the protein through amines. This contributed to a partial po- 60 Summary of the results 61 lymerization of streptavidin and tight binding to the surface. An average of 10 - 12 thiols per streptavidin were repeatedly coupled based on the quantification by Ellman’s reaction. The surfaces coated with this thiolated streptavidin yielded 2 - 3-fold higher binding capacities for Eu-biotin compared with the surfaces made from unmodified streptavidin. The leaching of adsorbed streptavidin from this surface was substantially reduced in comparison with several other high-capacity surfaces and the regular strep- tavidin surface. The preparation and optimization of this modified type will be dis- cussed more specifically elsewhere (Ylikotila et al., manuscript in preparation). 4.6 Immunoassays (I, III, IV) 4.6.1 Linearity and the dose-response relationship The high capturing capacity of the surface obviously increased the high-end dynamic range of the TR-IFMA assays and avoided or postponed the high-dose hook effect towards the higher concentrations. This was most clearly observed when using a com- bination of the high-capacity streptavidin surface and the site-specifically biotinylated Fab fragments (I). In the low-end region of the dose-response curves the higher total binding capacity did not guarantee higher signal levels. 4.6.2 Characteristics of various assay types Though the TR-IFMA- and the EIA-assays were in principle similar sandwich-type assays they differed in the reagent incubation sequences. The TR-IFMA-assays em- ployed a separate immobilization of the biotinylated capture antibody prior to the ac- tual assay. The differences between the plates possessing various capacity and desorp- tion features were actually quite small in the linear region of the assay (Fig. 16 A). The IFMA-assay, with separate immobilization of the capture antibody, represented a rela- tively reliable reaction setup in terms of the equilibrium reactions taking place in the well. Furthermore, the preceding immobilization and washing steps evidently elimi- nated some unwanted surface characteristics related to leaching and variation. The EIA-assays (III) using a one-step incubation of the capture antibody and the anti- gen (CA125-assay) or a one-step incubation of all binding partners (PSA-assay) repre- sented more demanding systems with more complicated equilibrium reactions. More substantial differences were observed between the streptavidin-coated surfaces in the linear assay range when using the EIA-assays (Fig. 16 B). This is most probably ex- plained by the different reaction schemes compared with the IFMA-assays and the absence of the prior immobilization and washing steps which would have relieved some of the surface defects. The features of the surfaces observed in the EIA-assays could be to some extent, but not definitely, related to the leaching characteristics of the plates. 61 Summary of the results 62 0 50 100 150 200 250 300 350 0 300 000 600 000 900 000 1 200 000 1 500 000 cp s TSH (mU/l) Bt-Reg. Bt-Mod. InnoSA96 DELFIA-SAv SigmaScreen Reacti-Bind StreptaWell Immobilizer A 0 10 20 30 40 50 60 0 0.500 1.000 1.500 2.000 2.500 3.000 3.500B Bt-Reg. Bt-Mod. InnoSA96 DELFIA-SAv SigmaScreen Reacti-Bind StreptaWell Immobilizer Bg r-c or re ct ed A 61 5 PSA (µg/l) Figure 16. The features of various streptavidin-coated surfaces in the immunoassays. A) An IFMA-type assay of TSH utilizes a separate immobilization of the biotinylated capture antibody prior to the actual assay. B) In the EIA-assay of PSA all reagents are simultaneously mixed in the well. More substantial differences are recognized between the surfaces in the EIA-assay. Modified from original publication III. 4.6.3 Non-specific binding and the sensitivity The background responses (absorbance or fluorescence) measured from the control reactions without an analyte consist of both the material background and the non- specific binding of the labeled reagents. Various inorganic materials exhibit different absorption and autofluorescence features. This is evident even within one material type, for example, polystyrene may have variable fluorescence properties depending on the source and treatment. Therefore, a plain comparison of the total background signals from different surfaces is only partially indicative of the degree of non-specific binding to the surface. Preferably, a number of unused wells should be measured for reference in order to distinguish the background signal of the material from the background caused by the non-specific binding. However, it is the total background response that is routinely employed to determine the specific net responses from the calibrators and the samples and, therefore, the distinction between non-specific binding and material background is typically omitted. This was also practiced in this work (I, III, IV). Various streptavidin-coated surfaces used in this work showed different background responses. The surfaces with the higher binding capacities tended to exhibit higher background levels, but the differences were no more than approximately 2-fold at the highest in comparison with the regular streptavidin surfaces. The analytical sensitivi- ties of the assays (the term sensitivity generally referring to the detection limits in this work), when defined by the method based on the standard deviation of the replicate background readings, are especially vulnerable to variation in background responses. The analytical detection limits of the assays were affected by the streptavidin-coated 62 Summary of the results 63 surfaces; the effect was mainly related to the magnitude and variation of the back- ground readings. 4.6.4 Kinetics (I, IV) The principal definer of the kinetics of an assay is the ratio of the binding surface area to the liquid volume of the reaction (Esser, 1992). The regular microtiter well pre- sented relatively beneficial surface-to-volume ratios so that the higher binding capacity provided by the streptavidin surface did not remarkably improve the assay kinetics, particularly when using intact antibodies for the capture of the antigen. The spot sur- faces introduced new challenges, with highly reduced surface-to-volume ratios. This was also well recognized in assay kinetics. The decreased surface area was compen- sated for by increasing the binding capacity of the streptavidin-coated surface and con- sequently increasing the immobilization capacity of the biotinylated capturing anti- body. The benefit remained marginal when using intact antibodies. A more pronounced effect has been observed by using combinations of high capacity streptavidin surfaces with site-specifically biotinylated fragments of antibodies which evidently exhibit the correct orientation on the surface (Ylikotila et al., 2005). 4.7 Spot-assay (IV) The assays based on the surface readout measurement of time-resolved fluorescence from a washed surface are dependent on the density of the label on a limited surface area defined by the excitation beam. The principle of the spot-assay was to condense the binding of the labeled antibodies into a dense layer under the measuring beam (Fig. 17). The regular binding surfaces manufactured using large volumes of the coat- ing solution have been non-optimal in terms of signal collection due to the spread of labeled detection antibodies outside the measuring beam. In the TR-IFMA-assay of TSH, condensed binding provided a 5 - 6-fold increase in the signal-to-background ratios and an equivalent improvement in the detection limits of the assay (Fig. 18). The positive gain in the signal-to-background ratio was essentially due to the increased response from the actual reactions containing the analyte while the background signals beneficially remained equal to the background levels of the reference assays employing conventionally coated surfaces. 63 Summary of the results 64 Figure 17. A) In the regular assay the binding surface area is large and, consequently, a signifi- cant proportion of the labeled antibody is bound outside the measured area. B) In the spot-assay the binding area is confined to a small spot and the labeled antibodies are captured more densely under the excitation beam. Modified from original publication IV. 0 10 20 30 40 50 60 70 0 100 000 200 000 300 000 400 000 500 000 600 000 A cp s TSH (mU/l) 0.01 0.1 1 10 100 10 100 1 000 10 000 100 000 1 000 000 B cp s TSH (mU/l) 0 20 40 60 80 100 C V % Figure 18. The dose-response curves of the TSH-assay using the reference assay employing a large binding area (■) and the spot-assay utilizing condensed binding (●). Axes in the linear (A) and logarithmic (B) scales. The variation (CV %) of the replicate wells in the reference assay (□) and the spot-assay (○) is shown in graph B. The results are from original publication IV. 64 Discussion 65 5 DISCUSSION There is ongoing active research and development accompanied by extensive literature in the field of the protein coatings. However, most research has reported the properties and performance of surfaces narrowly. Various surface coating methods have been used and reported in a number of studies but when the binding surface itself is not an issue, further evaluation and discussion on the essential features of the surface are omitted. Thereby, it is problematic, or almost impossible, to extract information about the suitability of a given method to the specified needs of other applications. The litera- ture review section of this thesis presented the use of streptavidin as a binding surface in solid-phase assays as well as introduced various advanced coating chemistries re- ported in the literature. The experimental part examined the coating and modifications of streptavidin and studied the effects of the specific properties of streptavidin surfaces to the performance of solid-phase immunoassays. The modeling of the surface and the quantification of adsorbed streptavidin were car- ried out to obtain an estimate of the surface coverage that can be reached by passive adsorption of unmodified streptavidin. One theoretical surface modeling approach, the random sequential adsorption (RSA) (Feder, 1980; Brosilow et al., 1991; Evans, 1993; Loscar et al., 2003) defines the jamming limits for the maximum adsorption of mole- cules into a monolayer. When the densest, well-ordered packing of a particular shape into a monolayer is represented as 100 %, the jamming limits defined by the RSA- model, 54.7 % for spheres and 56.2 % for squares, represent the limits beyond which adsorption into a single layer is no more possible in a random process. Streptavidin from one supplier yielded a 61 - 67 % coverage (II) which was fairly consistent with the RSA model. Streptavidin from the other supplier showed 93 % coverage which implies that the protein is adsorbed in multiple layers or it is organized in an ordered manner. The jamming limit for streptavidin, based on this 54.7 - 56.2 % occupancy and the defined maximum monolayer density of 6.2 pmol/cm2, should be equal to approxi- mately 3.5 pmol/cm2 (≈ 200 ng), which corresponds to about 5.4 pmol (320 ng) in a microtiter well with a coated area of 1.54 cm2. Consequently, assuming that two of the four binding sites of adsorbed streptavidin are available for the binding of biotin, the maximum biotin-binding capacity of the well attainable by plain adsorption should be equal to approximately 11 pmol. Interestingly, the maximum capacity for the binding of Eu-biotin was assayed approximately 10 - 11 pmol/well with the surface prepared by adsorption of unmodified core streptavidin (II). The biotinylated, Eu-labeled reporter molecules provided a direct means to monitor adsorption, determine binding capacities and even to study the localization of the molecules by imaging (IV). The determination of the total binding capacity was based on the quantification of the bound Eu concentration in the saturated region of the bind- ing isotherms against the Eu-calibrators and by proportioning it to the defined labeling molecule binding to streptavidin, where practically 100 % is bound in the linear re- 65 degree of the reporter molecule. Due to the high-affinity interaction (the biotinylated Discussion 66 gion), a graphical determination from the turning points of the binding isotherms would also have been feasible. This method should be, however, considered with cau- tion especially among low affinity interactions, since the turning point of the binding isotherm indicates input concentration where the maximum binding is reached rather than the actual bound quantity. It is also vulnerable to fluctuations in input quantities of the reporter molecule. The Eu concentration based method is dependent on the correct determination of the reporter molecule's degree of labeling. This can be facilitated by a fixed chemical structure as employed in Eu-biotin which provided only one site for the coupling of the chelate. When the degree of labeling is accurately defined, this is, how- ever, a reliable and reproducible method of capacity determination. The polymerization of streptavidin resulted in an up to 5-fold increase in biotin-binding capacity compared with surfaces of unmodified streptavidin when integrated from the entire coated area. The local binding site densities at the bottom surface of the wells were further increased as shown by the surface readout measurements. The fractiona- tion and the gel filtration analysis of the polymerized streptavidin (GA-SAv) revealed a considerable degree of polymerization which could be adjusted by changing the reac- tion conditions. Further polymerization beyond a certain level did not present increased capacities when coated on the surface, but rather a mixture of monomers and polymers of various sizes proved most efficient. Alternative thiol-based modification also re- sulted in increased binding capacities, but above all, the stability of the surface against leaching was improved. The studies with modified streptavidin preparations showed that relatively straightforward, cost-effective methods provide substantial improvement in the binding capacities and surface densities which can be applied first before pro- ceeding to more complicated techniques. The higher immobilization and the capturing capacities were helpful in avoiding or postponing the high-dose hook effect of the immunoassays. This was clearly observed when using a combination of the high-capacity streptavidin surface with the Fab frag- ments for the capture of the antigen. The model assays employed in the tests showed the linearity advantage mainly beyond regions of clinical relevance, but nevertheless, they showed the importance of the capturing capacity to guarantee the linearity of the assay, implying favorable effects on assays that are indeed susceptible to the high-dose hook effect. The binding capacity was not similarly relevant to assay sensitivity (detec- tion limits) and low-end performance. In a reagent-excess assay, only a small fraction of the binding sites are occupied by the antigen in the dilute region and the streptavidin surface of a moderate binding capacity binds a sufficient amount of antibodies to serve a satisfactory capture of the antigen, at least in the 96-well plate format. In the low end and the linear regions of the assays other determinants such as leaching, variation, sur- face homogeneity and non-specific binding were pronounced. The initially defined properties of the surfaces such as desorption and the homogeneity of the binding ca- pacities were not, by definition, predictive of the surface's performance in the immuno- assays. Some relation could be observed between the desorption degree and the re- sponses from the EIA-type assays which utilized simultaneous binding of the bioti- 66 Discussion 67 nylated capture antibody and the antigen. Variation of the replicate wells tended to increase in the plates exhibiting high desorption, most likely due to carryover by the washing device, and generally lower responses were measured because of the consid- erable leaching of the captured immunocomplexes. Desorption of the adsorbed protein and its correlation with immunoassay performance has occasionally been an issue of speculation. Adsorption of the proteins, the technique employed for the preparation of the most surfaces, is generally considered an irreversi- ble process, though some replacement has been regularly observed especially in the presence of replacing proteins (see section 1.3.3.1). The immunoassay buffer utilized in the desorption studies contained proteins, but at the same time it contained deter- gents which efficiently prevent adsorption and, therefore, the replacement was not obvious. The desorption observed can be considered a new process, independent of the preceding adsorption, where the rinsing and shearing forces introduced by the mobile liquid played a major role. The assays using a preceding immobilization and washing procedure did not seem particularly vulnerable to desorption. Weakly bound strepta- vidin was dissociated from the surface during the immobilization step and the remain- ing capture layer showed stable features. The assays that avoided the preceding immo- bilization step evidently suffered more from the instable surfaces, as shown by the EIA-type assays. An interesting observation made during this research (but which was not comprehensively studied) was the increased desorption of adsorbed streptavidin during storage. Freshly prepared surfaces clearly exhibited lower quantities of de- sorbed protein than those stored for months or years at +4 °C. This obstructed their use as reference wells in desorption studies. Increased desorption is likely indicative of the rearrangement and lateral diffusion processes taking place among adsorbed proteins, as reviewed for example by Dent and Aslam (1998). The recombinant Fab fragments of the antibodies presented superior antigen-binding performance in comparison with intact, whole-size antibodies. The higher capturing capacity of the Fab surfaces is brought about by two reasons; first the smaller size of the fragments and thereby the higher density of the antigen binding sites on the surface and second, evidently correct orientation of the fragments due to site-specific biotinyla- tion. The specific share of one or the other to binding performance was not further studied. It seems that the combinations of tailored antibody fragments with highly dense streptavidin surfaces are proving a promising binding surface option in the mi- cro- and nano-scale assay concepts where rapid results from small sample quantities are expected. The immunoassays based on the surface readout measurement of intrinsically fluores- cent lanthanide labels from a washed surface have suffered from the distribution of the immunocomplexes and the labeled detection antibodies outside the measuring beam. The purpose of the spot approach was to improve the sensitivity of these assays through condensed collection of the labeled antibodies to the surface and consequent increase in the signal-to-background ratios. Unmodified native streptavidin (SAv) and 67 Discussion 68 polymerized high-capacity streptavidin (GA-SAv) were successfully coated to the spe- cial spotted microtiter wells to serve the binding of the capture antibodies. The signifi- cantly higher biotin-binding capacities and the binding site densities obtained with the GA-SAv-coated spots were not completely beneficial in the model immunoassay of TSH. This may be partially due to the uneven distribution recognized in the binding which was more pronounced on the GA-SA-surface. Though incomplete in the sense of the measuring parameters, the increased concentration and density of the labeled antibodies under the measuring beam resulted in a significant increase in the signal-to- background ratios and improvement in the analytical detection limits. Interestingly, the central property of the microspot-based ambient analyte assay (Ekins et al., 1989; Ekins and Chu, 1991; Ekins, 1998), that is to say the increased label density in the measured area conferring to a more sensitive assay, was also recognized in the spot- assay of this work. The research methods used in the studies of this thesis provided information on surface binding capacities, binding site densities, surface stability and the effects of the binding surface on the solid-phase immunoassay. They provided little information on the other relevant surface features such as topography, layer thickness, orientation and homoge- neity (except the imaging, IV) and thereby these determinants remained mostly specu- lative. This raises interest to proceed to the next level, to relate the contribution of the specific surface properties defined by various surface science studies to the features of solid-phase assays. This approach would, for example, disclose the actual organization of variously coated streptavidin surfaces as well as the orientation and the densities of the capture antibodies and the Fab fragments on the surface. Several potential surface research methods and technologies exist which would provide further insights into the capturing surfaces. Scanning probe microscopy (SPM) is a common designation of surface imaging techniques that are based on the scanning of the specimen by a probe. Among these, atomic force microscopy (AFM) and scanning tunneling microscopy (STM) have been used for studying biomolecular layers (Leggett et al., 1996). In AFM, the probe is a sharp tip which moves over the surface and sensitively responds to the forces encountered at the interface, thus creating a three-dimensional image of the surface topography even in a sub-nanometer resolution (Davies et al., 1994b; Moll et al., 2002; Browne et al., 2004; Kim J. et al., 2004). In addition, AFM has been used to quantify the interactive forces between molecules on the interface, for example, be- tween streptavidin and biotin using a biotin-coated probe (Allen et al., 1996; Wong et al., 1999). Scanning tunneling microscopy (STM) is based on the current of electrodes (tunneling) through a small gap between an atomically sharp conducting probe and a conducting surface. While based on the electronic current, the support must be con- ducting or the surface must be overlaid with a metal layer. STM has been used to visu- alize adsorbed streptavidin (Cooper et al., 1994) and antigen-antibody interactions on the surface (Davies et al., 1994a; 1994b). Scanning electron microscopy (SEM) has also been used for studying the protein-coated solid phases (Butler et al., 1992; Assis, 2003). 68 Discussion 69 Among the optical measuring methods, ellipsometry is an efficient method to assay layer thicknesses ranging from less than a nanometer to several micrometers. When polarized light is directed to, and reflected from a sample surface, its polarization state changes as a function of the surface thickness and the refractive index. As a quite well- established and versatile method, ellipsometry and variations of it are widely used for studying biomolecular surfaces (Malmsten, 1995; Spaeth et al., 1997; Elwing, 1998; Tengvall et al., 1998; Bae et al., 2005). Another well established optical technique, surface plasmon resonance (SPR) has definitely gained popularity in studying protein layers and biomolecular interactions (Hodneland et al., 2002; Cui et al., 2003; Homola, 2003; Snopok and Kostyukevich, 2006). Of the reflection based methods, neutron re- flection has recently been shown to be a feasible technique for studying protein sur- faces (Su et al., 1998; Xu et al., 2006; Lu et al., 2007; Xu et al., 2007). Spectroscopic methods such as X-ray photoelectron spectroscopy (XPS) (Ruiz-Taylor et al., 2001a; Pradier et al., 2002b; Browne et al., 2004) and infrared reflection-absorption spectros- copy (IRRAS) (Mendelsohn and Flach, 2002; Pradier et al., 2002a; Liu and Amiridis, 2004; 2005) provide insights essentially into the chemical constitution of adsorbed layers. Quartz crystal microbalance (QCM) is a technique based on a piezoelectric quartz crystal the oscillation frequency of which changes upon a mass deposited on the crystal. The change in frequency can be quantitatively related to the applied mass and thus QCM serves as a sensitive method for quantifying adsorbed material (Rickert et al., 1997; Linder et al., 2002; Kim J. et al., 2004; Zhang et al., 2005; Shu et al., 2007). Most of the surface research techniques referred to require specific solid substrates or sample cells, such as conductive or reflective beds, and thus are not readily applicable to all desired settings, such as to polystyrene microtiter wells or array slides. They can serve, above all, as tools for studying the organization of surfaces and molecular inter- actions in environments which mimic the actual molecular level situation, as shown, for example, by using a polystyrene-coated SPR-chip for the study of adsorption and binding of streptavidin and antibodies (Davies et al., 1994b). Direct measurement from the microtiter wells is not, however, completely out of the question. Scanning electron microscopy images have been acquired from intact microtiter wells (Butler et al., 1992) and other, perhaps the most closely related setup, utilized coated wells from which the walls were excised and the bottom discs were subjected to AFM and STM analyses (Davies et al., 1994a; 1994b; Allen et al., 1996). 69 Conclusions 70 6 CONCLUSIONS The immobilization of proteins using streptavidin-biotin technology constitutes an advantage compared with direct adsorption in terms of preserving molecular activity and natural conformation. The binding surface is a significant determinant of solid- phase assays and therefore the specific features of streptavidin-coated surfaces relevant to solid-phase immunoassays were subjected to a comprehensive evaluation in this thesis. The work was accompanied by the establishment of a number of surface charac- terization tools and methods. The binding site densities of the streptavidin-coated mi- crotiter wells were improved by close to 10-fold and the leaching of the adsorbed pro- tein was decreased using chemically modified streptavidin. Furthermore, the sensitivity and the detection limits of the immunoassays based on the surface readout measure- ment of time-resolved fluorescence were improved through the condensed binding of the labeled antibodies enabled by the spotted streptavidin surfaces. The properties of the streptavidin surfaces significant to immunoassay performance were found to be the surface stability, the degree of non-specific binding and homogeneity. The increased binding capacity enabled wider dynamic ranges of the assays, while it was less relevant to assay sensitivity and low-end performance at least in the regular microtiter well format. The driving forces of several surface studies are, however, the high binding capacity and coating densities which are admittedly relevant issues in certain technolo- gies. Some central characteristics of the future assay concepts are high sensitivity and multi- plex assays as well as rapidity combined with simplicity to serve the instant point-of- care assays (some of the trends recently reviewed by Wu, 2006). Nanotechnologies such as nanoparticles and array-based approaches have an important role when work- ing on these objectives. In addition, the development in label technologies has opened up new opportunities for separation-free homogeneous assays (Kokko et al., 2004; Hemmilä and Laitala, 2005; Kuningas et al., 2005), especially in high-throughput screening where assay simplicity is valued, and gradually in clinical diagnostics as the sensitivities of homogeneous assays reach those of heterogeneous assays. The interac- tions between proteins and solid phases will still prevail, even in several homogeneous assays, where the biomolecular interactions after all take place at a solid-liquid inter- face such as on the nanoparticle surface. With respect to the array-type assays using spotted surfaces, Ekins and others have concluded that microarray assays should obey the rules of the ambient analyte assay (Ekins, 1999; Saviranta et al., 2004), an assay format that demands highly dense and homogeneous surfaces. Given these facts, there is no doubt that the features of the binding surface will have a dominant and even in- creasing influence on assay performance in the new forthcoming assay concepts. 70 Acknowledgements 71 7 ACKNOWLEDGEMENTS This work was carried out at the University of Turku, in the laboratory of Biotechnol- ogy at the Department of Biochemistry and Food Chemistry between 1999 and 2006. The work was financially supported by Innotrac Diagnostics Oy and the Finnish Fund- ing Agency for Technology and Innovation (Tekes). Part of the work was done as col- laboration between the University of Turku, Innotrac Diagnostics Oy and Nunc A/S under the EUREKA-status (project E!2217). I warmly thank my supervisor, Professor Timo Lövgren, PhD and the head of the labo- ratory, for giving me the opportunity to do the thesis work in the innovative and ambi- tious atmosphere of the Department of Biotechnology. His initial encouragement and ongoing support during the years have been valuable carrying forces in completing this work. I am deeply grateful to my main supervisor, Professor Kim Pettersson, PhD, for his continuous support and guidance to the right destinations in the world of the immu- noassays. His knowledge and skills in immunoassays and clinical diagnostics has al- lowed me to carry out this work under excellent supervision. I am grateful to Docent Markku Viander and PhD Petri Ihalainen for reviewing this thesis and giving valuable feedback and comments. I would like to thank Dr. Mike Nelson for revising the linguistic form of the thesis. I warmly thank the other co-authors of the original publications, Matti Karp, Markus Vehniäinen, Jaana Rosenberg, Katja Laurikainen, Johanna Ylikotila, Hannu Kojola, Tero Soukka and Harri Takalo, for their excellent contribution to the particular works. Special thanks to Johanna and Katja for the completion of the main laboratory work for two original publications. Innotrac Diagnostics Oy is thanked for the support and the long-term collaborative research with the University in the interesting "solid-phase" discipline. The employees of the company, Dr. Harri Takalo, the director of the research and development, Jarmo Rainaho, Annukka Mäki and all the others are thanked for the straightforward co- operation and practical organization of numerous issues and materials. I would like to thank Svend Erik Rasmussen, Lena Brandt Larsen and Marianne Pedersen from Nunc A/S, Denmark, for co-operation and assistance in planning and manufacturing the microtiter surfaces. I have learned a lot about surface sciences as well as the prepa- ration and properties of the medical plastics with them. I want to thank all the employees, collaborative scientists and students (not yet men- tioned) of the prior "solid phases" project. Tiina Kokko, Elina Harittu, Susann Eriks- son, Qiu-Ping Qin, Jukka Hellman, Eeva-Christine Brockmann, Johanna Hellström, Henna Heikonen, Antti Varho, Nicolas Isla, Hannu Leino and Heidi Virtanen are thanked for their contribution to the discipline. In addition, Johanna Ylikotila and Mari Peltola are thanked for the reading of a draft of this thesis and giving constructive 71 Acknowledgements 72 feedback. Jaana Rosenberg is thanked for assistance in drawing the pictures of biotins in this thesis. Furthermore, I want to thank the co-workers in my current project and facilities. Etvi Juntunen, Katja Niemelä, Maria Rissanen, Riina-Minna Väänänen, Juuso Huusko, Erik Jokisalo, Noelia Lozano Vidal, Pauliina Helo and Leni Mannermaa are thanked particularly for their kindness and understanding when writing the thesis. All the personnel at the Department of Biotechnology, past and present, are warmly thanked for creating an enjoyable, pleasant atmosphere to work. The administrative and the laboratory personnel especially are greatly acknowledged. Marja-Liisa Knuuti is thanked for assistance in administrative issues and Mirja Jaala, Pirjo Laaksonen, Martti Sointusalo, Marja Maula, Sari Lindgren, Veikko Wahlroos and Pirjo Pietilä are particularly thanked for organizing materials and reagents as well as for all other tech- nical support. My former office mates, Dr. Mika Tuomola and Markus Vehniäinen, are thanked for the delightful discussions and sharing of experiences in science. Dr. Ville Väisänen, a co-worker and a friend, is thanked for the valuable tips during the finaliza- tion of the thesis and for regular organization of lunch hour company within the work- ing days. I wish to thank all my friends. With the lasting friendships - some of them from child- hood - they have played an important role in my life enabling an invaluable balance to work. In addition, my time and responsibilities in the Scouts provided me with impor- tant relationships as well as several unforgettable training experiences at the mercy of nature and weather. I have had the privilege to live my youth in a warm, loving family. This is due to my parents, sister, deceased grandparents and all the other relatives and family friends around me. I am deeply grateful to my parents, Terttu and Jukka, for their enormous love and support. I warmly thank my younger sister, Marjut, especially for believing in my skills - she once said she is willing to see a doctor's hat on my head. It is probably close to happening soon. I am also very grateful to Leena's family for their kindness and support to me and our family. Finally, I give my warmest thanks to my wife Leena for the lasting love and support - and definitely, taking responsibility of the household particularly during the final stage of this work. I am also very grateful to our children Anni, Oskari and Elisa. You have shown me the real values of life, and the joy and fun that I feel when playing with you and the smiles on your faces is something beyond description. Lieto, May 2008 72 References 73 8 REFERENCES Allen, S., Davies, J., Dawkes, A.C., Davies, M.C., Edwards, J.C., Parker, M.C., Roberts, C.J., Sefton, J., Tendler, S.J. and Williams, P.M. (1996) In situ observation of streptavidin-biotin binding on an immunoassay well surface using an atomic force microscope. FEBS Lett. 390: 161-164. Anderson, G.P., Jacoby, M.A., Ligler, F.S. and King, K.D. (1997) Effectiveness of protein A for antibody immobilization for a fiber optic biosensor. Biosens. Bioelectron. 12: 329-336. Andrade, J.D. and Hlady, V. (1986) Protein adsorption and materials biocompatibility: a tutorial review and suggested hypotheses. Adv. Polym. Sci. 79: 1-63. Argaraña, C.E., Kuntz, I.D., Birken, S., Axel, R. and Cantor, C.R. (1986) Molecular cloning and nucleotide sequence of the streptavidin gene. Nucleic Acids Res. 14: 1871-1882. Assis, O.B. (2003) Scanning electron microscopy study of protein immobilized on SiO2 sol-gel surfaces. Braz. J. Chem. Eng. 20: 339-342. Bae, Y.M., Oh, B.K., Lee, W., Lee, W.H. and Choi, J.W. (2005) Study on orientation of immunoglobulin G on protein G layer. Biosens. Bioelectron. 21: 103-110. Bagchi, P. and Birnbaum, S.M. (1981) Effect of pH on the adsorption of immunoglobulin G on anionic poly(vinyltoluene) model latex particles. J. Colloid Interface Sci. 83: 460-478. Bayer, E.A., Zalis, M.G. and Wilchek, M. (1985) 3-(N-Maleimido-propionyl)biocytin: a versatile thiol- specific biotinylating reagent. Anal. Biochem. 149: 529-536. Bayer, E.A., Ben-Hur, H., Gitlin, G. and Wilchek, M. (1986) An improved method for the single-step purification of streptavidin. J. Biochem. Biophys. Methods 13: 103-112. Bayer, E.A., Ben-Hur, H., Hiller, Y. and Wilchek, M. (1989) Postsecretory modifications of streptavidin. Biochem. J. 259: 369-376. Bayer, E.A., Ben-Hur, H. and Wilchek, M. (1990) Isolation and properties of streptavidin. Methods Enzymol. 184: 80-89. Bayer, E.A. and Wilchek, M. (1990) Protein biotinylation. Methods Enzymol. 184: 138-160. Becker, J.M., Wilchek, M. and Katchalski, E. (1971) Irreversible inhibition of biotin transport in yeast by biotinyl-p-nitrophenyl ester. Proc. Natl. Acad. Sci. U. S. A. 68: 2604-2607. Benters, R., Niemeyer, C.M. and Wöhrle, D. (2001) Dendrimer-activated solid supports for nucleic acid and protein microarrays. ChemBioChem 2: 686-694. Berger, M., Deger, A. and Maier, J. (1993) Preparation of a solid phase matrix containing a bound specific binding pair. US Patent No. 5268306. Berson, S.A., Yalow, R.S., Bauman, A., Rothschild, M.A. and Newerly, K. (1956) Insulin-I131 metabolism in human subjects: demonstration of insulin binding globulin in the circulation of insulin treated subjects. J. Clin. Invest. 35: 170-190. Berson, S.A. and Yalow, R.S. (1958) Isotopic tracers in the study of diabetes. Adv. Biol. Med. Phys. 6: 349-430. Bicknell, G.R., Shaw, J.A., Pringle, J.H. and Furness, P.N. (1996) Amplification of specific mRNA from a single human renal glomerulus, with an approach to the separation of epithelial cell mRNA. J. Pathol. 180: 188-193. Bjarnason, B., Chimuka, L., Önnerfjord, P., Eremin, S., Jönsson, J.A., Johansson, G. and Emnéus, J. (2001) Enzyme flow immunoassay using a protein G column for the screening of triazine herbicides in surface and waste water. Anal. Chim. Acta 426: 197-207. Björck, L. and Kronvall, G. (1984) Purification and some properties of streptococcal protein G, a novel IgG-binding reagent. J. Immunol. 133: 969-974. Blawas, A.S., Oliver, T.F., Pirrung, M.C. and Reichert, W.M. (1998) Step-and-repeat photopatterning of protein features using caged-biotin-BSA: Characterization and resolution. Langmuir 14: 4243- 4250. 73 References 74 Boas, M.A. (1924) An observation on the value of egg-white as the sole source of nitrogen for young growing rats. Biochem. J. 18: 422-424. Boas, M.A. (1927) The effect of desiccation upon the nutritive properties of egg-white. Biochem. J. 21: 712-724. Bonroy, K., Frederix, F., Reekmans, G., Dewolf, E., De Palma, R., Borghs, G., Declerck, P. and Goddeeris, B. (2006) Comparison of random and oriented immobilisation of antibody fragments on mixed self-assembled monolayers. J. Immunol. Methods 312: 167-181. Bora, U., Chugh, L. and Nahar, P. (2002) Covalent immobilization of proteins onto photoactivated polystyrene microtiter plates for enzyme-linked immunosorbent assay procedures. J. Immunol. Methods 268: 171-177. Bora, U., Kannan, K. and Nahar, P. (2004) Heat-mediated enzyme-linked immunosorbent assay procedure on a photoactivated surface. J. Immunol. Methods 293: 43-50. Bora, U., Sharma, P., Kumar, S., Kannan, K. and Nahar, P. (2006) Photochemical activation of a polycarbonate surface for covalent immobilization of a protein ligand. Talanta 70: 624-629. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248-254. Brosilow, B.J., Ziff, R.M. and Vigil, R.D. (1991) Random sequential adsorption of parallel squares. Phys. Rev. A 43: 631-638. Browne, M.M., Lubarsky, G.V., Davidson, M.R. and Bradley, R.H. (2004) Protein adsorption onto polystyrene surfaces studied by XPS and AFM. Surf. Sci. 553: 155-167. Buckie, J.W. and Cook, G.M. (1986) Specific isolation of surface glycoproteins from intact cells by biotinylated concanavalin A and immobilized streptavidin. Anal. Biochem. 156: 463-472. Bugari, G., Poiesi, C., Beretta, A., Ghielmi, S. and Albertini, A. (1990) Quantitative immunoenzymatic assay of human lutropin, with use of a bi-specific monoclonal antibody. Clin. Chem. 36: 47-52. Buijs, J., Lichtenbelt, J.W.T., Norde, W. and Lyklema, J. (1995) Adsorption of monoclonal IgGs and their F(ab')2 fragments onto polymeric surfaces. Colloids Surf. B Biointerfaces 5: 11-23. Bull, H.B. (1956) Adsorption of bovine serum albumin on glass. Biochim. Biophys. Acta 19: 464-471. Bulmus, V., Ding, Z., Long, C.J., Stayton, P.S. and Hoffman, A.S. (2000) Site-specific polymer- streptavidin bioconjugate for pH-controlled binding and triggered release of biotin. Bioconjugate Chem. 11: 78-83. Butler, J.E., Spradling, J.E., Suter, M., Dierks, S.E., Heyermann, H. and Peterman, J.H. (1986) The immunochemistry of sandwich ELISAs–I. The binding characteristics of immunoglobulins to monoclonal and polyclonal capture antibodies adsorbed on plastic and their detection by symmetrical and asymmetrical antibody-enzyme conjugates. Mol. Immunol. 23: 971-982. Butler, J.E., Ni, L., Nessler, R., Joshi, K.S., Suter, M., Rosenberg, B., Chang, J., Brown, W.R. and Cantarero, L.A. (1992) The physical and functional behavior of capture antibodies adsorbed on polystyrene. J. Immunol. Methods 150: 77-90. Butler, J.E., Ni, L., Brown, W.R., Joshi, K.S., Chang, J., Rosenberg, B. and Voss, E.W. Jr. (1993) The immunochemistry of sandwich ELISAs–VI. Greater than 90% of monoclonal and 75% of polyclonal anti-fluorescyl capture antibodies (CAbs) are denatured by passive adsorption. Mol. Immunol. 30: 1165-1175. Butler, J.E., Lu, E.P., Navarro, P. and Christiansen, B. (1997a) Comparative studies on the interaction of proteins with a polydimethylsiloxane elastomer. I. Monolayer protein capture capacity (PCC) as a function of protein pI, buffer pH and buffer ionic strength. J. Mol. Recognit. 10: 36-51. Butler, J.E., Navarro, P. and Lu, E.P. (1997b) Comparative studies on the interaction of proteins with a polydimethylsiloxane elastomer. II. The comparative antigenicity of primary and secondarily adsorbed IgG1 and IgG2a and their non-adsorbed counterparts. J. Mol. Recognit. 10: 52-62. immunoassays. Methods 22: 4-23. 74 Butler, J.E. (2000) Solid supports in enzyme-linked immunosorbent assay and other solid-phase References 75 Cantarero, L.A., Butler, J.E. and Osborne, J.W. (1980) The adsorptive characteristics of proteins for polystyrene and their significance in solid-phase imunoassays. Anal. Biochem. 105: 375-382. Carlsson, J., Mosbach, K. and Bülow, L. (1996) Affinity precipitation and site-specific immobilization of proteins carrying polyhistidine tails. Biotechnol. Bioeng. 51: 221-228. Catt, K., Niall, H.D. and Tregear, G.W. (1966) Solid-phase radioimmunoassay of human growth hormone. Biochem. J. 100: 31c-33c. Catt, K.J., Niall, H.D. and Tregear, G.W. (1967) A disc method for solid-phase radioimmunoassay of proteins. Aust. J. Exp. Biol. Med. Sci. 45: 703-706. Catt, K. and Tregear, G.W. (1967) Solid-phase radioimmunoassay in antibody-coated tubes. Science 158: 1570-1572. Chaiet, L., Miller, T.W., Tausig, F. and Wolf, F.J. (1963) Antibiotic Msd-235. II. Separation and purification of synergistic components. Antimicrob. Agents Chemother. 161: 28-32. Chaiet, L. and Wolf, F.J. (1964) The Properties of streptavidin, a biotin-binding protein produced by streptomycetes. Arch. Biochem. Biophys. 106: 1-5. Chen, H.M. and Lin, C.W. (2007) Hydrogel-coated streptavidin piezoelectric biosensors and applications to selective detection of Strep-tag displaying cells. Biotechnol. Prog. 23: 741-748. Chilkoti, A., Boland, T., Ratner, B.D. and Stayton, P.S. (1995a) The relationship between ligand-binding thermodynamics and protein-ligand interaction forces measured by atomic force microscopy. Biophys. J. 69: 2125-2130. Chilkoti, A. and Stayton, P.S. (1995) Molecular origins of the slow streptavidin-biotin dissociation kinetics. J. Am. Chem. Soc. 117: 10622-10628. Chilkoti, A., Tan, P.H. and Stayton, P.S. (1995b) Site-directed mutagenesis studies of the high-affinity streptavidin-biotin complex: contributions of tryptophan residues 79, 108, and 120. Proc. Natl. Acad. Sci. U. S. A. 92: 1754-1758. Cho, I.H., Paek, E.H., Lee, H., Kang, J.Y., Kim, T.S. and Paek, S.H. (2007) Site-directed biotinylation of antibodies for controlled immobilization on solid surfaces. Anal. Biochem. 365: 14-23. Choi, H.J., Kim, N.H., Chung, B.H. and Seong, G.H. (2005) Micropatterning of biomolecules on glass surfaces modified with various functional groups using photoactivatable biotin. Anal. Biochem. 347: 60-66. Chu, V., Freitag, S., Le Trong, I., Stenkamp, R.E. and Stayton, P.S. (1998) Thermodynamic and structural consequences of flexible loop deletion by circular permutation in the streptavidin-biotin system. Protein Sci. 7: 848-859. Cooper, J.M., Shen, J., Young, F.M., Connolly, P., Barker, J.R. and Moores, G. (1994) The imaging of streptavidin and avidin using scanning tunneling microscopy. J. Mater. Sci.-Mater. Electron. 5: 106-110. Costello, S.M., Felix, R.T. and Giese, R.W. (1979) Enhancement of immune cellular agglutination by use of an avidin-biotin system. Clin. Chem. 25: 1572-1580. Cui, X., Pei, R., Wang, Z., Yang, F., Ma, Y., Dong, S. and Yang, X. (2003) Layer-by-layer assembly of multilayer films composed of avidin and biotin-labeled antibody for immunosensing. Biosens. Bioelectron. 18: 59-67. Dapron, J., Kappel, W.K. and Li, H. (2005) High capacity assay platforms. US Patent Application Publication No. US 2005/0118709 A1. Davies, J., Dawkes, A.C., Haymes, A.G., Roberts, C.J., Sunderland, R.F., Wilkins, M.J., Davies, M.C., Tendler, S.J., Jackson, D.E. and Edwards, J.C. (1994a) A scanning tunnelling microscopy comparison of passive antibody adsorption and biotinylated antibody linkage to streptavidin on microtiter wells. J. Immunol. Methods 167: 263-269. Davies, J., Roberts, C.J., Dawkes, A.C., Sefton, J., Edwards, J.C., Glasbey, T.O., Haymes, A.G., Davies, M.C., Jackson, D.E., Lomas, M., Shakesheff, K.M., Tendler, S.J.B., Wilkins, M.J. and Williams, P.M. (1994b) Use of scanning probe microscopy and surface plasmon resonance as analytical tools in the study of antibody-coated microtiter wells. Langmuir 10: 2654-2661. 75 References 76 DeLange, R.J. and Huang, T.S. (1971) Egg white avidin. III. Sequence of the 78-residue middle cyanogen bromide peptide. Complete amino acid sequence of the protein subunit. J. Biol. Chem. 246: 698- 709. Dent, A.H. and Aslam, M. (1998) Physisorption of proteins to solid phases. In: Aslam, M. and Dent, A. (Eds), Bioconjugation. Protein Coupling Techniques for the Biomedical Sciences. Macmillan Reference Ltd, London, p. 506-531. Desai, S., Rickerd, M. and Horneij, I. (2003) Coated surfaces with high capacity for capturing target molecules. US Patent No. US 6638728 B1. Diamandis, E.P. and Christopoulos, T.K. (1991) The biotin-(strept)avidin system: principles and applications in biotechnology. Clin. Chem. 37: 625-636. Ding, Z., Long, C.J., Hayashi, Y., Bulmus, E.V., Hoffman, A.S. and Stayton, P.S. (1999) Temperature specific conjugate. Bioconjugate Chem. 10: 395-400. Ding, Z., Fong, R.B., Long, C.J., Stayton, P.S. and Hoffman, A.S. (2001) Size-dependent control of the binding of biotinylated proteins to streptavidin using a polymer shield. Nature 411: 59-62. Djavadi-Ohaniance, L., Friguet, B. and Goldberg, M.E. (1984) Structural and functional influence of enzyme-antibody interactions: effects of eight different monoclonal antibodies on the enzymatic activity of Escherichia coli tryptophan synthase. Biochemistry 23: 97-104. Domen, P.L., Nevens, J.R., Mallia, A.K., Hermanson, G.T. and Klenk, D.C. (1990) Site-directed immobilization of proteins. J. Chromatogr. 510: 293-302. Dontha, N., Nowall, W.B. and Kuhr, W.G. (1997) Generation of biotin/avidin/enzyme nanostructures with maskless photolithography. Anal. Chem. 69: 2619-2625. Douglas, A.S. and Monteith, C.A. (1994) Improvements to immunoassays by use of covalent binding assay plates. Clin. Chem. 40: 1833-1837. du Vigneaud, V. (1942) Structure of biotin. Science 96: 454-461. Dudin, G., Steegmayer, E.W., Vogt, P., Schnitzer, H., Diaz, E., Howell, K.E., Cremer, T. and Cremer, C. (1988) Sorting of chromosomes by magnetic separation. Hum. Genet. 80: 111-116. Duncan, R.J., Weston, P.D. and Wrigglesworth, R. (1983) A new reagent which may be used to introduce sulfhydryl groups into proteins, and its use in the preparation of conjugates for immunoassay. Anal. Biochem. 132: 68-73. Eakin, R.E., McKinley, W.A. and Williams, R.J. (1940a) Egg-white injury in chicks and its relation to a deficiency of vitamin H (biotin). Science 92: 224-225. Eakin, R.E., Snell, E.E. and Williams, R.J. (1940b) A constituent of raw egg white capable of inactivating biotin in vitro. J. Biol. Chem. 136: 801-802. Eakin, R.E., Snell, E.E. and Williams, R.J. (1941) The concentration and assay of avidin, the injury- producing protein in raw egg white. J. Biol. Chem. 140: 535-543. Eisen, M.B. and Brown, P.O. (1999) DNA arrays for analysis of gene expression. Methods Enzymol. 303: 179-205. Ekins, R.P. (1960) The estimation of thyroxine in human plasma by an electrophoretic technique. Clin. Chim. Acta 5: 453-459. Ekins, R., Chu, F. and Biggart, E. (1989) Development of microspot multianalyte ratiometric immunoassay using dual fluorescent-labelled antibodies. Anal. Chim. Acta 227: 73-96. Ekins, R.P. and Chu, F.W. (1991) Multianalyte microspot immunoassay--microanalytical "compact disk" of the future. Clin. Chem. 37: 1955-1967. Ekins, R.P. (1998) Ligand assays: from electrophoresis to miniaturized microarrays. Clin. Chem. 44: 2015-2030. Ekins, R.P. (1999) Immunoassay, DNA analysis, and other ligand binding assay techniques: From electropherograms to multiplexed, ultrasensitive microarrays on a chip. J. Chem. Educ. 76: 769- 780. 76 control of biotin binding and release with a streptavidin-poly(N-isopropylacrylamide) site- References 77 Eliasson, M., Olsson, A., Palmcrantz, E., Wiberg, K., Inganäs, M., Guss, B., Lindberg, M. and Uhlén, M. (1988) Chimeric IgG-binding receptors engineered from staphylococcal protein A and streptococcal protein G. J. Biol. Chem. 263: 4323-4327. Eliasson, M., Andersson, R., Olsson, A., Wigzell, H. and Uhlén, M. (1989) Differential IgG-binding characteristics of staphylococcal protein A, streptococcal protein G, and a chimeric protein AG. J. Immunol. 142: 575-581. Elwing, H. (1998) Protein absorption and ellipsometry in biomaterial research. Biomaterials 19: 397-406. Eriksson, S., Vehniäinen, M., Jansén, T., Meretoja, V., Saviranta, P., Pettersson, K. and Lövgren, T. (2000) Dual-label time-resolved immunofluorometric assay of free and total prostate-specific antigen based on recombinant Fab fragments. Clin. Chem. 46: 658-666. Esser, P. (1992) The surface/volume ratio in solid phase assays. Nunc Bulletin No. 10. Nunc A/S, Roskilde, Denmark. Esser, P. (1997a) Adsorption geometry in Nunc products for solid phase assays. Nunc Bulletin No. 1. Second edition. Nunc A/S, Roskilde, Denmark. Esser, P. (1997b) Principles in adsorption to polystyrene. Nunc Bulletin No. 6. Second edition. Nunc A/S, Roskilde, Denmark. Evans, J.W. (1993) Random and cooperative sequential adsorption. Rev. Mod. Phys. 65: 1281-1329. Feder, J. (1980) Random sequential adsorption. J. Theor. Biol. 87: 237-254. Feder, J. and Giaever, I. (1980) Adsorption of ferritin. J. Colloid Interface Sci. 78: 144-154. Forsgren, A. and Sjöquist, J. (1966) "Protein A" from S. aureus. I. Pseudo-immune reaction with human gamma-globulin. J. Immunol. 97: 822-827. Forster, A.C., McInnes, J.L., Skingle, D.C. and Symons, R.H. (1985) Non-radioactive hybridization probes prepared by the chemical labelling of DNA and RNA with a novel reagent, photobiotin. Nucleic Acids Res. 13: 745-761. Freitag, S., Le Trong, I., Klumb, L., Stayton, P.S. and Stenkamp, R.E. (1997) Structural studies of the streptavidin binding loop. Protein Sci. 6: 1157-1166. Freitag, S., Le Trong, I., Chilkoti, A., Klumb, L.A., Stayton, P.S. and Stenkamp, R.E. (1998) Structural studies of binding site tryptophan mutants in the high-affinity streptavidin-biotin complex. J. Mol. Biol. 279: 211-221. Freitag, S., Chu, V., Penzotti, J.E., Klumb, L.A., To, R., Hyre, D., Le Trong, I., Lybrand, T.P., Stenkamp, R.E. and Stayton, P.S. (1999a) A structural snapshot of an intermediate on the streptavidin- biotin dissociation pathway. Proc. Natl. Acad. Sci. U. S. A. 96: 8384-8389. Freitag, S., Le Trong, I., Klumb, L.A., Chu, V., Chilkoti, A., Stayton, P.S. and Stenkamp, R.E. (1999b) X- ray crystallographic studies of streptavidin mutants binding to biotin. Biomol. Eng. 16: 13-19. Friguet, B., Djavadi-Ohaniance, L. and Goldberg, M.E. (1984) Some monoclonal antibodies raised with a native protein bind preferentially to the denatured antigen. Mol. Immunol. 21: 673-677. Fujita, K. and Silver, J. (1993) Surprising lability of biotin-streptavidin bond during transcription of biotinylated DNA bound to paramagnetic streptavidin beads. Biotechniques 14: 608-617. Gao, F., Pan, B.-F., Zheng, W.-M., Ao, L.-M. and Gu, H.-C. (2005) Study of streptavidin coated onto PAMAM dendrimer modified magnetite nanoparticles. J. Magn. Magn. Mater. 293: 48-54. Gershon, P.D. and Khilko, S. (1995) Stable chelating linkage for reversible immobilization of oligohistidine tagged proteins in the BIAcore surface plasmon resonance detector. J. Immunol. Methods 183: 65-76. Gersten, D.M. and Marchalonis, J.J. (1978) A rapid, novel method for the solid-phase derivatization of IgG antibodies for immune-affinity chromatography. J. Immunol. Methods 24: 305-309. Giebel, L.B., Cass, R.T., Milligan, D.L., Young, D.C., Arze, R. and Johnson, C.R. (1995) Screening of cyclic peptide phage libraries identifies ligands that bind streptavidin with high affinities. Biochemistry 34: 15430-15435. 77 References 78 González, M., Bagatolli, L.A., Echabe, I., Arrondo, J.L., Argaraña, C.E., Cantor, C.R. and Fidelio, G.D. (1997) Interaction of biotin with streptavidin. Thermostability and conformational changes upon binding. J. Biol. Chem. 272: 11288-11294. González, M., Argaraña, C.E. and Fidelio, G.D. (1999) Extremely high thermal stability of streptavidin and avidin upon biotin binding. Biomol. Eng. 16: 67-72. Green, N.M. (1963) Avidin. 1. The use of [14C]biotin for kinetic studies and for assay. Biochem. J. 89: 585-591. Green, N.M. and Toms, E.J. (1973) The properties of subunits of avidin coupled to sepharose. Biochem. J. 133: 687-700. Green, N.M. (1975) Avidin. Adv. Protein Chem. 29: 85-133. Green, N.M. (1990) Avidin and streptavidin. Methods Enzymol. 184: 51-67. Griffin, J. and Odell, W.D. (1987) Ultrasensitive immunoradiometric assay for chorionic gonadotropin which does not cross-react with luteinizing hormone nor free beta chain of hCG and which detects hCG in blood of non-pregnant humans. J. Immunol. Methods 103: 275-283. Gross, J., Moller, R., Henke, W. and Hoesel, W. (2006) Detection of anti-EPO antibodies in human sera by a bridging ELISA is much more sensitive when coating biotinylated rhEPO to streptavidin rather than using direct coating of rhEPO. J. Immunol. Methods 313: 176-182. Guesdon, J.L., Ternynck, T. and Avrameas, S. (1979) The use of avidin-biotin interaction in immunoenzymatic techniques. J. Histochem. Cytochem. 27: 1131-1139. Gunaratna, P.C. and Wilson, G.S. (1990) Optimization of multienzyme flow reactors for determination of acetylcholine. Anal. Chem. 62: 402-407. Gunaratna, P.C. and Wilson, G.S. (1992) Use of monoclonal anti-enzyme antibodies for analytical purposes. Biotechnol. Prog. 8: 268-274. György, P., Melville, D.B., Burk, D. and du Vigneaud, V. (1940) Possible identity of vitamin H with biotin and coenzyme R. Science 91: 243-245. György, P., Rose, C.S., Eakin, R.E., Snell, E.E. and Williams, R.J. (1941) Egg-white injury as the result of nonabsorption or inactivation of biotin. Science 93: 477-478. Haab, B.B., Dunham, M.J. and Brown, P.O. (2001) Protein microarrays for highly parallel detection and quantitation of specific proteins and antibodies in complex solutions. Genome Biol. 2: 1-13. Haeuptle, M.T., Aubert, M.L., Djiane, J. and Kraehenbuhl, J.P. (1983) Binding sites for lactogenic and somatogenic hormones from rabbit mammary gland and liver. J. Biol. Chem. 258: 305-314. Harris, S.A., Wolf, D.E., Mozingo, R., Anderson, R.C., Arth, G.E., Easton, N.R., Heyl, D., Wilson, A.N. and Folkers, K. (1944) Biotin. II. Synthesis of biotin. J. Am. Chem. Soc. 66: 1756-1757. Haynes, C.A. and Norde, W. (1994) Globular proteins at solid/liquid interfaces. Colloids Surf. B Biointerfaces 2: 517-566. Heitzmann, H. and Richards, F.M. (1974) Use of the avidin-biotin complex for specific staining of biological membranes in electron microscopy. Proc. Natl. Acad. Sci. U. S. A. 71: 3537-3541. Hektor, H.J. and Scholtmeijer, K. (2005) Hydrophobins: proteins with potential. Curr. Opin. Biotechnol. 16: 434-439. Hemmilä, I., Dakubu, S., Mukkala, V.-M., Siitari, H. and Lövgren, T. (1984) Europium as a label in time- resolved immunofluorometric assays. Anal. Biochem. 137: 335-343. Hemmilä, I. and Laitala, V. (2005) Progress in lanthanides as luminescent probes. J. Fluoresc. 15: 529- 542. Hendrickson, W.A., Pähler, A., Smith, J.L., Satow, Y., Merritt, E.A. and Phizackerley, R.P. (1989) Crystal structure of core streptavidin determined from multiwavelength anomalous diffraction of synchrotron radiation. Proc. Natl. Acad. Sci. U. S. A. 86: 2190-2194. Hengsakul, M. and Cass, A.E. (1996) Protein patterning with a photoactivatable derivative of biotin. Bioconjugate Chem. 7: 249-254. 78 References 79 Herbert, V., Lau, K.S., Gottlieb, C.W. and Bleicher, S.J. (1965) Coated charcoal immunoassay of insulin. J. Clin. Endocrinol. Metab. 25: 1375-1384. Hermanson, G.T. (1996a) Modification of amino acids, peptides, and proteins. Bioconjugate Techniques. Academic Press, Inc., San Diego, pp. 3-26. Hermanson, G.T. (1996b) Protein cross-linking methods. Bioconjugate Techniques. Academic Press, Inc., San Diego, pp. 23-26. Hermanson, G.T. (1996c) The chemistry of reactive groups. Bioconjugate Techniques. Academic Press, Inc., San Diego, pp. 137-166. Hermanson, G.T. (1996d) Photoreactive chemical reactions. Bioconjugate Techniques. Academic Press, Inc., San Diego, pp. 163-166. Hermanson, G.T. (1996e) Heterobifunctional cross-linkers. Bioconjugate Techniques. Academic Press, Inc., San Diego, pp. 228-286. Hirsch, J.D., Eslamizar, L., Filanoski, B.J., Malekzadeh, N., Haugland, R.P. and Beechem, J.M. (2002) Easily reversible desthiobiotin binding to streptavidin, avidin, and other biotin-binding proteins: uses for protein labeling, detection, and isolation. Anal. Biochem. 308: 343-357. Hodneland, C.D., Lee, Y.S., Min, D.H. and Mrksich, M. (2002) Selective immobilization of proteins to self-assembled monolayers presenting active site-directed capture ligands. Proc. Natl. Acad. Sci. U. S. A. 99: 5048-5052. Hoffman, W.L. and O'Shannessy, D.J. (1988) Site-specific immobilization of antibodies by their oligosaccharide moieties to new hydrazide derivatized solid supports. J. Immunol. Methods 112: 113-120. Hofmann, K. and Axelrod, A.E. (1950) Microbiological activity of the guanidino analogs of biotin and oxybiotin. J. Biol. Chem. 187: 29-33. Hofmann, K., Wood, S.W., Brinton, C.C., Montibeller, J.A. and Finn, F.M. (1980) Iminobiotin affinity columns and their application to retrieval of streptavidin. Proc. Natl. Acad. Sci. U. S. A. 77: 4666-4668. Holden, M.A. and Cremer, P.S. (2003) Light activated patterning of dye-labeled molecules on surfaces. J. Am. Chem. Soc. 125: 8074-8075. Holden, M.A., Jung, S.Y. and Cremer, P.S. (2004) Patterning enzymes inside microfluidic channels via photoattachment chemistry. Anal. Chem. 76: 1838-1843. Holmberg, A., Blomstergren, A., Nord, O., Lukacs, M., Lundeberg, J. and Uhlén, M. (2005) The biotin- streptavidin interaction can be reversibly broken using water at elevated temperatures. Electrophoresis 26: 501-510. Homola, J. (2003) Present and future of surface plasmon resonance biosensors. Anal. Bioanal. Chem. 377: 528-539. Hong, M.-Y., Yoon, H.C. and Kim, H.-S. (2003) Protein-ligand interactions at poly(amidoamine) dendrimer monolayers on gold. Langmuir 19: 416-421. Hong, M.Y., Lee, D. and Kim, H.S. (2005) Kinetic and equilibrium binding analysis of protein-ligand interactions at poly(amidoamine) dendrimer monolayers. Anal. Chem. 77: 7326-7334. Howarth, M., Chinnapen, D.J., Gerrow, K., Dorrestein, P.C., Grandy, M.R., Kelleher, N.L., El-Husseini, A. and Ting, A.Y. (2006) A monovalent streptavidin with a single femtomolar biotin binding site. Nat. Methods 3: 267-273. Huang, J., Ichinose, I. and Kunitake, T. (2006) Biomolecular modification of hierarchical cellulose fibers through titania nanocoating. Angew. Chem.-Int. Edit. 45: 2883-2886. Huang, S.C., Swerdlow, H. and Caldwell, K.D. (1994) Binding of biotinylated DNA to streptavidin-coated polystyrene latex. Anal. Biochem. 222: 441-449. Huang, S.C., Stump, M.D., Weiss, R. and Caldwell, K.D. (1996) Binding of biotinylated DNA to streptavidin-coated polystyrene latex: effects of chain length and particle size. Anal. Biochem. 237: 115-122. 79 References 80 Huber, C., Liu, J., Egelseer, E.M., Moll, D., Knoll, W., Sleytr, U.B. and Sára, M. (2006) Heterotetramers formed by an S-layer-streptavidin fusion protein and core-streptavidin as a nanoarrayed template for biochip development. Small 2: 142-150. Hultman, T., Ståhl, S., Hornes, E. and Uhlén, M. (1989) Direct solid phase sequencing of genomic and plasmid DNA using magnetic beads as solid support. Nucleic Acids Res. 17: 4937-4946. Hyre, D.E., Le Trong, I., Merritt, E.A., Eccleston, J.F., Green, N.M., Stenkamp, R.E. and Stayton, P.S. (2006) Cooperative hydrogen bond interactions in the streptavidin-biotin system. Protein Sci. 15: 459-467. Härmä, H., Soukka, T. and Lövgren, T. (2001) Europium nanoparticles and time-resolved fluorescence for ultrasensitive detection of prostate-specific antigen. Clin. Chem. 47: 561-568. Ihalainen, P. and Peltonen, J. (2004) Immobilization of streptavidin onto biotin-functionalized Langmuir- Schaefer binary monolayers chemisorbed on gold. Sens. Actuators B Chem. 102: 207-218. Jacobsen, M.H. and Koch, T. (2000) Method of photochemical immobilization of ligands using quinones. US Patent No. 6033784. Jaenicke, R., Welsch, R., Sára, M. and Sleytr, U.B. (1985) Stability and self-assembly of the S-layer protein of the cell wall of Bacillus stearothermophilus. Biol. Chem. Hoppe-Seyler 366: 663-670. Jenne, A. and Famulok, M. (1999) Disruption of the streptavidin interaction with biotinylated nucleic acid probes by 2-mercaptoethanol. Biotechniques 26: 249-252, 254. Johnson, D.L. and Martin, L.L. (2005) Controlling protein orientation at interfaces using histidine tags: an alternative to Ni/NTA. J. Am. Chem. Soc. 127: 2018-2019. Joshi, K.S., Hoffmann, L.G. and Butler, J.E. (1992) The immunochemistry of sandwich ELISAs–V. The capture antibody performance of polyclonal antibody-enriched fractions prepared by various methods. Mol. Immunol. 29: 971-981. Jung, S.-H., Son, H.-Y., Yuk, J.S., Jung, J.-W., Kim, K.H., Lee, C.-H., Hwang, H. and Ha, K.-S. (2006) Oriented immobilization of antibodies by a self-assembled monolayer of 2- (biotinamido)ethanethiol for immunoarray preparation. Colloids Surf. B Biointerfaces 47: 107- 111. Kang, J.H., Choi, H.J., Hwang, S.Y., Han, S.H., Jeon, J.Y. and Lee, E.K. (2007) Improving immunobinding using oriented immobilization of an oxidized antibody. J. Chromatogr. A 1161: 9-14. Katz, B.A. (1997) Binding of biotin to streptavidin stabilizes intersubunit salt bridges between Asp61 and His87 at low pH. J. Mol. Biol. 274: 776-800. Keane, P.M., Pearson, J. and Walker, W.H. (1968) Dextran-coated charcoal immunoassay of insulin. Diabetologia 4: 339-344. Keefe, A.D., Wilson, D.S., Seelig, B. and Szostak, J.W. (2001) One-step purification of recombinant proteins using a nanomolar-affinity streptavidin-binding peptide, the SBP-Tag. Protein Expr. Purif. 23: 440-446. Kim, J., Yamasaki, R., Park, J., Jung, H., Lee, H. and Kawai, T. (2004) Highly dense protein layers confirmed by atomic force microscopy and quartz crystal microbalance. J. Biosci. Bioeng. 97: 138-140. Kim, K., Yang, H., Jon, S., Kim, E. and Kwak, J. (2004) Protein patterning based on electrochemical activation of bioinactive surfaces with hydroquinone-caged biotin. J. Am. Chem. Soc. 126: 15368-15369. Kim, Y.-P., Hong, M.-Y., Shon, H.K., Moon, D.W., Kim, H.-S. and Lee, T.G. (2006) Quantitative ToF- SIMS study of surface-immobilized streptavidin. Appl. Surf. Sci. 252: 6801-6804. Kingshott, P. and Griesser, H.J. (1999) Surfaces that resist bioadhesion. Curr. Opin. Solid State Mat. Sci. 4: 403-412. Klumb, L.A., Chu, V. and Stayton, P.S. (1998) Energetic roles of hydrogen bonds at the ureido oxygen binding pocket in the streptavidin-biotin complex. Biochemistry 37: 7657-7663. Knowles, J.R. (1989) The mechanism of biotin-dependent enzymes. Annu. Rev. Biochem. 58: 195-221. 80 References 81 Kochwa, S., Brownell, M., Rosenfield, R.E. and Wasserman, L.R. (1967) Adsorption of proteins by polystyrene particles. I. Molecular unfolding and acquired immunogenicity of IgG. J. Immunol. 99: 981-986. Kokko, L., Sandberg, K., Lövgren, T. and Soukka, T. (2004) Europium(III) chelate-dyed nanoparticles as donors in a homogeneous proximity-based immunoassay for estradiol. Anal. Chim. Acta 503: 155-162. Koyano, T., Saito, M., Miyamoto, Y., Kaifu, K. and Kato, M. (1996) Development of a technique for microimmobilization of proteins on silicon wafers by a streptavidin-biotin reaction. Biotechnol. Prog. 12: 141-144. Kumada, Y., Tokunaga, Y., Imanaka, H., Imamura, K., Sakiyama, T., Katoh, S. and Nakanishi, K. (2006) Screening and characterization of affinity peptide tags specific to polystyrene supports for the orientated immobilization of proteins. Biotechnol. Prog. 22: 401-405. Kumada, Y., Katoh, S., Imanaka, H., Imamura, K. and Nakanishi, K. (2007) Development of a one-step ELISA method using an affinity peptide tag specific to a hydrophilic polystyrene surface. J. Biotechnol. 127: 288-299. Kuningas, K., Rantanen, T., Ukonaho, T., Lövgren, T. and Soukka, T. (2005) Homogeneous assay technology based on upconverting phosphors. Anal. Chem. 77: 7348-7355. Kurzban, G.P., Gitlin, G., Bayer, E.A., Wilchek, M. and Horowitz, P.M. (1990) Biotin binding changes the conformation and decreases tryptophan accessibility of streptavidin. J. Protein Chem. 9: 673-682. Kurzban, G.P., Bayer, E.A., Wilchek, M. and Horowitz, P.M. (1991) The quaternary structure of streptavidin in urea. J. Biol. Chem. 266: 14470-14477. Kusnezow, W., Jacob, A., Walijew, A., Diehl, F. and Hoheisel, J.D. (2003) Antibody microarrays: an evaluation of production parameters. Proteomics 3: 254-264. Kyte, J. and Doolittle, R.F. (1982) A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157: 105-132. Ladd, J., Boozer, C., Yu, Q., Chen, S., Homola, J. and Jiang, S. (2004) DNA-directed protein immobilization on mixed self-assembled monolayers via a streptavidin bridge. Langmuir 20: 8090-8095. Laitinen, O.H., Hytönen, V.P., Nordlund, H.R. and Kulomaa, M.S. (2006) Genetically engineered avidins and streptavidins. Cell. Mol. Life Sci. 63: 2992-3017. Laitinen, O.H., Nordlund, H.R., Hytönen, V.P. and Kulomaa, M.S. (2007) Brave new (strept)avidins in biotechnology. Trends Biotechnol. 25: 269-277. Lamla, T. and Erdmann, V.A. (2004) The Nano-tag, a streptavidin-binding peptide for the purification and detection of recombinant proteins. Protein Expr. Purif. 33: 39-47. Langmuir, I. (1916) Constitution and fundamental properties of solids and liquids. I. Solids. J. Am. Chem. Soc. 38: 2221-2295. Langone, J.J. (1978) [125I]protein A: A tracer for general use in immunoassay. J. Immunol. Methods 24: 269-285. Lee, S.H. and Ruckenstein, E. (1988) Adsorption of proteins onto polymeric surfaces of different hydrophilicities - a case study with bovine serum albumin. J. Colloid Interface Sci. 125: 365- 379. Leggett, G.J., Davies, M.C., Jackson, D.E. and Tendler, S.J.B. (1996) Scanning probe microscopy of biomolecules and polymeric biomaterials. J. Electron Spectrosc. Relat. Phenom. 81: 249-268. Lehtonen, O.P. (1981) Immunoreactivity of solid phase hapten binding plasmacytoma protein (ABPC 24). Mol. Immunol. 18: 323-329. Leininger, R.I., Cooper, C.W., Falb, R.D. and Grode, G.A. (1966) Nonthrombogenic plastic surfaces. Science 152: 1625-1626. Li, S., Hu, J. and Liu, B. (2005) A study on the adsorption behavior of protein onto functional microspheres. J. Chem. Technol. Biotechnol. 80: 531-536. 81 References 82 Li, Y.J., Bi, L.J., Zhang, X.E., Zhou, Y.F., Zhang, J.B., Chen, Y.Y., Li, W. and Zhang, Z.P. (2006) Reversible immobilization of proteins with streptavidin affinity tags on a surface plasmon resonance biosensor chip. Anal. Bioanal. Chem. 386: 1321-1326. Linder, M., Selber, K., Nakari-Setälä, T., Qiao, M., Kula, M.R. and Penttilä, M. (2001) The hydrophobins HFBI and HFBII from Trichoderma reesei showing efficient interactions with nonionic surfactants in aqueous two-phase systems. Biomacromolecules 2: 511-517. Linder, M., Szilvay, G.R., Nakari-Setälä, T., Söderlund, H. and Penttilä, M. (2002) Surface adhesion of fusion proteins containing the hydrophobins HFBI and HFBII from Trichoderma reesei. Protein Sci. 11: 2257-2266. Lis, H. and Sharon, N. (1986) Lectins as molecules and as tools. Annu. Rev. Biochem. 55: 35-67. Liu, Z. and Amiridis, M.D. (2004) FT-IRRAS spectroscopic studies of the interaction of avidin with biotinylated dendrimer surfaces. Colloids Surf. B Biointerfaces 35: 197-203. Liu, Z. and Amiridis, M.D. (2005) FT-IRRAS quantitative analysis of specific avidin adsorption on biotinylated Au surfaces. Surf. Sci. 596: 117-125. Loomans, E.E., Petersen-van Ettekoven, A., Bloemers, H.P. and Schielen, W.J. (1997) Direct coating of poly(lys) or acetyl-thio-acetyl peptides to polystyrene: the effects in an enzyme-linked immunosorbent assay. Anal. Biochem. 248: 117-129. Loomans, E.E., Gribnau, T.C., Bloemers, H.P. and Schielen, W.J. (1998a) The influence of binding capacity and affinity on the improved performance of N-terminally extended hCG peptides, determined by ELISA-based procedures. J. Immunol. Methods 221: 119-130. Loomans, E.E.M.G., Gribnau, T.C.J., Bloemers, H.P.J. and Schielen, W.J.G. (1998b) Adsorption studies of tritium-labeled peptides on polystyrene surfaces. J. Immunol. Methods 221: 131-139. Loscar, E.S., Borzi, R.A. and Albano, E.V. (2003) Fluctuations of jamming coverage upon random sequential adsorption on homogeneous and heterogeneous media. Phys. Rev. E 68: 041106-1– 041106-9. Lu, B., Xie, J., Lu, C., Wu, C. and Wei, Y. (1995) Oriented Immobilization of Fab' Fragments on Silica Surfaces. Anal. Chem. 67: 83-87. Lu, B., Smyth, M.R. and O'Kennedy, R. (1996) Oriented immobilization of antibodies and its applications in immunoassays and immunosensors. Analyst 121: 29R-32R. Lu, J.R., Zhao, X. and Yaseen, M. (2007) Protein adsorption studied by neutron reflection. Curr. Opin. Colloid Interface Sci. 12: 9-16. Lyklema, J. (1984) Proteins at solid-liquid interfaces. A colloid-chemical review. Colloids Surf. 10: 33-42. Lövgren, J., Piironen, T., Övermo, C., Dowell, B., Karp, M., Pettersson, K., Lilja, H. and Lundwall, A. (1995) Production of recombinant PSA and HK2 and analysis of their immunologic cross- reactivity. Biochem. Biophys. Res. Commun. 213: 888-895. Malmsten, M. (1995) Ellipsometry studies of the effects of surface hydrophobicity on protein adsorption. Colloids Surf. B Biointerfaces 3: 297-308. Mark, S.S., Sandhyarani, N., Zhu, C., Campagnolo, C. and Batt, C.A. (2004) Dendrimer-functionalized self-assembled monolayers as a surface plasmon resonance sensor surface. Langmuir 20: 6808- 6817. Melville, D.B., Moyer, A.W., Hofmann, K. and du Vigneaud, V. (1942) The structure of biotin: The formation of thiophenevaleric acid from biotin. J. Biol. Chem. 146: 487-492. Mendelsohn, R. and Flach, C.R. (2002) Infrared reflection-absorption spectroscopy of lipids, peptides, and proteins in aqueous monolayers. Curr. Top. Membr. 52: 57-88. Michaeli, I., Absolom, D.R. and Van Oss, C.J. (1980) Diffusion of adsorbed protein within the plane of adsorption. J. Colloid Interface Sci. 77: 586-587. Miles, L.E. and Hales, C.N. (1968) Labelled antibodies and immunological assay systems. Nature 219: 186-189. 82 References 83 Moll, D., Huber, C., Schlegel, B., Pum, D., Sleytr, U.B. and Sára, M. (2002) S-layer-streptavidin fusion proteins as template for nanopatterned molecular arrays. Proc. Natl. Acad. Sci. U. S. A. 99: 14646-14651. Morag, E., Bayer, E.A. and Wilchek, M. (1996a) Reversibility of biotin-binding by selective modification of tyrosine in avidin. Biochem. J. 316: 193-199. Morag, E., Bayer, E.A. and Wilchek, M. (1996b) Immobilized nitro-avidin and nitro-streptavidin as reusable affinity matrices for application in avidin-biotin technology. Anal. Biochem. 243: 257- 263. Moseley, J.M., Marchalonis, J.J., Harris, A.W. and Pye, J. (1977) Molecular properties of T lymphoma immunoglobulin. I. Serological and general physicochemical properties. J. Immunogenet. 4: 233-248. Mouton, C.A., Pang, D., Natraj, C.V. and Shafer, J.A. (1982) A reagent for covalently attaching biotin to proteins via a cleavable connector arm. Arch. Biochem. Biophys. 218: 101-108. Mukkala, V.M., Mikola, H. and Hemmilä, I. (1989) The synthesis and use of activated N-benzyl derivatives of diethylenetriaminetetraacetic acids: alternative reagents for labeling of antibodies with metal ions. Anal. Biochem. 176: 319-325. Mukkala, V.-M., Hänninen, E. and Hemmilä, I. (1993) Synthesis of two biotin isothiocyanate derivatives. Eighth European Symposium on Organic Chemistry. Book of Abstracts. Universitat Autònoma de Barcelona, Barcelona, Spain. Nahar, P., Wali, N.M. and Gandhi, R.P. (2001) Light-induced activation of an inert surface for covalent immobilization of a protein ligand. Anal. Biochem. 294: 148-153. Nakajima, H., Ishino, S., Masuda, H., Nakagama, T., Shimosaka, T. and Uchiyama, K. (2006) Photochemical immobilization of protein on the inner wall of a microchannel and its application in a glucose sensor. Anal. Chim. Acta 562: 103-109. Nakanishi, K., Sakiyama, T. and Imamura, K. (2001) On the adsorption of proteins on solid surfaces, a common but very complicated phenomenon. J. Biosci. Bioeng. 91: 233-244. Naqvi, A. and Nahar, P. (2004) Photochemical immobilization of proteins on microwave-synthesized photoreactive polymers. Anal. Biochem. 327: 68-73. Neurath, H. and Bull, H.B. (1938) The surface activity of proteins. Chem. Rev. 23: 391-435. Ngai, P.K., Ackermann, F., Wendt, H., Savoca, R. and Bosshard, H.R. (1993) Protein A antibody-capture ELISA (PACE): an ELISA format to avoid denaturation of surface-adsorbed antigens. J. Immunol. Methods 158: 267-276. Nieba, L., Nieba-Axmann, S.E., Persson, A., Hämäläinen, M., Edebratt, F., Hansson, A., Lidholm, J., Magnusson, K., Karlsson, A.F. and Plückthun, A. (1997) BIACORE analysis of histidine-tagged proteins using a chelating NTA sensor chip. Anal. Biochem. 252: 217-228. Niemeyer, C.M., Sano, T., Smith, C.L. and Cantor, C.R. (1994) Oligonucleotide-directed self-assembly of proteins: semisynthetic DNA-streptavidin hybrid molecules as connectors for the generation of macroscopic arrays and the construction of supramolecular bioconjugates. Nucleic Acids Res. 22: 5530-5539. Niemeyer, C.M., Boldt, L., Ceyhan, B. and Blohm, D. (1999) DNA-Directed immobilization: efficient, reversible, and site-selective surface binding of proteins by means of covalent DNA-streptavidin conjugates. Anal. Biochem. 268: 54-63. Niemeyer, C.M., Wacker, R. and Adler, M. (2003) Combination of DNA-directed immobilization and immuno-PCR: very sensitive antigen detection by means of self-assembled DNA-protein conjugates. Nucleic Acids Res. 31: e90. Niveleau, A., Sage, D., Reynaud, C., Bruno, C., Legastelois, S., Thomas, V. and Dante, R. (1993) Covalent linking of haptens, proteins and nucleic acids to a modified polystyrene support. J. Immunol. Methods 159: 177-187. Nock, S., Spudich, J.A. and Wagner, P. (1997) Reversible, site-specific immobilization of polyarginine- tagged fusion proteins on mica surfaces. FEBS Lett. 414: 233-238. 83 References 84 Norde, W. and Lyklema, J. (1979) Thermodynamics of protein adsorption. Theory with special reference to the adsorption of human plasma albumin and bovine pancreas ribonuclease at polystyrene surfaces. J. Colloid Interface Sci. 71: 350-366. Norde, W. (1986) Adsorption of proteins from solution at the solid-liquid interface. Adv. Colloid Interface Sci. 25: 267-340. Nourse, A., Millar, D.B. and Minton, A.P. (2000) Physicochemical characterization of generation 5 polyamidoamine dendrimers. Biopolymers 53: 316-328. Nurmikko, P., Väisänen, V., Piironen, T., Lindgren, S., Lilja, H. and Pettersson, K. (2000) Production and characterization of novel anti-prostate-specific antigen (PSA) monoclonal antibodies that do not detect internally cleaved Lys145-Lys146 inactive PSA. Clin. Chem. 46: 1610-1618. Nuzzo, R.G. and Allara, D.L. (1983) Adsorption of bifunctional organic disulfides on gold surfaces. J. Am. Chem. Soc. 105: 4481-4483. Nygren, H., Czerkinsky, C. and Stenberg, M. (1985) Dissociation of antibodies bound to surface- immobilized antigen. J. Immunol. Methods 85: 87-95. Nygren, H. (1988) Experimental demonstration of lateral cohesion in a layer of adsorbed protein and in layers of gold-antibody complexes bound to surface immobilised antigen. J. Immunol. Methods 114: 107-114. Odell, W.D., Griffin, J. and Zahradnik, R. (1986) Two-monoclonal-antibody sandwich-type assay for thyrotropin, with use of an avidin-biotin separation technique. Clin. Chem. 32: 1873-1878. Odell, W.D. and Griffin, J. (1987) Two-monoclonal-antibody "sandwich"-type assay of human lutropin, with no cross reaction with choriogonadotropin. Clin. Chem. 33: 1603-1607. O'Shannessy, D.J., Dobersen, M.J. and Quarles, R.H. (1984) A novel procedure for labeling immunoglobulins by conjugation to oligosaccharide moieties. Immunol. Lett. 8: 273-277. O'Shannessy, D.J. and Quarles, R.H. (1985) Specific conjugation reactions of the oligosaccharide moieties of immunoglobulins. J. Appl. Biochem. 7: 347-355. Page, J.D., Derango, R. and Huang, A.E. (1998) Chemical modification of polystyrene's surface and its effect on immobilized antibodies. Colloid Surf. A-Physicochem. Eng. Asp. 132: 193-201. Peluso, P., Wilson, D.S., Do, D., Tran, H., Venkatasubbaiah, M., Quincy, D., Heidecker, B., Poindexter, K., Tolani, N., Phelan, M., Witte, K., Jung, L.S., Wagner, P. and Nock, S. (2003) Optimizing antibody immobilization strategies for the construction of protein microarrays. Anal. Biochem. 312: 113-124. Pérez-Luna, V.H., O'Brien, M.J., Opperman, K.A., Hampton, P.D., López, G.P., Klumb, L.A. and Stayton, P.S. (1999) Molecular recognition between genetically engineered streptavidin and surface- bound biotin. J. Am. Chem. Soc. 121: 6469-6478. Peterman, J.H., Tarcha, P.J., Chu, V.P. and Butler, J.E. (1988) The immunochemistry of sandwich- ELISAs. IV. The antigen capture capacity of antibody covalently attached to bromoacetyl surface-functionalized polystyrene. J. Immunol. Methods 111: 271-275. Piironen, T., Villoutreix, B.O., Becker, C., Hollingsworth, K., Vihinen, M., Bridon, D., Qiu, X., Rapp, J., Dowell, B., Lövgren, T., Pettersson, K. and Lilja, H. (1998) Determination and analysis of antigenic epitopes of prostate specific antigen (PSA) and human glandular kallikrein 2 (hK2) using synthetic peptides and computer modeling. Protein Sci. 7: 259-269. Piran, U. and Riordan, W.J. (1990) Dissociation rate constant of the biotin-streptavidin complex. J. Immunol. Methods 133: 141-143. Pirrung, M.C. and Huang, C.Y. (1996) A general method for the spatially defined immobilization of biomolecules on glass surfaces using "caged" biotin. Bioconjugate Chem. 7: 317-321. Porter, M.D., Bright, T.B., Allara, D.L. and Chidsey, C.E.D. (1987) Spontaneously organized molecular assemblies. 4. Structural characterization of n-alkyl thiol monolayers on gold by optical ellipsometry, infrared spectroscopy, and electrochemistry. J. Am. Chem. Soc. 109: 3559-3568. Pradier, C.M., Salmain, M., Liu, Z. and Méthivier, C. (2002a) Comparison of different procedures of biotin immobilization on gold for the molecular recognition of avidin: an FT-IRRAS study. Surf. Interface Anal. 34: 67-71. 84 References 85 Pradier, C.-M., Salmain, M., Zheng, L. and Jaouen, G. (2002b) Specific binding of avidin to biotin immobilized on modified gold surfaces. Fourier transform infrared reflection absorption spectroscopy analysis. Surf. Sci. 502-503: 193-202. Prisyazhnoy, V.S., Fusek, M. and Alakhov, Y.B. (1988) Synthesis of high-capacity immunoaffinity sorbents with oriented immobilized immunoglobulins or their Fab' fragments for isolation of proteins. J. Chromatogr. 424: 243-253. Pritchard, D.J., Morgan, H. and Cooper, J.M. (1995) Patterning and regeneration of surfaces with antibodies. Anal. Chem. 67: 3605-3607. Pum, D. and Sleytr, U.B. (1999) The application of bacterial S-layers in molecular nanotechnology. Trends Biotechnol. 17: 8-12. Pyun, J.C., Kim, S.D. and Chung, J.W. (2005) New immobilization method for immunoaffinity biosensors by using thiolated proteins. Anal. Biochem. 347: 227-233. Pähler, A., Hendrickson, W.A., Kolks, M.A., Argaraña, C.E. and Cantor, C.R. (1987) Characterization and crystallization of core streptavidin. J. Biol. Chem. 262: 13933-13937. Qin, Q.P., Lövgren, T. and Pettersson, K. (2001) Development of highly fluorescent detection reagents for the construction of ultrasensitive immunoassays. Anal. Chem. 73: 1521-1529. Qureshi, M.H. and Wong, S.L. (2002) Design, production, and characterization of a monomeric streptavidin and its application for affinity purification of biotinylated proteins. Protein Expr. Purif. 25: 409-415. Rabbany, S.Y., Piervincenzi, R., Judd, L., Kusterbeck, A.W., Bredehorst, R., Hakansson, K. and Ligler, F.S. (1997) Assessment of heterogeneity in antibody-antigen displacement reactions. Anal. Chem. 69: 175-182. Rao, S.V., Anderson, K.W. and Bachas, L.G. (1998) Oriented immobilization of proteins. Microchim. Acta 128: 127-143. Rasmussen, S.E. (1990) Covalent immobilization of biomolecules onto polystyrene MicroWells for use in biospecific assays. Ann. Biol. Clin. 48: 647-650. Reh, E., Birkner, C. and Weigert, M.P. (1986) Possibilities and limitations of hydrophobic interaction chromatography evaluation with streptavidin. Chromatographia 22: 391-398. Reis, K.J., Ayoub, E.M. and Boyle, M.D. (1984) Streptococcal Fc receptors. I. Isolation and partial characterization of the receptor from a group C streptococcus. J. Immunol. 132: 3091-3097. Reznik, G.O., Vajda, S., Smith, C.L., Cantor, C.R. and Sano, T. (1996) Streptavidins with intersubunit crosslinks have enhanced stability. Nat. Biotechnol. 14: 1007-1011. Reznik, G.O., Vajda, S., Cantor, C.R. and Sano, T. (2001) A streptavidin mutant useful for directed immobilization on solid surfaces. Bioconjugate Chem. 12: 1000-1004. Rickert, J., Brecht, A. and Göpel, W. (1997) Quartz crystal microbalances for quantitative biosensing and characterizing protein multilayers. Biosens. Bioelectron. 12: 567-575. Ruiz-Taylor, L.A., Martin, T.L. and Wagner, P. (2001a) X-ray photoelectron spectroscopy and radiometry studies of biotin-derivatized poly(L-lysine)-grafted-poly(ethylene glycol) monolayers on metal oxides. Langmuir 17: 7313-7322. Ruiz-Taylor, L.A., Martin, T.L., Zaugg, F.G., Witte, K., Indermuhle, P., Nock, S. and Wagner, P. (2001b) Monolayers of derivatized poly(L-lysine)-grafted poly(ethylene glycol) on metal oxides as a class of biomolecular interfaces. Proc. Natl. Acad. Sci. U. S. A. 98: 852-857. Sabanayagam, C.R., Smith, C.L. and Cantor, C.R. (2000) Oligonucleotide immobilization on micropatterned streptavidin surfaces. Nucleic Acids Res. 28: e33. Sakiyama, T., Ueno, S., Imamura, K. and Nakanishi, K. (2004) Use of a novel affinity tag selected with a bacterial random peptide library for improving activity retention of glutathione S-transferase adsorbed on a polystyrene surface. J. Mol. Catal. B-Enzym. 28: 207-214. Sano, T. and Cantor, C.R. (1990) Cooperative biotin binding by streptavidin. Electrophoretic behavior and subunit association of streptavidin in the presence of 6 M urea. J. Biol. Chem. 265: 3369-3373. 85 References 86 Sano, T., Smith, C.L. and Cantor, C.R. (1992) Immuno-PCR: very sensitive antigen detection by means of specific antibody-DNA conjugates. Science 258: 120-122. Sano, T. and Cantor, C.R. (1995) Intersubunit contacts made by tryptophan 120 with biotin are essential for both strong biotin binding and biotin-induced tighter subunit association of streptavidin. Proc. Natl. Acad. Sci. U. S. A. 92: 3180-3184. Sano, T., Pandori, M.W., Chen, X., Smith, C.L. and Cantor, C.R. (1995) Recombinant core streptavidins. A minimum-sized core streptavidin has enhanced structural stability and higher accessibility to biotinylated macromolecules. J. Biol. Chem. 270: 28204-28209. Sano, T., Vajda, S., Smith, C.L. and Cantor, C.R. (1997) Engineering subunit association of multisubunit proteins: a dimeric streptavidin. Proc. Natl. Acad. Sci. U. S. A. 94: 6153-6158. Sano, T. and Stayton, P.S. (1999) (Guest Eds.) (Strept)avidin biotin. Biomol. Eng. 16: 1-140 (special issue). Sarikaya, M., Tamerler, C., Jen, A.K.Y., Schulten, K. and Baneyx, F. (2003) Molecular biomimetics: nanotechnology through biology. Nat. Mater. 2: 577-585. Saviranta, P., Okon, R., Brinker, A., Warashina, M., Eppinger, J. and Geierstanger, B.H. (2004) Evaluating sandwich immunoassays in microarray format in terms of the ambient analyte regime. Clin. Chem. 50: 1907-1920. Schetters, H. (1999) Avidin and streptavidin in clinical diagnostics. Biomol. Eng. 16: 73-78. Schiestel, T., Brunner, H. and Tovar, G.E. (2004) Controlled surface functionalization of silica nanospheres by covalent conjugation reactions and preparation of high density streptavidin nanoparticles. J. Nanosci. Nanotechnol. 4: 504-511. Schmidt, T.G. and Skerra, A. (1993) The random peptide library-assisted engineering of a C-terminal affinity peptide, useful for the detection and purification of a functional Ig Fv fragment. Protein Eng. Des. Sel. 6: 109-122. Schmidt, T.G., Koepke, J., Frank, R. and Skerra, A. (1996) Molecular interaction between the Strep-tag affinity peptide and its cognate target, streptavidin. J. Mol. Biol. 255: 753-766. Schmidt, T.G. and Skerra, A. (2007) The Strep-tag system for one-step purification and high-affinity detection or capturing of proteins. Nat. Protoc. 2: 1528-1535. Schmitt, U., Maurer, E., Rüdinger, W., Deeg, R., Geiger, T. and Engel, W.D. (1994) Process for the determination of an immunologically detectable substance and a suitable reaction vessel therefor. US Patent No. 5362624. Schwidop, W.D., Klossek, P., Müller, R. and Claus, R. (1990) Procedure for the purification of streptavidin by hydrophobic interaction chromatography. J. Chromatogr. 520: 325-331. Scorilas, A., Bjartell, A., Lilja, H., Moller, C. and Diamandis, E.P. (2000) Streptavidin-polyvinylamine conjugates labeled with a europium chelate: applications in immunoassay, immunohistochemistry, and microarrays. Clin. Chem. 46: 1450-1455. Scouten, W.H. and Konecny, P. (1992) Reversible immobilization of antibodies on magnetic beads. Anal. Biochem. 205: 313-318. Sharma, H.K. and Tihon, C. (1988) Determination of avidin and streptavidin by a modified Bradford method. Anal. Biochem. 170: 135-139. Shmanai, V.V., Nikolayeva, T.A., Vinokurova, L.G. and Litoshka, A.A. (2001) Oriented antibody immobilization to polystyrene macrocarriers for immunoassay modified with hydrazide derivatives of poly(meth)acrylic acid. BMC Biotechnol. 1: 4. Shriver-Lake, L.C., Donner, B., Edelstein, R., Breslin, K., Bhatia, S.K. and Ligler, F.S. (1997) Antibody immobilization using heterobifunctional crosslinkers. Biosens. Bioelectron. 12: 1101-1106. Shu, W., Laue, E.D. and Seshia, A.A. (2007) Investigation of biotin-streptavidin binding interactions using microcantilever sensors. Biosens. Bioelectron. 22: 2003-2009. Siitari, H., Hemmilä, I., Soini, E., Lövgren, T. and Koistinen, V. (1983) Detection of hepatitis B surface antigen using time-resolved fluoroimmunoassay. Nature 301: 258-260. 86 References 87 Singh, P., Moll, F. 3rd, Lin, S.H., Ferzli, C., Yu, K.S., Koski, R.K., Saul, R.G. and Cronin, P. (1994) Starburst dendrimers: enhanced performance and flexibility for immunoassays. Clin. Chem. 40: 1845-1849. Singh, P. (1998) Terminal groups in Starburst dendrimers: activation and reactions with proteins. Bioconjugate Chem. 9: 54-63. Skerra, A. and Schmidt, T.G. (1999) Applications of a peptide ligand for streptavidin: the Strep-tag. Biomol. Eng. 16: 79-86. Sleytr, U.B. and Beveridge, T.J. (1999) Bacterial S-layers. Trends Microbiol. 7: 253-260. Sleytr, U.B., Messner, P., Pum, D. and Sára, M. (1999) Crystalline bacterial cell surface layers (S layers): from supramolecular cell structure to biomimetics and nanotechnology. Angew. Chem.-Int. Edit. 38: 1034-1054. Sleytr, U.B., Egelseer, E.M., Ilk, N., Pum, D. and Schuster, B. (2007) S-Layers as a basic building block in a molecular construction kit. FEBS J. 274: 323-334. Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J. and Klenk, D.C. (1985) Measurement of protein using bicinchoninic acid. Anal. Biochem. 150: 76-85. Snopok, B.A. and Kostyukevich, E.V. (2006) Kinetic studies of protein-surface interactions: A two-stage model of surface-induced protein transitions in adsorbed biofilms. Anal. Biochem. 348: 222-231. Soini, E. and Lövgren, T. (1987) Time-resolved fluorescence of lanthanide probes and applications in biotechnology. Crit. Rev. Anal. Chem. 18: 105-154. Soukka, T., Paukkunen, J., Härmä, H., Lönnberg, S., Lindroos, H. and Lövgren, T. (2001) Supersensitive time-resolved immunofluorometric assay of free prostate-specific antigen with nanoparticle label technology. Clin. Chem. 47: 1269-1278. Soukup, G.A., Cerny, R.L. and Maher, L.J. 3rd. (1995) Preparation of oligonucleotide-biotin conjugates with cleavable linkers. Bioconjugate Chem. 6: 135-138. Spaeth, K., Brecht, A. and Gauglitz, G. (1997) Studies on the biotin-avidin multilayer adsorption by spectroscopic ellipsometry. J. Colloid Interface Sci. 196: 128-135. Spitznagel, T.M. and Clark, D.S. (1993) Surface-density and orientation effects on immobilized antibodies and antibody fragments. Biotechnology (N. Y.) 11: 825-829. Spitznagel, T.M., Jacobs, J.W. and Clark, D.S. (1993) Random and site-specific immobilization of catalytic antibodies. Enzyme Microb. Technol. 15: 916-921. Starodub, N.F., Pirogova, L.V., Demchenko, A. and Nabok, A.V. (2005) Antibody immobilisation on the metal and silicon surfaces. The use of self-assembled layers and specific receptors. Bioelectrochemistry 66: 111-115. Stayton, P.S., Ding, Z. and Hoffman, A.S. (2004) Smart polymer-streptavidin conjugates. In: Niemeyer, C.M. (Ed.) Bioconjugation Protocols. Strategies and Methods. In: Walker, J.M. (Series Ed.) Methods in Molecular BiologyTM. Vol. 283, Humana Press Inc., Totowa, NJ, pp. 37-43. Streit, W.R. and Entcheva, P. (2003) Biotin in microbes, the genes involved in its biosynthesis, its biochemical role and perspectives for biotechnological production. Appl. Microbiol. Biotechnol. 61: 21-31. Strohner, P. (2000) Sensitive DNA-mutation analysis on BioTeZ thermostable streptavidin-coated microplates. Biomol. Eng. 16: 175. Abstract issue of the First International Conference on (Strept) Avidin-Biotin Technologies, The Banff Centre, Canada, 18 - 21 June, 2000. Strohner, P. and Immer, U. (2001) Surfaces coated with streptavidin/avidin. US Patent No. US 6270983 B1. Su, T.J., Lu, J.R., Thomas, R.K., Cui, Z.F. and Penfold, J. (1998) The effect of solution pH on the structure of lysozyme layers adsorbed at the silica-water interface studied by neutron reflection. Langmuir 14: 438-445. Sundberg, S.A., Barrett, R.W., Pirrung, M., Lu, A.L., Kiangsoontra, B. and Holmes, C.P. (1995) Spatially- addressable immobilization of macromolecules on solid supports. J. Am. Chem. Soc. 117: 12050-12057. 87 References 88 Suonpää, M., Markela, E., Ståhlberg, T. and Hemmilä, I. (1992) Europium-labelled streptavidin as a highly sensitive universal label. Indirect time-resolved immunofluorometry of FSH and TSH. J. Immunol. Methods 149: 247-253. Suter, M. and Butler, J.E. (1986) The immunochemistry of sandwich ELISAs. II. A novel system prevents the denaturation of capture antibodies. Immunol. Lett. 13: 313-316. Suter, M., Butler, J.E. and Peterman, J.H. (1989) The immunochemistry of sandwich ELISAs–III. The stoichiometry and efficacy of the protein-avidin-biotin capture (PABC) system. Mol. Immunol. 26: 221-230. Sutton, A., Vann, W.F., Karpas, A.B., Stein, K.E. and Schneerson, R. (1985) An avidin-biotin based ELISA for quantitation of antibody to bacterial polysaccharides. J. Immunol. Methods 82: 215- 224. Suzuki, N., Quesenberry, M.S., Wang, J.K., Lee, R.T., Kobayashi, K. and Lee, Y.C. (1997) Efficient immobilization of proteins by modification of plate surface with polystyrene derivatives. Anal. Biochem. 247: 412-416. Szilvay, G.R., Nakari-Setälä, T. and Linder, M.B. (2006) Behavior of Trichoderma reesei hydrophobins in solution: interactions, dynamics, and multimer formation. Biochemistry 45: 8590-8598. Szilvay, G.R., Paananen, A., Laurikainen, K., Vuorimaa, E., Lemmetyinen, H., Peltonen, J. and Linder, M.B. (2007) Self-assembled hydrophobin protein films at the air-water interface: structural analysis and molecular engineering. Biochemistry 46: 2345-2354. Takalo, H., Mukkala, V.M., Mikola, H., Liitti, P. and Hemmilä, I. (1994) Synthesis of europium(III) chelates suitable for labeling of bioactive molecules. Bioconjugate Chem. 5: 278-282. Tausig, F. and Wolf, F.J. (1964) Streptavidin - a substance with avidin-like properties produced by microorganisms. Biochem. Biophys. Res. Commun. 14: 205-209. Tengvall, P., Lundström, I. and Liedberg, B. (1998) Protein adsorption studies on model organic surfaces: an ellipsometric and infrared spectroscopic approach. Biomaterials 19: 407-422. Terpe, K. (2003) Overview of tag protein fusions: from molecular and biochemical fundamentals to commercial systems. Appl. Microbiol. Biotechnol. 60: 523-533. Tischer, W., Maier, J. and Deeg, R. (1991) Process for preparing a carrier useful in immunoassays by deposition of a complex of a specifically binding substance with hydrophobic protein, and the resulting carrier. US Patent No. 5061640. Tong, X. and Smith, L.M. (1992) Solid-phase method for the purification of DNA sequencing reactions. Anal. Chem. 64: 2672-2677. Torkkeli, M., Serimaa, R., Ikkala, O. and Linder, M. (2002) Aggregation and self-assembly of hydrophobins from Trichoderma reesei: low-resolution structural models. Biophys. J. 83: 2240- 2247. Traut, R.R., Bollen, A., Sun, T.T., Hershey, J.W., Sundberg, J. and Pierce, L.R. (1973) Methyl 4- mercaptobutyrimidate as a cleavable cross-linking reagent and its application to the Escherichia coli 30S ribosome. Biochemistry 12: 3266-3273. Turková, J. (1999) Oriented immobilization of biologically active proteins as a tool for revealing protein interactions and function. J. Chromatogr. B 722: 11-31. Underwood, P.A. and Steele, J.G. (1991) Practical limitations of estimation of protein adsorption to polymer surfaces. J. Immunol. Methods 142: 83-94. Updyke, T.V. and Nicolson, G.L. (1984) Immunoaffinity isolation of membrane antigens with biotinylated monoclonal antibodies and immobilized streptavidin matrices. J. Immunol. Methods 73: 83-95. Wacker, R. and Niemeyer, C.M. (2004) DDI-microFIA--A readily configurable microarray-fluorescence immunoassay based on DNA-directed immobilization of proteins. ChemBioChem 5: 453-459. Wacker, R., Schröder, H. and Niemeyer, C.M. (2004) Performance of antibody microarrays fabricated by either DNA-directed immobilization, direct spotting, or streptavidin-biotin attachment: a comparative study. Anal. Biochem. 330: 281-287. 88 References 89 Wahlgren, M.C., Arnebrant, T. and Paulsson, M.A. (1993) The adsorption from solutions of b- lactoglobulin mixed with lactoferrin or lysozyme onto silica and methylated silica surfaces. J. Colloid Interface Sci. 158: 46-53. Vallina-García, R., Del Mar García-Suárez, M., Fernández-Abedul, M.T., Méndez, F.J. and Costa-García, A. (2007) Oriented immobilisation of anti-pneumolysin Fab through a histidine tag for electrochemical immunosensors. Biosens. Bioelectron. 23: 210-217. Wang, G.P., Song, E.Q., Xie, H.Y., Zhang, Z.L., Tian, Z.Q., Zuo, C., Pang, D.W., Wu, D.C. and Shi, Y.B. (2005) Biofunctionalization of fluorescent-magnetic-bifunctional nanospheres and their applications. Chem. Commun. 4276-4278. Wang, J., Bhattacharyya, D. and Bachas, L.G. (2001) Improving the activity of immobilized subtilisin by site-directed attachment through a genetically engineered affinity tag. Fresenius J. Anal. Chem. 369: 280-285. Wang, Z.H. and Jin, G. (2004) Covalent immobilization of proteins for the biosensor based on imaging ellipsometry. J. Immunol. Methods 285: 237-243. Weber, P.C., Cox, M.J., Salemme, F.R. and Ohlendorf, D.H. (1987) Crystallographic data for Streptomyces avidinii streptavidin. J. Biol. Chem. 262: 12728-12729. Weber, P.C., Ohlendorf, D.H., Wendoloski, J.J. and Salemme, F.R. (1989) Structural origins of high- affinity biotin binding to streptavidin. Science 243: 85-88. Weber, P.C., Pantoliano, M.W. and Thompson, L.D. (1992) Crystal structure and ligand-binding studies of a screened peptide complexed with streptavidin. Biochemistry 31: 9350-9354. Vermeer, A.W., Giacomelli, C.E. and Norde, W. (2001) Adsorption of IgG onto hydrophobic teflon. Differences between the F(ab) and F(c) domains. Biochim. Biophys. Acta 1526: 61-69. Wide, L. and Porath, J. (1966) Radioimmunoassay of proteins with the use of Sephadex-coupled antibodies. Biochim. Biophys. Acta 130: 257-260. Wide, L., Bennich, H. and Johansson, S.G. (1967) Diagnosis of allergy by an in-vitro test for allergen antibodies. Lancet 2: 1105-1107. Wide, L. (1969) Radioimmunoassays employing immunosorbents. Acta Endocrinol. Suppl. (Copenh.) 142: 207-221. Wide, L. (2005) Inventions leading to the development of the diagnostic test kit industry - from the modern pregnancy test to the sandwich assays. Ups. J. Med. Sci. 110: 193-216. Widjojoatmodjo, M.N., Fluit, A.C., Torensma, R. and Verhoef, J. (1993) Comparison of immunomagnetic beads coated with protein A, protein G, or goat anti-mouse immunoglobulins. Applications in enzyme immunoassays and immunomagnetic separations. J. Immunol. Methods 165: 11-19. Vijayendran, R.A. and Leckband, D.E. (2001) A quantitative assessment of heterogeneity for surface- immobilized proteins. Anal. Chem. 73: 471-480. Wilbur, D.S., Pathare, P.M., Hamlin, D.K., Stayton, P.S., To, R., Klumb, L.A., Buhler, K.R. and Vessella, R.L. (1999) Development of new biotin/streptavidin reagents for pretargeting. Biomol. Eng. 16: 113-118. Wilchek, M. and Bayer, E.A. (1988) The avidin-biotin complex in bioanalytical applications. Anal. Biochem. 171: 1-32. Wilchek, M. and Bayer, E.A. (1990) (Eds.) Avidin-Biotin Technology. In: Methods in Enzymology, Vol. 184. Academic Press, Inc., San Diego. Wilde, L.M., Farace, G., Roberts, C.J., Davies, M.C., Sanders, G.H., Tendler, S.J. and Williams, P.M. (2001) Molecular patterning on carbon based surfaces through photobiotin activation. Analyst 126: 195-198. Wilkins Stevens, P., Hansberry, M.R. and Kelso, D.M. (1995) Assessment of adsorption and adhesion of proteins to polystyrene microwells by sequential enzyme-linked immunosorbent assay analysis. Anal. Biochem. 225: 197-205. Wilkins Stevens, P. and Kelso, D.M. (1995) Estimation of the protein-binding capacity of microplate wells using sequential ELISAs. J. Immunol. Methods 178: 59-70. 89 References 90 von Lode, P., Rosenberg, J., Pettersson, K. and Takalo, H. (2003) A europium chelate for quantitative point-of-care immunoassays using direct surface measurement. Anal. Chem. 75: 3193-3201. Wong, J., Chilkoti, A. and Moy, V.T. (1999) Direct force measurements of the streptavidin-biotin interaction. Biomol. Eng. 16: 45-55. Voss, S. and Skerra, A. (1997) Mutagenesis of a flexible loop in streptavidin leads to higher affinity for the Strep-tag II peptide and improved performance in recombinant protein purification. Protein Eng. 10: 975-982. Vroman, L. and Adams, A.L. (1969) Findings with the recording ellipsometer suggesting rapid exchange of specific plasma proteins at liquid/solid interfaces. Surf. Sci. 16: 438-446. Vroman, L. and Adams, A.L. (1986) Adsorption of proteins out of plasma and solutions in narrow spaces. J. Colloid Interface Sci. 111: 391-402. Wu, A.H. (2006) A selected history and future of immunoassay development and applications in clinical chemistry. Clin. Chim. Acta 369: 119-124. Wu, F.B., Han, S.Q., Xu, T. and He, Y.F. (2003) Sensitive time-resolved fluoroimmunoassay for simultaneous detection of serum thyroid-stimulating hormone and total thyroxin with Eu and Sm as labels. Anal. Biochem. 314: 87-96. Wu, S.C. and Wong, S.L. (2005) Engineering soluble monomeric streptavidin with reversible biotin binding capability. J. Biol. Chem. 280: 23225-23231. Wösten, H.A. (2001) Hydrophobins: multipurpose proteins. Annu. Rev. Microbiol. 55: 625-646. Xia, N., Shumaker-Parry, J.S., Zareie, M.H., Campbell, C.T. and Castner, D.G. (2004) A streptavidin linker layer that functions after drying. Langmuir 20: 3710-3716. Xu, H., Lu, J.R. and Williams, D.E. (2006) Effect of surface packing density of interfacially adsorbed monoclonal antibody on the binding of hormonal antigen human chorionic gonadotrophin. J. Phys. Chem. B 110: 1907-1914. Xu, H., Zhao, X., Lu, J.R. and Williams, D.E. (2007) Relationship between the structural conformation of monoclonal antibody layers and antigen binding capacity. Biomacromolecules 8: 2422-2428. Yalow, R.S. and Berson, S.A. (1959) Assay of plasma insulin in human subjects by immunological methods. Nature 184: 1648-1649. Yalow, R.S. and Berson, S.A. (1960) Immunoassay of endogenous plasma insulin in man. J. Clin. Invest. 39: 1157-1175. Yam, C.M., Pradier, C.M., Salmain, M., Marcus, P. and Jaouen, G. (2001) Binding of biotin to gold surfaces functionalized by self-assembled monolayers of cystamine and cysteamine: Combined FT-IRRAS and XPS characterization. J. Colloid Interface Sci. 235: 183-189. Yang, C. (1998) Statistical mechanical study on the freundlich isotherm equation. J. Colloid Interface Sci. 208: 379-387. Yang, Z., Frey, W., Oliver, T. and Chilkoti, A. (2000) Light-activated affinity micropatterning of proteins on self-assembled monolayers on gold. Langmuir 16: 1751-1758. Ye, L., Pelton, R. and Brook, M.A. (2007) Biotinylation of TiO2 nanoparticles and their conjugation with streptavidin. Langmuir 23: 5630-5637. Ylikotila, J., Välimaa, L., Vehniäinen, M., Takalo, H., Lövgren, T. and Pettersson, K. (2005) A sensitive TSH assay in spot-coated microwells utilizing recombinant antibody fragments. J. Immunol. Methods 306: 104-114. Zahradnik, R., Brennan, G., Hutchison, J.S. and Odell, W.D. (1989) Immunoradiometric assay of corticotropin with use of avidin-biotin separation. Clin. Chem. 35: 804-807. Zhang, Y., Fung, Y., Sun, H., Zhu, D. and Yao, S. (2005) Study of protein adsorption on polymer coatings surface by combining quartz crystal microbalance with electrochemical impedance methods. Sens. Actuator B-Chem. B108: 933-942. Zhang, Z., Zhu, H., Tang, Y., Cui, T., Geng, T., Chen, C. and Cui, Y. (2007) Preparation and application of streptavidin magnetic particles. Sci. China Ser. B-Chem. 50: 127-134. 90