ADSORPTION OF BOVINE SERUM ALBUMIN TO POLYETHYLENE TUBING REVERSIBILITY AND pH-DEPENDENCE By JUDY NEEDHAM B.Sc, The University of Leeds, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Chemistry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September 1988 0 JUDY NEEDHAM, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^ T « - V The University of British Columbia Vancouver, Canada Date QctoVaer; \^>g^ DE-6 (2/88) ABSTRACT This thesis i s concerned with the adsorption of bovine serum albumin to polyethylene tubing. A method using radioiodinated protein was developed to measure the surface concentration taking into account the d i l u t i o n e f f e c t f o r miscible displacement i n a c a p i l l a r y . A steady-state surface concentration was established within 2 hours. Adsorption did not depend on the r a t i o of r a d i o l a b e l l e d to unlabel led protein. The adsorption isotherm was Langmuir -1ike with a plateau concentration of approximately 2 0.2 ug/cm . Two methods were used to c a l c u l a t e the surface concentration i n the desorption study. The surface concentration calculated by depletion of the t o t a l r a d i o a c t i v i t y was always higher than that calculated from assaying the r a d i o a c t i v i t y associated with the tubing. Desorption of at least 5% of the loosely bound protein occurs. The surface concentration-pH data show two maxima. The f i r s t i s at the i s o e l e c t r i c point of the albumin while the second i s at pH 9.5-10. The second maximum seems to be due to p r e f e r e n t i a l adsorption of the higher molecular weight oligomers i n the protein sample. TABLE OF CONTENTS Page ABSTRACT i i LIST OF TABLES v LIST OF FIGURES v i LIST OF APPENDICES ix ACKNOWLEDGEMENTS x CHAPTER 1 INTRODUCTION 1 1.1 Protein structure 3 1.1.1 Bovine serum albumin i n s o l u t i o n 4 1.1.2 Albumin oligomers 5 1.2 Driving forces f o r adsorption 6 1.3 Mechanisms f o r adsorption 8 1.4 C h a r a c t e r i s t i c s of protein adsorption 11 1.4.1 Protein desorption and exchange 17 1.5 Objectives and methods 21 CHAPTER 2 EXPERIMENTAL 24 2.1 Ra d i o l a b e l l i n g 24 2.1.1 TCA p r e c i p i t a t i o n assay 25 2. 1.2 Gel f i l t r a t i o n 25 2.1.3 Thin-layer chromatography 26 2.2 Protein electrophoresis 26 2.1.1 Materials 27 2.2.2 Sample preparation 27 2.2.3 Method 28 2.2.4 Gel s t a i n i n g 30 2.3 Protein concentration 30 2.4 Adsorption experiments 31 2.4.1 Methods and materials 31 2.4.2 Surface concentration as a function of time 32 2.4.3 Surface concentration as a function of the r a t i o of r a d i o l a b e l l e d to unlabel led BSA 33 Page 2.4.4 Adsorption isotherm 33 2.4.5 Surface concentration as a function of pH 33 CHAPTER 3 RESULTS AND DISCUSSION 34 3.1 R a d i o l a b e l l i n g 34 3.1.1 Degree of r a d i o l a b e l l i n g 34 3.2 E x t i n c t i o n c o e f f i c i e n t 36 3.3 SDS-PAGE 38 3.4 Adsorption experiments 49 3.4.1 Surface concentration as a function of the r a t i o of r a d i o l a b e l l e d to unlabel led BSA 49 3.4.2 Surface soncentration as a function of time 50 3.4.3 Adsorption isotherm 52 3.4.4 R e v e r s i b i l i t y of the adsorbed BSA 59 3.4.5 E f f e c t of pH on BSA adsorption 66 CHAPTER 4 CONCLUSIONS 72 REFERENCES 75 iv LIST OF TABLES Table Page 1.1 The adsorption of human serum albumin (HSA) on various surfaces. 13 2.1 Composition of buffers and 3.75% gels f o r SDS-PAGE. 29 3. 1 Molecular weight assignments f or the protein standards used on the 3.75% SDS-PAGE gels. 39 3.2 The e f f e c t of pH on the amount of BSA polymer present i n the stock solutions. The amount of polymer i s represented as a percentage of the t o t a l p r otein ± error. The errorwas c a l c u l a t e d from the a c t i v i t y ; see Appendix 2. 43 3.3 The amount of BSA polymer present i n various samples represented as a percentage of the t o t a l protein ± error. 46 3.4 The e f f e c t of l a b e l l e d BSA content on the adsorption to polyethylene from a 0.5 mg/ml solution, (see Appendix 2 fo r the c a l c u l a t i o n of AD. 50 3.5 Dimensions of bovine serum albumin. 58 2 3.6 Comparison of the surface concentration (ug/cm ) of BSA at pH 7.4 c a l c u l a t e d from cutt i n g up the tubing, T (tube), and from by depletion of the t o t a l counts, T (cpm); see Appendix 2 f o r error analysis. 62 3.7 Surface concentration ( D of BSA at pH 7.4 from a 0.5 mg/ml sol u t i o n . 67 3.8 Surface concentration ( D calculated from the ri n s e d tube and from depletion of t o t a l cpm. The pH was adjusted with borax or NaOH. 69 v LIST OF FIGURES Figure 1.1 A schematic view of a protein i n t e r a c t i n g with a well-characterized surface. The protein has a number of surface domains with hydrophobic, charged and polar character. The s o l i d may have a s i m i l a r domain-like character. (Taken from Andrade, 1985, p.4). 3.1 R a d i o a c t i v i t y vs f r a c t i o n number for the samples c o l l e c t e d from a Sephadex G-25 column following r a d i o l a b e l l i n g . 3.2 C a l i b r a t i o n curve of BSA. Optical density vs BSA concentration at 278 nm. 3.3 The molecular weight on a semi-log scale i s pl o t t e d against the r e l a t i v e mobility (R ) f o r a va r i e t y of SDS-protein complexes run on 3.75% gels. 3.4 SDS-PAGE rod gels stained with coomassie blue. A low molecular weight standards B high molecular weight C unlabel led stock BSA D r a d i o l a b e l l e d BSA 3.5 The mobility of BSA in 3.75% SDS-PAGE gels: ( ) densitometric scan of BSA. The protein was coomassie blue stained and scanned at 595 nm. TD = tracking dye. ( ) r e l a t i v e mobility vs cpm f o r r a d i o l a b e l l e d BSA. 3. 6a Relat ive mobi1ity vs r e l a t ive radioact i v i t y of BSA, pH 7. 4. 3. 6b Re1 at i ve mobi1ity vs r e l a t ive radioact i v i t y of BSA, pH 4.4. 3. 6c Re 1 at i ve mobi1ity vs r e l a t ive r a d i o a c t i v i t y of BSA, pH 9.4. 3. 6d Re 1 at i ve mobi1ity vs r e l a t ive radioact i v i t y of BSA, pH 12.0 3-7a Re 1 at i ve mobi1ity vs r e l a t ive r a d i o a c t i v i t y of sample ( i ) . 3. 7b Re 1 at i ve mobi1ity vs r e l a t i v e radioact i v i t y of sample ( i i ) . 3. 7c Re 1 at i ve mobi1ity vs r e l a t ive radioact i v i t y of sample ( i i i ) 3. 7d Re 1 at i ve mobi1ity vs r e l a t i v e r a d i o a c t i v i t y of sample ( i v ) . vi Figure Page 3.8 Time dependence f or the adsorption of BSA on polyethylene at a s o l u t i o n concentration of 0.19 mg/ml. 51 3.9 Adsorption isotherm f o r albumin on polyethylene at 23°C. 53 3. 10a Surface concentration vs tube section f o r BSA adsorbed from a 8 ug/ml solution. In a l l cases tube s e c t i o n 1 i s the f r a c t i o n c o l l e c t o r end i.e. , output end, and tube section 25 i s the syringe end i.e. , input end. 54 3.10b Surface concentration vs tube s e c t i o n f o r BSA adsorbed from a 17 fig/ml solution. 54 3. 10c Surface concentration vs tube s e c t i o n f o r BSA adsorbed from a 44 ug/ml solution. 55 3. lOd Surface concentration vs tube s e c t i o n f o r BSA adsorbed from a 67 ug/ml solution. 55 3. lOe Surface concentration vs tube s e c t i o n f o r BSA adsorbed from a 178 fug/ml solution. 56 3. lOf Surface concentration vs tube s e c t i o n f o r BSA adsorbed from a 326 fig/ml solution. 56 3. lOd Surface concentration vs tube section f o r BSA adsorbed from a 493 ug/ml solution. 57 3. lOe Surface concentration vs tube section f o r BSA adsorbed from a 2.694 mg/ml solution. 57 3.11 The concentration of the BSA s o l u t i o n displaced from an adsorption experiment vs the volume c o l l e c t e d . The i n i t i a l bulk concentrations of the BSA sol u t i o n s were (A) 274 and (B) 761 pg/ml. 63 3. 12 Adsorption isotherms f o r BSA on polyethylene. The surface concentration was calculated from the rinsed tubing (*) and by depletion of the t o t a l r a d i o a c t i v i t y (o). 64 3.13 Cb/r versus Cb f o r adsorption data c a l c u l a t e d from the rin s e d tubing (*) and by depletion of the t o t a l r a d i o a c t i v i t y (o). 65 3. 14 The surface concentration of BSA adsorbed to polyethylene tubing p l o t t e d against the pH. The bulk concentration i n each case was 0.5 mg/ml. 68 3.15 Adsorption of BSA on polyethylene as a function of pH from a 0.5 mg/ml solution. The surface concentrations were c a l c u l a t e d from the rinsed tubing (*) and from depletion of vi i Figure Page the t o t a l r a d i o a c t i v i t y ( o ) . 71 A3.1 C/C Q vs x f o r the miscible displacement of BSA with buffer i n a c a p i l l a r y tube. 93 v i i i LIST OF APPENDICES Page APPENDIX 1 ABBREVIATIONS 81 APPENDIX 2 CALCULATIONS AND ERROR ANNALYSIS 83 A2.1 Error analysis 83 A2.2 Imprecision i n the a c t i v i t y 85 A2.3 SDS-PAGE GELS, '/.monomer or polymer ± error 87 A2.4 Surface concentration from c u t t i n g up the tubing 88 A2.5 Surface concentration calculated from the t o t a l counts 88 APPENDIX 3 MISCIBLE DISPLACEMENT IN A CAPILLARY 90 APPENDIX 4 STATISTICAL ANALYSIS 95 A4.1 Testing the differ e n c e between the two regression l i n e s 98 A4.2 The paired-sample test f o r the adsorption isotherm data 99 A4.3 The paired-sample test f o r the pH data 100 ix ACKNOWLEDGEMENTS I would l i k e to thank Dr. D. E. Brooks, my supervisor, f o r his help and guidance throughout t h i s study; Johann Janzen f o r his help with some of the experimental work and invaluable discussion; Charles Ramey f o r h i s discussion and suggestions. A s p e c i a l thank-you goes to a l l the members of the Brooks' lab f o r a l l t h e i r help and support. F i n a l l y I wish to thank Terry J a r v i s and Tim Haddad f o r t h e i r patience during the preparation of t h i s thesis. x CHAPTER 1 INTRODUCTION Protein adsorption to s o l i d surfaces i s of great b i o l o g i c a l , medical and technological s i g n i f i c a n c e . The blood c o m p a t i b i l i t y or i n c o m p a t i b i l i t y with non-biological materials i s important e s p e c i a l l y with the increasing use of p r o s t h e t i c materials i n the body and t h i s i s generally considered to be r e l a t e d to protein adsorption. When a synthetic material i s introduced into the cardiovascular system the i n i t i a l event i n a complex seri e s of reactions i s the rapid adsorption of a proteinaceous layer (Baier and Dutton, 1969). Subsequent c e l l u l a r i n t e r a c t i o n s lead to thrombus formation, the entrapment of erthyrocytes and other formed blood elements i n a f i b r i n network, and coagulation as determined by the adsorbed proteins (Baier, 1977; Brash, 1981). It has been observed that d i f f e r e n t materials have d r a s t i c a l l y d i f f e r e n t thrombogenic a c t i v i t y . This suggests that the c h a r a c t e r i s t i c s of the p r o t e i n layers are d i f f e r e n t for d i f f e r e n t materials. It has been shown that by precoating surfaces with plasma proteins, p l a t e l e t adhesion, a p r e r e q u i s i t e to thrombus formation, i s g r e a t l y altered. An albumin coated surface reduces p l a t e l e t adhesion (Lyman et al., 1971; Packman et al., 1969), while fibrinogen greatly enhances i t and 9r-globulin a c t i v a t e s the release reaction (Packman et al., 1969; Jenkins et al., 1973). Protein adsorption i s a key event i n the blood-surface i n t e r a c t i o n 1 that .nust be understood before we understand the mechanisms of surface-induced thrombosis. Over the past 30 years many investigators have studied protein adsorption to a v a r i e t y of non-biological surfaces. Many techniques have been developed providing information regarding the absolute q u a n t i t i e s adsorbed from s i n g l e solutions (Bull, 1956; Brash and Uniyal, 1979), the r e l a t i v e q u a n t i t i e s adsorbed from complex solutions (Brash and Davidson, 1976; Lee et al., 1974), the number of surface attachments (Morrissey and Stromberg, 1974), desorption and exchange (Brash and Samak, 1978; Chuang et al., 1978), and measurements of the adsorbed layer thickness (Morrissey et al., 1976; Cuypers et al., 1977). There i s s t i l l a lot of controversy regarding protein adsorption at s o l i d / l i q u i d interfaces, p a r t i c u l a r l y concerning the r e v e r s i b i l i t y and the c o n f i g u r a t i o n of the adsorbed molecules. The present study was aimed at c l a r i f y i n g the question of r e v e r s i b i l i t y . Various techniques have been employed to study protein desorption. Solution-depletion methods followed by d i l u t i o n were used to study albumin desorption from glass (Bull, 1956), and from s i l i c a (MacRitchie, 1972). Radiolabelled proteins were used to give a d i r e c t measure of the amount of adsorbed protein. Desorption studies usually followed a r i n s i n g period (Brash et al. , 1974). In t h i s study a more d i r e c t technique using a r a d i o l a b e l l e d protein was developed to measure the amount of protein desorbed and information regarding albumin adsorption to polyethylene and r e v e r s i b i l i t y was obtained. Many f a c t o r s contribute to determining the c h a r a c t e r i s t i c s of protein 2 adsorption including the nature of the protein, the medium in which i t i s located and the nature of the s o l i d surface. These properties give an insight into the d r i v i n g forces and the mechanisms of adsorption. The information required to give a complete picture of protein adsorption includes the amount of protein adsorbed as a function of s o l u t i o n concentration and time (adsorption k i n e t i c s ) . The o r i e n t a t i o n and conformation of the protein upon adsorption i s important since a conformational change, known as denaturation, a l t e r s the properties of the protein. Other desired information includes the capacity of the protein i n s o l u t i o n to compete f o r the surface, and the a b i l i t y of the protein to desorb or exchange. 1.1 Protein structure Protein structure i s largely determined by the interactions among the amino acids which comprise i t , and between the protein molecule and the environment. Because of the structure-function r e l a t i o n s h i p s of proteins t h e i r three-dimensional structure i s of interest. Proteins are high molecular weight polyamides b u i l t up by the s p e c i f i c copolymerization of amino acids f o r p a r t i c u l a r functions. Each protein has a unique amino a c i d sequence known as the primary structure. The secondary structures, ordered three-dimensional regions, are the a-helix and (3-sheet r e s u l t i n g from hydrogen bonding in the protein backbone. The protein's t e r t i a r y structure i s the complete three-dimensional structure and i s the r e s u l t of intramolecular interactions such as i o n i c or e l e c t r o s t a t i c i n t e r a c t i o n s , hydrogen bonding, hydrophobic interactions, s a l t bridges and covalent d i s u l f i d e bonds. The noncovalent a s s o c i a t i o n of independent 3 t e r t i a r y structure gives the quarternary structure. 1.1.1 Bovine serum albumin i n s o l u t i o n Serum albumin i s the protein that i s present in the largest amount in blood plasma, where approximately h a l f of the protein i s albumin. It i s one of the most intensely studied proteins. Globular proteins i n aqueous solutions near t h e i r i s o e l e c t r i c point have compact configurations with low permeability to water. The nonpolar groups are l a r g e l y excluded from the surface of the protein while the polar and charged amino acids are at the surface and interact strongly with water. S h i f t i n g the pH of a protein s o l u t i o n away from the i s o e l e c t r i c point can decrease the s t a b i l i t y of the protein. The protein unfolds exposing the inner part of the molecule. Bovine serum albumin (BSA) expands very r e a d i l y below the i s o e l e c t r i c point of pH 4.9. This expansion at pH 4.5-3.5 depends on the fact that the carboxyl groups are s u c c e s s i v e l y transformed into the uncharged form, leaving the p o s i t i v e l y charged groups i n excess. The repulsion between various segments of the peptide chains increases. This conformational change i s usually c a l l e d the N-F transformation (Foster, 1960). BSA i s somewhat more stable on the a l k a l i n e side of the i s o e l e c t r i c point. Expansion s i m i l a r to the conformational change i n acid takes place i n a l k a l i n e pH, but the a l t e r a t i o n i n s i z e and shape does not begin to occur u n t i l pH 10.3 (Tanford et al., 1955). A small conformational change also has been shown to occur in the pH i n t e r v a l 7-8 (Leonard et al., 1963; Harmsen et al., 1971). The s o l u b i l i t y of BSA i s a function of pH. As the pH moves away from the i s o e l e c t r i c point the net charge on the protein increases thus 4 increasing the s o l u b i l i t y . 1.1.2 Albumin oligomers Albumin samples are heterogeneous in the sense that they contain mercaptalbumin, the f r a c t i o n having a f r e e l y reactive s u l f h y d r y l group, and nonmercaptalbumin, the f r a c t i o n showing no sulfhy d r y l a c t i v i t y . Most c a r e f u l l y prepared albumin preparations have a sulfhy d r y l content of 0.65-0.70 su l f h y d r y l groups per albumin molecule. In most regards mercaptalbumin and nonmercaptalbumin are remarkably s i m i l a r . Physical chemical studies on whole serum albumin y i e l d r e s u l t s i n d i s t i n g u i s h a b l e from those on mercaptalbumin. In v i r t u a l l y a l l samples of BSA, dimers and higher oligomers, aggregates of monomers, ex i s t . It has been shown that such dimers and oligomers a r i s e as a r t i f a c t s during and a f t e r i s o l a t i o n and are not present i n the bloodstream (Andersson, 1966). Albumin preparations contain variable amounts of dimers and higher polymers depending on the source of plasma ( F r i e d l i and K i s t l e r , 1970), the f r a c t i o n a t i o n procedure (Smith et al., 1972; S o l l i and B e r t o l i n i , 1977), the storage conditions (Finlayson et al., 1960) and the length of time of storage (Finlayson et al., 1960; Finlayson, 1965). The most l i k e l y source of dimerization would be the d i r e c t formation of d i s u l f i d e linkages through oxidation reactions involving the s u l f h y d r y l residues of two mercaptalbumin monomers. If t h i s i s true then a t h i o l reagent would break the d i s u l f i d e linkage, but i t i s well known that t y p i c a l albumin samples contain a portion of dimeric forms which are not broken down by reduction with t h i o l reagents (Hartley et al., 1962; 5 Janatova et al., 1968). Andersson showed that dimer i s o l a t e d from serum albumin was heterogeneous (Andersson, 1966). Approximately one-third of the dimer was s p l i t into monomer by mercaptoethanol, a reducing agent, at pH 8 or by standing i n a l k a l i n e solution, pH 11.4, f o r 2 days. The part of the dimer not s p l i t by mercaptoethanol was r e l a t i v e l y stable. A small degree of cleavage res u l t e d when the dimer was treated with dioxan or a detergent s o l u t i o n i n d i c a t i n g that hydrophobic bonding i s of l i m i t e d importance i n holding the dimer together. The s t a b i l i t y of the mercaptoethanol r e s i s t a n t dimer at low and high pH values indicates that e l e c t r o s t a t i c bonds cannot be the explanation. Andersson suggested that both types of dimers are held together by d i s u l f i d e bonds. In one dimer the d i s u l f i d e bond i s sit u a t e d i n the i n t e r i o r of the molecule and therefore not accessible to react with mercaptoethanol. It was also proposed that hydrogen bonding may be responsible f o r the s t a b i l i t y of the dimer not s p l i t by mercaptoethanol (Andersson, 1966). 1.2 Driving forces f o r adsorption Before looking at the mechanism of protein adsorption one should look at the d r i v i n g forces f o r the process. For protein adsorption to be spontaneous the change i n free energy AG = AH - TAS , must be negative, where AH i s the enthalpy of ads ads ads ads adsorption, T the absolute temperature and AS i s the entropy of ads adsorption. Calorimetric measurements (Norde and Lyklema, 1978; Nyilas et al., 1974), give a d i r e c t measure of enthalpy and the data show that both e n t h a l p i c a l l y and e n t r o p i c a l l y driven adsorption occur since enthalpy changes ranged from p o s i t i v e to negative depending on pH f o r albumin 6 adsoroed on negatively charged polystyrene. Interactions such as covalent, e l e c t r o s t a t i c and hydrogen bonding between the pr o t e i n and the surface are l i k e l y to be exothermic while changes i n hydrophobic interactions, which are a r e s u l t of the ordering of water molecules near the surface of the protein or adsorbent, contribute to the changes i n entropy. Proteins have low s o l u b i l i t i e s which r a r e l y exceed 1% by weight due to high molecular weights. It has been shown (B u l l , 1956), that at the i s o e l e c t r i c point proteins usually display minimum s o l u b i l i t y and maximum adsorption. The s o l u b i l i t y of a protein i s determined by the balance of the a t t r a c t i o n of the protein molecules f o r each other, which tends to prevent s o l u t i o n and the a t t r a c t i o n of the solvent molecules f o r the protein, which tends to promote solution. At the i s o e l e c t r i c point the protein has a net neutral charge and the a t t r a c t i o n of the protein molecules f o r each other i s maximal. When the pH i s s h i f t e d away from the i s o e l e c t r i c point the protein molecule becomes charged. This decreases the a t t r a c t i o n of the pr o t e i n molecules f o r one another and leads to an increase i n s o l u b i l i t y since the proteins charged groups are more solvated. The s o l u b i l i t y and adsorption of proteins i s analogous to some aspects of synthetic polymer adsorption. An increase i n adsorption with decreasing s o l u b i l i t y has been shown f o r various polymers on glass (Rowland and E i r i c h , 1966). For synthetic polymers the amount adsorbed per unit area increases with increasing molecular weight ( G i l l i l a n d and Gutoff, 1960). From a mixture of polymers of varying molecular weight the larger molecules are adsorbed p r e f e r e n t i a l l y since they can form more bonds per molecule with the surface. 7 Proteins adsorb to non-biological surfaces due to t h e i r amphipathic nature, high molecular weight, limited s o l u b i l i t y and a b i l i t y to change con f i g u r a t i o n at an interface. The decrease i n free energy may re s u l t from a gain i n entropy due to the disorder of water released from the surface or prot e i n but i t may also be due to exothermic events. 1. 3 Mechanisms f o r adsorption Adsorption from an aqueous s o l u t i o n i s a competitive process, since, when p r o t e i n molecules are adsorbed solvent molecules are displaced. When determining the mode of adsorption a l l the in t e r a c t i o n s i n the system must be taken into account. When a protein s o l u t i o n flows past a s o l i d surface the protein reaches the surface by a diffusion-convection process, then binds. The i n i t i a l rate of adsorption depends on the transport and binding. Once protein has been adsorbed onto the surface the surface a v a i l a b i l i t y becomes the dominant f a c t o r and therefore r a t e - c o n t r o l l i n g and now protein-protein i n t e r a c t i o n s may become important. Proteins may bind to the surface v i a ion i c or e l e c t r o s t a t i c i n t e r a c t i o n s , hydrophobic interactions, hydrogen bonding and by charge-transfer or p a r t i a l donor-acceptor interactions. Covalent bonding does not r e s u l t from adsorption under b i o l o g i c a l conditions. Figure 1.1 depicts the d i f f e r e n t types of regions of a protein molecule that may be involved i n the adsorption process. Ionic or e l e c t r o s t a t i c interactions, due to the a t t r a c t i o n or repu l s i o n of two or more groups carrying a net charge, are important i n many systems. Proteins may bind to an oppositely charged surface v i a e l e c t r o s t a t i c bonds. However, a t t r a c t i v e e l e c t r o s t a t i c bonds may also be 8 formed between a protein carrying a net charge equal to that of the surface. For instance, i t has been shown that the adsorption of negatively charged proteins on a negative polystyrene latex occurs spontaneously and exothermically (Norde and Lyklema, 1978). SOLID—SOLUTION INTERFACE Fig. 1. 1. A schematic view of a protein interacting with a well-characterized surface. The protein has a number of surface domains with hydrophobic, charged and polar character. The solid may have a similar domain-like character. (Taken from Andrade, 1985, p.4). Charges on a protein surface are surrounded by unlike charges in a 9 d i f f u s e double layer rather than being only solvated by water (Wada and Nakamora, 1981). The s t a b i l i t y of a protein depends in some cases on intramolecular e l e c t r o s t a t i c i nteractions (Perutz, 1978). If a charged group important f o r pr o t e i n s t a b i l i t y i nteracts with a charged surface a conformational change may r e s u l t due to a change i n the e l e c t r o s t a t i c i n t e r a c t i o n s i n the protein. Proteins adsorbed e l e c t r o s t a t i c a l l y should be s e n s i t i v e to changes i n the i o n i c composition and pH. One would expect r e v e r s i b l e adsorption since continuous exchange of the protein with other ions i n the blood ought to occur. Protein adsorption i s thought to be i r r e v e r s i b l e or p a r t i a l l y r e v e r s i b l e on many surfaces, however, i n d i c a t i n g that e l e c t r o s t a t i c i n t e r a c t i o n s are only of minor importance. Another protein-surface i n t e r a c t i o n i s the hydrogen bond, a predominantly e l e c t r o s t a t i c i n t e r a c t i o n . The dipole/dipole i n t e r a c t i o n s may, i n an extreme case, give r i s e to i n t e r a c t i o n energies s i m i l a r to weak covalent bonds and due to the small s i z e of the hydrogen atom a small binding distance r e s u l t s . However, in proteins the binding energies are much smaller and the binding distance larger. Hydrogen bonds are important i n proteins and they contribute to the s t a b i l i t y of the inter n a l structure and s t a b i l i z a t i o n of the a-helix and /3-sheet structures. Again, competition from surface hydrogen bonding groups can cause conformational changes on binding. The hydrophobic patches on a protein can interact with hydrophobic polymer surfaces such as polyethylene or Teflon. The hydrophobic i n t e r a c t i o n i s an e n t r o p i c a l l y driven i n t e r a c t i o n r e s u l t i n g from a gain i n free energy caused by the loss of structured water at the hydrophobic 10 i n t e r f a c e when two such surfaces come together. This i s of great importance i n protein adsorption since the s t a b i l i t y and i n t e r a c t i o n s of proteins depend on the o v e r a l l free energy. The ordering of water at a i r or apolar i n t e r f a c e s i s e n t r o p i c a l l y undesirable. To keep these i n t e r f a c i a l areas at a minimum the hydrophobic amino a c i d side chains are excluded from the protein surface. Globular proteins have a hydrophobic core and a r e l a t i v e l y h ydrophilic s h e l l i n aqueous solutions but complete burying of the hydrophobic regions i s generally not possible. Intramolecular hydrophobic bonding i n dissol v e d proteins may a f f e c t p r o t e i n adsorption e s p e c i a l l y when intramolecular hydrophobic bonding i s required f o r the s t a b i l i z a t i o n of the protein structure. Rearrangement of str u c t u r e upon adsorption i s now probable ( B i r d i , 1973). Charge t r a n s f e r i n t e r a c t i o n s i n aqueous solutions are due to n-n e l e c t r o n e f f e c t s and these are important i n protein s t a b i l i z a t i o n and surface i n t e r a c t i o n . Excess e l e c t r o n density can be donated to an e l e c t r o p h i 1 i c species or e l e c t r o n density can serve as an acceptor f o r p o s i t i v e charge. 1.4 C h a r a c t e r i s t i c s of protein adsorption Many in v e s t i g a t o r s have found that the adsorption of proteins from s o l u t i o n to non-biological surfaces i s apparently of the Langmuir type ( B u l l , 1956; Oreskes and Singer, 1961; Cheng et al., 1978; Young et al.. 1988). The surface concentration increases asymptotically with an increase i n s o l u t i o n concentration u n t i l a steady state, plateau value i s reached. This i s assumed to be associated with the formation of a complete monolayer. The amount adsorbed i s not s i g n i f i c a n t l y d i f f e r e n t from that 11 expec';ed f o r a close-packed monolayer of native protein i n a side-on or end-on conformation depending on the system and conditions. Even though the adsorption isotherm i s of the Langmuir type there i s no reason to accept the a p p l i c a b i l i t y of t h i s model of adsorption because many of the Langmuir assumptions are not s a t i s f i e d , e.g., adsorption does not take place on s i t e s ; adsorption i s not f u l l y reversible. M u l t i l a y e r adsorption has been demonstrated (Oreskes and Singer, 1961; P i t t and Cooper, 1986; Young et al., 1988). Adsorption experiments were c a r r i e d out on various surfaces and the data plotted using modified versions of the Langmuir equation (Oreskes and Singer, 1961; Young et al., 1988). M u l t i l a y e r adsorption was suggested since the data were f i t t e d by two or more regions of d i f f e r e n t slopes, each slope representing a d i f f e r e n t binding constant. The f i r s t steeper slope was interpreted as representing the i n i t i a l p r o tein layer bound to the polymer. The second l i n e can be interpreted as a second layer of protein due to prot e i n - p r o t e i n int e r a c t i o n s or to a reorganization of the monolayer from a side-on to an end-on d i s p o s i t i o n thereby increasing the amount adsorbed. Protein adsorption has been shown to be pH dependent ( B u l l , 1956, Morrissey and Stromberg, 1974); see e a r l i e r discussion (Section 1.2). Maximum adsorption occurs near the i s o e l e c t r i c point of the protein. A predominantly nonionic, hydrophobic mechanism i s suggested since the prot e i n has no net charge at t h i s pH. Protein adsorption to various surfaces has been shown to depend on the surface. Table 1.1 shows the plateau values, following r i n s i n g with buffer, f o r the adsorption of albumin onto various hydrophobic and o hydrophilic surfaces from a 1 mg/ml solution, pH 7.4 at 23 C. 12 Table 1. 1 The adsorption of human serum albumin (HSA) on various surfaces. Surface Plateau surface concentration (pg/cm ) Ref. Polyurethane 1540 (hydrophilic) 0. 02 1 Polyurethane 600 (hydrophilic) 0. 04 1 Glass 0. 04 2 Si 1ica 0. t 09 3 Collagen-coated glass 0. 09 1 P o l y ( v i n y l chloride) 0. 17 4 S i l i c o n i z e d glass 0. 18 1 Polyethylene 0. 18 2 Polystyrene 0. 20 1 Polyurethane (hydrophobic) 0. 57 1 t surface concentration f o r bovine serum albumin Key to references: 1 Brash and Uniyal, 1979. 2 Brash and Davidson, 1976. 3 Morrissey and Stromberg, 1974. 4 Young et al., 1988. The h y d r o p h i l i c surfaces show low surface concentrations. On these surfaces desorption occurs, therefore there may be some uncertainty with respect to the surface concentration because r i n s i n g may remove adsorbed pr o t e i n (Chan and Brash, 1981; MacRitchie, 1972). The hydrophobic surfaces show a varying range of plateau values. Polyethylene, polystyrene and s i l i c o n i z e d glass show a steady state surface concentration of 13 approximately 0.2 ug/cm . These are a l l e f f e c t i v e l y hydrocarbon polymers. The hydrophobic polyurethane shows a higher plateau surface concentration. This polymer contains a high proportion of ether oxygen and urethane f u n c t i o n a l i t i e s as well as hydrocarbon. The surface has been shown to posses domains i.e. , polyether-rich and urethane-rich regions, which may be responsible f o r the high albumin adsorption. Many adsorption studies have been c a r r i e d out on tubing . Infrared i n t e r n a l r e f l e c t i o n techniques were used by Lee and Kim to study the e f f e c t of time and flow rate on the adsorption of serum albumin, y- g l o b u l i n and prothrombin to s i l i c o n e rubber (SR), f l u o r i n a t e d ethylene-propylene copolymer (FEP) and segmented polyether urethane (PEUU) (Lee and Kim, 1974). The adsorption was rapid and dependent on the substrate not on the protein. The plateau value concentration was shown to depend on the flow rate with SR but not with PEUU. Increasing the flow rate can delay the plateau time because of shear forces opposing the d i f f u s i o n of protein molecules to the surface. There was a s i x f o l d increase i n the plateau concentration f o r albumin on SR when the flow rate was increased from 0 to 12 ml/sec. This was explained i n terms of surface roughness. A rough surface such as SR f o r example, has a greater surface area, therefore more anchoring s i t e s are av a i l a b l e f o r the greater number of protein molecules i n the v i c i n i t y with an increased flow rate. Perhaps t h i s may be due to the formation of a thick adsorption-entanglement layer along the wall analogous to the ones observed i n flowing high-molecular weight polymer so l u t i o n s (Hikmet et al., 1985). FTIR-ATR ( f o u r i e r transform infrared spectroscopy coupled with attenuated t o t a l reflectance optics) studies on albumin adsorbed onto 14 polyeinaneurea showed that the adsorption k i n e t i c s were not s i g n i f i c a n t l y a f f e c t e d by the shear rate over a range 0-1800 s V However, protein adsorbed at 0 to 200 s 1 desorbed more r a p i d l y than that adsorbed at 1800 s 1. This suggests that the protein adsorbed under higher shear rates may bind more t i g h t l y ( P i t t and Cooper, 1986). The conformation and conformational changes upon the adsorption of plasma proteins may be a way of p r e d i c t i n g the e f f e c t of i n t e r a c t i o n s with the surface and have received considerable attention. Polymers are adsorbed to surfaces by the attachment of various segments along the chain which may occur s i n g l y or i n runs which have loops extending from the surface into the solution. S t a t i s t i c a l mechanics has been applied to independent polymer molecules adsorbed to planar surfaces and one can predict the f r a c t i o n of bound segments, the d i s t r i b u t i o n of the segments normal to the adsorbing surface and the average number of loops (Cohen Stuart et al., 1986). However, f o r adsorbed proteins such p r e d i c t i o n s are not possible due to the numerous intermolecular i n t e r a c t i o n s described e a r l i e r . Morrissey and Stromberg used in f r a r e d d i f f e r e n c e spectroscopy to study p r o t e i n adsorption on s i l i c a p a r t i c l e s . By observing a s h i f t of 20 cm 1 i n the amide I band upon adsorption, the f r a c t i o n of adsorbed protein carbonyl groups bound to the surface could be measured and used to c a l c u l a t e the number of surface attachments (Morrissey and Stromberg, 1974). The conformation of serum albumin, fibrinogen and prothrombin was studied as a f u n c t i o n of surface concentration, time of adsorption, pD and ioni c strength. There was no change i n the bound f r a c t i o n of albumin or prothrombin along the isotherm. Both these proteins have an average bound 15 f r a c t i o n of 0.11 i n d i c a t i n g about 80 carbonyl attachments to the surface but most of the molecule i s i n s o l u t i o n and away from the s o l i d surface. The bound f r a c t i o n does not change with the time of adsorption. This data suggests that conformational changes do not occur or are minimal upon adsorption, therefore i t may be s a i d that the internal bonding i n these proteins i s s u f f i c i e n t to prevent s t r u c t u r a l changes. The bound f r a c t i o n of fibrinogen was found to increase with increasing adsorption which may suggest i n t e r f a c i a l aggregation. Studies on cross-linked and denatured albumin showed that c r o s s - l i n k e d albumin gave a bound f r a c t i o n s i m i l a r to that of the native protein, while unfolded albumin gave an increase of 55 contacts and aggregated albumin re s u l t e d i n a decrease of 50 surface contacts. It i s concluded that no aggregation or conformational changes occur upon adsorption of the native serum albumin. Other investigators have also concluded by i n f r a r e d spectroscopic techniques that in general plasma proteins were not dimensionally denatured, i.e . , no change i n conformation occurred on adsorption to the surfaces studied (Brash and Lyman, 1969). It has been shown that adsorbed fibrinogen and j - g l o b u l i n are required f o r p l a t e l e t adhesion and are therefore important for surface induced thrombosis (Zucker and Vroman, 1969; Kim et al., 1974). A study of these from a protein mixture or plasma may help predict biocompatibi1ity. Radiolabelled proteins have to be used since spectroscopic techniques do not discriminate between d i f f e r e n t proteins. Lee and coworkers studied the competitive e f f e c t s of plasma proteins adsorbed to hydrophobic polymer surfaces (Lee et al. , 1974). The rates of 16 adsorption of I l a b e l l e d albumin, y-globulin and fibrinogen were measured separately and from a mixed solution. The composition of the adsorbed layer, following r i n s i n g , at the plateau value was determined. The amount of each protein adsorbed from a mixed s o l u t i o n was less than that compared to the adsorption from a s i n g l e protein solution. The time to reach the plateau concentration f o r albumin was doubled while that f o r f i b r i n o g e n and y- g l o b u l i n decreased by a f a c t o r of ten. Lyman and coworkers t r i e d to e s t a b l i s h a r e l a t i o n s h i p between the adsorbed protein composition and the extent of surface-induced thrombosis on PEUU, SR and FEP (Lyman et al., 1974). R e c i r c u l a t i o n tubes were implanted i n dogs f o r varying periods of time. The tubes were rinsed and then soaked i n a detergent to remove the adsorbed protein. The amount of albumin, ^ - g l o b u l i n and other globulins was determined using acrylamide gel electrophoresis. The r e s u l t s indicated that low thrombogenic surfaces adsorbed mainly albumin while thrombogenic materials adsorbed l a r g e l y globulins. 1.4.1 Protein desorption and exchange The r e v e r s i b i l i t y of protein adsorbed has to be established i n order to j u s t i f y using thermodynamic equations to describe the adsorbed phase. It i s also of in t e r e s t to e s t a b l i s h any changes in the composition of the pr o t e i n f i l m s adsorbed from mixtures. If protein adsorption i s assumed to occur at multiple s i t e s on the protein, desorption r e s u l t s only when a l l these s i t e s simultaneously detach. This would be expected to be a very improbable event and therefore one would expect s i g n i f i c a n t desorption not to occur. 17 The r e v e r s i b i l i t y of protein adsorption i s somewhat c o n t r o v e r s i a l . Adsorption data has frequently been found to be in good agreement with the Langmuir model ( B u l l , 1956; Brash and Lyman, 1969; Chuang et al., 1978; Young et al., 1988). This might be interpreted as supporting r e v e r s i b i l i t y i n terms of a dynamic equilibrium. However, there i s l i t t l e evidence that supports the Langmuir assumptions of r e v e r s i b l e binding to s i n g l e s i t e s per protein. Generally i t has been observed that s i g n i f i c a n t desorption from a hydrophobic surface does not occur, while f o r hydrophilic surfaces both r e v e r s i b i l i t y and i r r e v e r s i b i l i t y of the adsorbed protein has been found. Various methods have been employed in order to study the desorption of proteins from s o l i d surfaces and care must be taken to ensure that one measures the amount of protein a c t u a l l y desorbed. Bul l used s o l u t i o n - d e p l e t i o n methods to study the adsorption of bovine serum albumin on glass ( B u l l , 1956). The surface concentration i s determined by monitoring a change in the bulk s o l u t i o n concentration. In t h i s way the adsorption at equilibrium i s measured. The absorbance near 278 nm i s measured following e q u i l i b r a t i o n and c e n t r i f u g a t i o n of the suspended s o l i d . A large surface area to s o l u t i o n volume r a t i o was required to produce a s i g n i f i c a n t decrease in s o l u t i o n concentration. Desorption was studied by allowing two samples of pyrex glass to adsorb pr o t e i n at pH 5.05. The amount of protein adsorbed from a 0.0140% protein s o l u t i o n was 0.78 mg per gram of glass. One of the samples was then d i l u t e d with buffer to give an equilibrium protein concentration of 0.0065% and the amount of p r o t e i n adsorbed was found to be 0.77 mg per gram of glass. The amount of p r o t e i n adsorbed i n each case was found to be approximately the same. It was concluded that no protein was removed from 18 the glass surface by d i l u t i n g the protein solution. B u l l c a r r i e d out an experiment to determine i f protein could be removed from the surface by extensive washing. Glass powder was suspended in a protein s o l u t i o n and allowed to eq u i l i b r a t e . The pro t e i n s o l u t i o n was removed and the glass was resuspended i n buffer at selected pH values. The resuspension procedure was repeated f i v e times. The protein was removed from the glass by suspending i t in one molar sodium acetate and i t s concentration determined. The r e s u l t s show that a considerable amount of the adsorbed p r o t e i n i s removed by extensive washing. The removal i s pH dependent and i t i s more d i f f i c u l t to remove the protein at pH values near the i s o e l e c t r i c point of albumin. MacRitchie used s i m i l a r solution-depletion techniques and d i l u t i o n with buffer to study the desorption of bovine serum albumin from hydrophobic and hydr o p h i l i c s i l i c a p a r t i c l e s (MacRitchie, 1972). It was shown that at pH 7.5 albumin adsorption to hydrophilic s i l i c a was completely r e v e r s i b l e , but at the albumin i s o e l e c t r i c point, pH 4.9, adsorption was not r e v e r s i b l e . R e v e r s i b i l i t y was not observed with the hydrophobic s i l i c a surface. Brash and his coworkers studied the desorption and exchange of serum albumin on polyethylene and cuprophane (Brash et al, 1974). The polymers were i n the form of tubes. Radiolabelled serum albumin was pumped through the tubes. Af t e r a 24 hour r i n s i n g period the tube was assayed v i a gamma counting (gamma counted) to give the surface concentration. The adsorption and exchange of albumin adsorbed from a 0.1 mg/ml s o l u t i o n on polyethylene was studied. No desorption into water was detected and exchange of the r a d i o l a b e l l e d with nonlabelled albumin at 0.1 mg/ml s o l u t i o n concentration 19 d i d rot occur under s t a t i c conditions. Rapid desorption would not be detected i n t h i s way since the tubing i s washed before the "equilibrium" adsorption i s determined by gamma counting. Exchange of r a d i o l a b e l l e d p r o t e i n was detected at a so l u t i o n concentration of 3.7 mg/ml. The turnover was 10% i n the f i r s t hour and 85% i n 220 hours. Grant et al. and Stromberg et al. used rapid r i n s i n g techniques to study desorption (Grant et al., 1977; Stromberg et al., 1975). Adsorption was c a r r i e d out using r a d i o l a b e l l e d albumin. The substrate was removed and immersed i n a rin s e vessel and continuously washed with water. Desorption from polyethylene was not detected. Albumin adsorbed on chromium showed r e v e r s i b i l i t y with up to 25% of the protein being removed i n the f i r s t minute. Chuang et al. used polymer discs to study desorption and exchange (Chuang et al., 1978). Cuprophane and po l y ( v i n y l c h l o r i d e ) , (PVC), discs 125 were precoated with I-protein by incubating the disc i n a protein s o l u t i o n f o r 30 minutes. The discs were washed by dipping i n Tyrode's buffer. Desorption and exchange studies were c a r r i e d out by incubating the precoated d i s c s i n Tyrode's buffer or in homologous unlabel led proteins f o r 24 hours at room temperature. The residual r a d i o a c t i v i t y was gamma counted. It was demonstrated that desorption and exchange was dependent on both the s p e c i f i c p r o t e i n species and the type of polymer surface. Albumin adsorbed to cuprophane did not desorb into Tyrode's buffer but 38% of the r a d i o l a b e l l e d albumin was found to exchange with unlabel led albumin from a 1 mg/ml so l u t i o n . For fibrinogen adsorbed on cuprophane 23% desorbed into Tyrode's buffer and 42% exchanged with unlabelled fibrinogen at a f a s t e r 20 rate than f o r albumin under s i m i l a r conditions. I-IgG adsorbed on cuprophane showed a 19% desorption and gave an exchange of 30% with 125 unlabel led IgG, whereas f o r I _IgG adsorbed on PVC the desorption was 4% and exchange was 6%, both ocurring at a much slower rate. Double isotope l a b e l l i n g experiments have shown exchange of the adsorbed p r o t e i n with the protein i n s o l u t i o n even though the quantity of p r o t e i n adsorbed remains the same (Brash and Samak, 1978; Chan and Brash, 1981). In these experiments the polymer surface was rinsed with buffer before exchange runs were c a r r i e d out. The r e s u l t s indicate a constant exchange of p r o t e i n between the surface and s o l u t i o n . The l e v e l l i n g o f f o the p r o t e i n loss and gain curves suggest that there i s a f r a c t i o n of the adsorbed p r o t e i n that i s exchangeable and a f r a c t i o n that i s not and that t h i s v aries with conditions. The rates and extent of exchange have been shown to be greater f o r glass and hydrophilic surfaces (Chan and Brash, 1981) than f o r hydrophobic surfaces (Brash and Samak, 1978; Cheng et al., 1987). This would suggest stronger binding f o r hydrophobic surfaces. 1. 5 Objectives and methods In many studies on protein adsorption, and in a l l studies on adsorption to tubing, the protein layer i s washed before determining the "equilibrium" adsorbed protein concentration. Desorption studies are then c a r r i e d out on the remaining adsorbed layer. The problem with t h i s protocol i s that any weakly bound protein w i l l be washed off. It i s t h i s weakly bound p r o t e i n with which t h i s thesis i s concerned. Such material i of i n t e r e s t because weakly adsorbed macromolecules are known to play a 21 c e n t r a l r o l e i n some blood c e l l adherence phenomena (Brooks et al., 1980). The tubing form for the substrate i s important because i t i s t h i s form that i s u t i l i z e d i n many blood c o m p a t i b i l i t y applications. Desorption studies have also been c a r r i e d out using s o l u t i o n - d e p l e t i o n techniques. A s o l i d i s placed i n a protein s o l u t i o n and a protein layer i s adsorbed to the surface. The protein s o l u t i o n i s then d i l u t e d . A decrease i n surface concentration following d i l u t i o n would indicate desorption. Solution-depletion techniques have had lim i t e d a p p l i c a t i o n since a f i n e l y d i vided substrate has to be used and there may be some uncertainty i n determining the av a i l a b l e surface area per gram of material. Also, the geometry of the substrate has been shown to af f e c t the amount of pro t e i n adsorbed (Oreskes and Singer, 1961) and i t i s not possible, i n general, to produce dispersions and tubular geometries of the same material with the same surface properties. In the adsorption experiments of t h i s study, BSA was adsorbed to a length of polyethylene tubing. Following e q u i l i b r a t i o n of the r a d i o l a b e l l e d BSA with the surface the protein s o l u t i o n was displaced with buffer. A l l the f r a c t i o n s c o l l e c t e d were gamma counted. In an experiment of t h i s type the surface concentration can be calculated by a v a r i e t y of methods. F i r s t l y , a minimum surface concentration i s determined by c u t t i n g up the polyethylene tubing following r i n s i n g and gamma counting. In t h i s case any r e v e r s i b l y adsorbed protein w i l l have been washed o f f . The amount of p r o t e i n displaced from the tube, taking into account the d i l u t i o n e f f e c t , can be calculated from the a c t i v i t y of the c o l l e c t e d samples. From the t o t a l a c t i v i t y added and the a c t i v i t y of the protein displaced the surface concentration i n the tube can be calculated. This s o l u t i o n 22 depletion value gives the maximum amount of protein adsorbed at e q u i l i b r i u m and the difference between these two values, i f any, represents the loosely bound protein. 23 CHAPTER 2 EXPERIMENTAL 2. 1 Rad i o1abe11i ng R a d i o l a b e l l i n g of bovine serum albumin (BSA) was c a r r i e d out using iodo-beads, a commercial s o l i d state reagent (Markwell, 1982). The iodo-beads have N-chloro-benzenesulfonamide, an oxidant, Immobilized on 2.8 mm diameter non-porous polystyrene spheres. The i o d i n a t i o n involves the oxidation of the radioiodide which then reacts with tyrosine (4-hydroxyphenylalanine) residues of the protein by the e l e c t r o p h i 1 i c s u b s t i t u t i o n of the ortho hydrogens on the phenolic r i n g (Regoeczi, 1984). Bovine serum albumin (Fraction V, code no. 81-003, Miles S c i e n t i f i c , Rexdale, Ont.), the protein to be iodinated, was dissolved i n phosphate buffered s a l i n e (PBS)/azide, pH 7.4, to give a concentration of 1 to 4 mg/ml. Isotonic PBS/azide pH 7.4 consisted of Na2HP0^ 2.367 g/1, NaH2P0Ii 0.400 g/1, NaN3 0.195 g/1 and NaCl 7.621 g/1. The iodo-beads (Pierce Chemical Company, Rockford, 111.) with an oxidative capacity of 0.45 umol/bead f o r tyrosine-containing peptides, were washed twice i n PBS/azide and blotted dry on f i l t e r paper. To a 1.5 ml Eppendorf micro test tube 2-4 iodo-beads and 0.5 ml of the protein s o l u t i o n were added. The r e a c t i o n was 125 i n i t i a t e d by the add i t i o n of 10-20 ul (200-400 pCi) of c a r r i e r free Na I (Amersham, Arli n g t o n Heights, 111.). The capped tube was rotated f o r 30 O minutes at room temperature (19 C) a f t e r which the reaction was monitered by a t r i c h l o r o a c e t i c a c i d (TCA) p r e c i p i t a t i o n assay to determine the amount 24 of free label hence the completeness of the reaction. 2.1.1 TCA p r e c i p i t a t i o n assay A small sample, 1 u l , of the reaction mixture was added to 1 ml of a 1 mg/ml BSA s o l u t i o n i n a polypropylene test tube. To t h i s 1 ml of 0.5 M t r i c h l o r o a c e t i c a c i d (TCA) s o l u t i o n was added to p r e c i p i t a t e the protein. The sample was centrifuged at 4500 x g f o r 10 minutes and 1 ml of the supernatant was pipetted into a second tube and both samples were counted i n a LKB-Wallac 1282 Compu Gamma gamma counter. The r a d i o l a b e l l e d protein 125 p e l l e t e d with the albumin while the free I was d i s t r i b u t e d evenly between the p e l l e t and the supernatant. The percent of pro t e i n bound was calc u l a t e d using the following equation P e l l e t - Supernatant % Bound = Pel l e t + Supernatant 2.1.2 Gel f i l t r a t i o n Following i o d i n a t i o n the free label was separated from the ra d i o l a b e l l e d p r o t e i n by gel f i l t r a t i o n . A Bio-Rad column (1 x 20 cm) was packed with Sephadex G-25 (Pharmacia, Uppsala, Sweeden) and e q u i l i b r a t e d with PBS/azide at pH 7.4 . Radiolabel free BSA, (2 ml at 1 mg/ml) was put on the column f i r s t to reduce binding of the l a b e l l e d protein to the gel. The r a d i o l a b e l l e d protein was loaded onto the column and eluted with PBS/azide buffer pH 7.4 and the eluate was c o l l e c t e d i n 20, 25 or 30 drop f r a c t i o n s . The f r a c t i o n s were sampled with 1 ul Drummond micro c a p i l l a r i e s (Fisher S c i e n t i f i c ) and gamma counted. The appropriate f r a c t i o n s were 25 pooled and before use the samples were eit h e r put through a second Sephadex G-25 column or dialysed against PBS/azide using an u l t r a f i l t r a t i o n unit (molecular weight cut o f f = 10,000, M i l l i p o r e Ltd., Mississauga, Ont.) to remove more free label. The sample was s p l i t into convenient aliquots and o stored at -20 C. The amount of free label was checked by TCA p r e c i p i t a t i o n and instant t h i n layer chromatography (TLC). 2.1.3 Thin-layer chromatography A small amount, 1 u l , of the r a d i o l a b e l l e d protein was added near the base of a 10 x 1.5 cm s t r i p of p o l y s i l i c i c a c i d gel chromatography media (Gelman Sciences Inc., Ann Arbor, Mi.). This was put into a chamber and developed with 1:1 (v/v) acetone:methanol. The s t r i p was a i r - d r i e d , cut up into 1 cm sections perpendicular to the d i r e c t i o n of migration and placed i n gamma tubes containing 2 ml of 10 mM NaOH and counted. The amount of free label can be calculated since the free -label migrates up the s t r i p and the r a d i o l a b e l l e d protein remains where spotted. 2.2 Protein Electrophoresis The p u r i t y of the r a d i o l a b e l l e d BSA was determined using polyacrylamide gel electrophoresis i n sodium dodecyl s u l f a t e (SDS-PAGE), (Ornstein 1964, Davis 1964). The mobility of a protein i n a polyacrylamide gel i s governed mainly by the protein molecular weight (Sharpiro et al., 1967; Weber and Osborn, 1969). In the presence of SDS, a l l proteins whatever t h e i r o r i g i n a l charge, are converted to complexes having strong negative charges. This causes them to behave as rods of constant diameter. Electrophoresis c a r r i e d out in gels with pores small enough to r e s t r i c t 26 m o b i l i t y shows that the observed mobility i s r e l a t e d very nearly l i n e a r l y to the log of the molecular weight of the protein. This i s unaffected by the proteins o r i g i n a l charge. The p r o t e i n sample i s layered on the polyacrylamide gel and a voltage gradient i s applied. The macromolecules migrate at d i f f e r e n t (constant) rates i n the gel and t h e i r location i n the gel i s determined a f t e r the experiment by s t a i n i n g with Coomassie Blue, a c a t i o n i c dye that binds mainly to amines (Fazekas et a l . , 1963), or by gel s l i c i n g and counting the s l i c e s f o r gamma radiation. 2.2.1 Materials The following were of electrophoresis p u r i t y from Bio-Rad laboratories, Richmond, CA.: acrylamide, N,N-methylene-bis-acrylamide (BIS), sodium dodecyl sulphate (SDS), N,N,N',N'-tetramethylenediamine (TEMED) and ammonium persulphate. The disodium ethylenediamine-tetraacetate (EDTA) was from Fisher S c i e n t i f i c Company, F a i r Lawn, N.J.. Tris(hydroxymethyl)aminomethane (tris-base) and Tris(hydroxymethyl)aminomethane hydrochloride ( t r i s - H C l ) were obtained from Sigma Chemical Company, St.Louis Mo.. Sucrose was from Baker and Adamson, Morristown, N.J.. Pyronin Y (C.I. 45005) was from J.T. Baker Chemical Company, P h i l l i p s b u r g N.J.. Coomassie B r i l l i a n t Blue G-250 (42655) and Photo-Flo 200 s o l u t i o n were obtained from Eastman Kodak Co. Rochester N.Y.. 2.2.2 Sample preparation The samples for electrophoresis contained 1 mg/ml of protein or had approximately 2000 cpm/ul. 27 following an adsorption experiment the tubing was placed in a hot water bath (80 C) f o r 5 minutes. The s o l u t i o n in the tube was displaced with a hot a l k a l i n e s o l u t i o n of SDS (4%). The hot SDS s o l u t i o n was allowed to s i t f o r 5 minutes and was then displaced. This procedure was repeated u n t i l the counts coming o f f the tube were n e g l i g i b l e . The displaced so l u t i o n s were pooled and concentrated using a M i l l i p o r e u l t r a f i l t r a t i o n unit ( M i l l i p o r e Corporation, Bedford, Mass.). The u l t r a f i l t r a t i o n unit (10,000 molecular weight cut o f f ) connected v i a s i l i c o n e tubing to a 50 ml syringe was lowered into the sample to be concentrated in a 15 ml polypropylene tube. A vacuum was applied using the syringe which allowed the f i l t r a t e to pass through the membrane into the syringe. The r a d i o l a b e l l e d BSA remained in the tube.The sample was then used f o r SDS-PAGE. 2.2.3 Method The procedure was a modification of that used by Fairbanks (Fairbanks et a l . , 1971). The solutions and concentrated stock solutions were mixed i n the order and proportions given i n Table 2.1. The 3.75% gel system was prepared and cast as rod gels in acid cleaned 125 x 7 mm inside diameter glass tubes, 2 ml gel s o l u t i o n was used per tube. The gels were overlayed with buffer to produce a f l a t surface on the gel on which to layer the sample. When polymerization was complete, 40 minutes at room temperature or overnight, the gels were mounted in a Bio-Rad model IS0A electrophoresis chamber and the overlay s o l u t i o n was flushed away.with f r e s h r e s e r v o i r buffer. The samples were mixed 1:1 (v/v) with the sample reagent and 20 ul 28 Table 2.1 Composition of buffers and 3.75% gels f o r SDS-PAGE. 10 X Acrvlamide-Bis Acrylamide 40.0 g Bis 1.5 g Water to 100 ml Fairbanks Reservoir Buffer 10 X Buffer 200 ml 4 % SDS 100 ml Water 1700 ml Fairbanks Gel System Fairbanks Sample Reagent 10 X Acrylamide-bis 10 X Buffer Water 4 % SDS 0.5 % TEMED 1.5 % Ammonium persulphate 3.0 ml 3.0 ml 18.0 ml 1.5 ml 1.5 ml 3.0 ml Tris-HCl EDTA SDS Sucrose Pyronin Y Water to 50 ml 0.1211 g 0.0372 g 1.0 g 7.0 g Overlay Solution The same as the gel s o l u t i o n with the acrylamide-bis replaced with water Fairbanks 10 X Buffer Tris-base 24.23 g Sodium acetate 13.61 g EDTA 3. 72 g Water to 500 ml pH to 7.4 with g l a c i a l a c e t i c a c i d a l i q u o t s were layered on the gel surface with the displacement pipettor. For the protein washed o f f the tubing 60 ul-was used to provide s u f f i c i e n t r a d i o a c t i v i t y . Avoltage was applied across the gels at a constant current of 0.5 mA per tube u n t i l the sample entered the gel and then the current o was increased to 6 mA per tube. The gels were run at 4 C f o r approximately 2 hours u n t i l the tracer dye was near the bottom of the gels. A syringe was f i l l e d with water and a few drops of g l y c e r i n and a hypodermic needle was attached. The gels were removed from the glass tubes which were rotated as the hypodermic needle was pushed between the tube and the gel while expressing a f i n e stream of the g l y c e r i n solution. The 29 p o s i t i o n of the tracking dye was marked in each gel by p r i c k i n g i t with a needle dipped i n Indian ink. 2.2.4 Gel s t a i n i n g The gels were stained f o r protein with coomassie blue f o r 1-2 hours. Coomassie blue s t a i n consisted of coomassie blue G-250 0.20 g, methanol 28 ml, g l a c i a l a c e t i c a c i d 5 ml, perc h l o r i c acid (70%) 25 ml, water to 500 ml. Destained with methanol f i x (methanol 300 ml, g l a c i a l a c e t i c a c i d 50 ml, water 650 ml) f o r one hour and the f i n a l c l e a r i n g i n 7% g l a c i a l a c e t i c a c i d overnight u n t i l the background was clear. The s t a i n i n g was repeated i f necessary. Densitometry was performed with a Auto Scanner F l u r - V i s equipped with a 595 nm f i l t e r and a zig-zag time base integrator. Gels run with r a d i o l a b e l l e d samples were s l i c e d using a Bio-Rad model 195 e l e c t r i c gel s l i c e r with a 1 mm blade-to-blade separation. The s l i c e s were t r a n s f e r r e d to gamma tubes and counted f o r radi a t i o n . 2.3 Protein Concentration Proteins show strong absorption at a wavelength of approximately 280 nm due to the residues of phenylalanine, tyrosine and tryptophane. The prot e i n concentration can be determined by measuring the o p t i c a l density (OD) at about 280 nm i f the molar e x t i n c t i o n c o e f f i c i e n t i s known. The e x t i n c t i o n c o e f f i c i e n t of the BSA sample used at 278 nm was determined. BSA (5 g) was freeze dried f o r 24 hours and then dri e d over phosporous pentoxide f o r 3 days u n t i l no change in weight was observed. A 10 mg/ml stock s o l u t i o n of BSA i n PBS/azide was prepared. A s e r i e s of s e r i a l d i l u t i o n s were c a r r i e d out and the 0D of the solutions at 278 nm 30 were .neasured and the e x t i n c t i o n c o e f f i c i e n t calculated. The concentrations of r a d i o l a b e l l e d protein solutions were determined by measuring the OD at 278 nm. This was found to be more r e l i a b l e and consistent than the colourimetric method used by Smith et al.,1985. 2.4. Adsorption Experiments The aim of the work f o r t h i s thesis was to develop a technique that would detect, i f any, the r e v e r s i b l y adsorbed protein bound to a p l a s t i c surface. The technique developed was to use a long tube of small diameter that would give a large surface area. The r a d i o l a b e l l e d protein was pumped into the tube, l e f t to e q u i l i b r a t e and pumped out. By c o l l e c t i n g a l l f r a c t i o n s , the amount of pr o t e i n i s calculated. The experiments were performed to obtain the surface concentration as a function of time of contact, s p e c i f i c a c t i v i t y , pH and concentration. 2.4.1 Methods and Materials 125 The BSA was l a b e l l e d with I using iodo-beads; the amount of free label was less than 1% as checked by TLC. The average degree of i o d i n a t i o n 125 was less than one I atom per molecule of protein and t h i s degree of s u b s t i t u t i o n has been shown to leave the protein properties b i o l o g i c a l l y unaltered, (McFarlane, 1963; Harwig et al., 1975). The polyethylene tubing of inside diameter 0.038 cm was obtained from Intramedic. This material i s intended f or c l i n i c a l use and i s made from low density polyethylene. It alle g e d l y contains no additives or p l a s t i c i z e r s . The tubing was prepared f o r adsorption experiments by 31 pumping methanol and then d i s t i l l e d water through the tubing. Solutions of BSA were made up in PBS/azide, pH 7.4, to the desired s p e c i f i c a c t i v i t y and concentration. Protein concentrations were ca l c u l a t e d by measuring the OD at 278 nm or from a known s p e c i f i c a c t i v i t y . A 5 m length of tubing was connected v i a a three-way valve to a syringe on a Harvard pump. The valve allowed the removal of a i r bubbles through the side arm before introducing protein s o l u t i o n or buffer. Adsorption runs were c a r r i e d out by f i l l i n g the e n t i r e system with PBS/azide buffer, d i s p l a c i n g with protein s o l u t i o n and c o l l e c t i n g a l l f r a c t i o n s using a Gil s o n micro-fraction c o l l e c t o r . The protein s o l u t i o n O was l e f t to e q u i l i b r a t e f o r 4 hours at room temperature (23 C), except i n the case of the time dependence experiment. The contents of the tube were displaced and the tube rinsed with PBS/azide and 3-drop f r a c t i o n s were c o l l e c t e d . A chart recorder was connected to the f r a c t i o n c o l l e c t o r to provide a time base and enable the volume of the samples to be determined. The tubing was f i n a l l y cut into 20 cm segments. A l l the f r a c t i o n s and tubing were counted f o r gamma radiation. 2.4.2 Surface concentration as a function of time To determine the time f o r the system to reach steady state the e q u i l i b r a t i o n time was varied from 0.5 to 24 hours. One stock protein s o l u t i o n was used f o r the serie s of experiments to ensure a constant concentrat ion. 32 2.4.3 Surface concentration as a function of the r a t i o of r a d i o l a b e l l e d to unlabel led BSA The e f f e c t of the r a t i o of radiolabelled:unlabel led protein on the surface concentration was determined. Using a saturating concentration of 12S BSA, 500 ug/ml, the r a t i o of I-BSA to unlabel led BSA was varied, a se r i e s of experiments were c a r r i e d out and the surface concentrations calculated. 2.4.4 Adsorption Isotherm Adsorption was studied as a function of concentration. Protein s o l u t i o n s were prepared at various concentrations with a s p e c i f i c a c t i v i t y 4 of approximately 2x10 cpm/ug. Protein concentrations were determined from the OD or from a known s p e c i f i c a c t i v i t y . 2.4.5 Surface concentration as a function of pH A s e r i e s of protein solutions was prepared at d i f f e r e n t pH values by mixing r a d i o l a b e l l e d and unlabelled BSA with NaOH or HC1 at a f i n a l p r otein concentration of 500 ug/ml. A stock s o l u t i o n of BSA was adjusted to the appropriate pH using a pHM63 D i g i t a l pH meter. Approximately 1.5 ml of the stock BSA was weighed and a small amount of r a d i o l a b e l l e d BSA was added. The pH was checked using pH paper since the s o l u t i o n contained r a d i o l a b e l l e d BSA. 33 CHAPTER 3 RESULTS AND DISCUSSION 3. 1 Radiolabel1ing A plot of the r a d i o a c t i v i t y (counts per minute (cpm)) from a 1 ul sample of each f r a c t i o n against the f r a c t i o n c o l l e c t e d from a Sephadex G-25 column (June 23, 1987) following radiolabel1ing i s shown i n Figure 3.1. The f i r s t peak corresponds to the r a d i o l a b e l l e d protein. The plot shows that the r a d i o l a b e l l e d BSA and the free iodide f r a c t i o n s are well separated. In t h i s example a TCA p r e c i p i t a t i o n assay of the pooled f r a c t i o n s (12 to 14) showed that 6% of the a c t i v i t y was due to the free l a b e l . The amount of free label was reduced to less than 1%, as tested by TLC, by a second column or by d i a l y s i n g against PBS/azide when a more concentrated s o l u t i o n was required. 3.1.1 Degree of Radiolabel1ing The degree of radiolabel1ing was found to be less than one molecule of 125 I per molecule of BSA and t h i s has been reported to have no e f f e c t on the b i o l o g i c a l a c t i v i t y , (McFarlane, 1963; Harwig et al. , 1975). For a t y p i c a l l a b e l l i n g experiment the degree of r a d i o l a b e l 1 i n g i s 125 c a l c u l a t e d as follows. The Na I on June 1, 1987 had an a c t i v i t y of 16.6 125 mCi/ug of iodide. The atomic weight of I i s 126.9 g/mole and 9 125 18 1 mCi = 2.2 x 10 dpm, therefore the a c t i v i t y of the I was 4.63 x 10 125 dpm/mole. The dpm of the I on June 24, 1987 i s calculated from the 34 20 Fraction number Fig. 3.1. Radioactivity vs fraction number for the samples collected from a Sephadex G—25 column following radiolabelling. following equation A = A 0e (3.1) where A 0 and A are the a c t i v i t i e s at times t 0 and t r e s p e c t i v e l y and A i s the decay constant which i s given by A = l n 2 ( t / 2 ) _ 1 (3.2) where t/2 i s the h a l f - l i f e . Substituting into equations 3.2 and 3.3 using 1 PS 12B 60 days as the ha l f l i f e of I, on June 24, 1987 the a c t i v i t y of I was 18 3.55 x 10 dpm/mole. The a c t i v i t y of the r a d i o l a b e l l e d BSA on June 24, 1987 was 196,000 cpm/pg or assuming 77% e f f i c i e n c y of the gamma counter (Janzen, 1985) 255,000 dpm/ug. Using 66,000 as the molecular weight f o r BSA i t s a c t i v i t y 16 was about 1.68 x 10 dpm/mole. Dividing the two a c t i v i t i e s gives a degree 125 of r a d i o l a b e l l i n g of 211 moles of BSA per mole of I. 3.2 E x t i n c t i o n c o e f f i c i e n t The o p t i c a l d e n s i t i e s of a s e r i e s of solutions of known BSA concentration were measured. A plot of BSA concentration against 0D at 278 nm i s given i n Figure 3.2. The plot indicates a l i n e a r r e l a t i o n s h i p between the 0D and a concentration of 0-1.63 mg/ml. The e x t i n c t i o n c o e f f i c i e n t i s given by the slope and i s calculated to be 0.641 ± 0.002. This value i s s i m i l a r to the BSA e x t i n c t i o n c o e f f i c i e n t of 0.66 given i n the l i t e r a t u r e (Cohn et al., 1947). 36 Concentration (mg/ml) Fig. 3.2. Calibration curve of BSA. Optical density vs BSA concentration at 278nm. 3.3 SDS-PAGE The p u r i t y of the BSA used i n the adsorption experiments was determined using SDS-PAGE. The gels were c a l i b r a t e d using Pharmacia elect r o p h o r e s i s c a l i b r a t i o n k i t s . In both the high molecular weight k i t (HMWK) and the low molecular weight k i t (LMWK) several SDS-denatured proteins were run on the same gel. The r e l a t i v e mobility (FLJ of a protein i s c a l c u l a t e d as distance of protein migration Relative mobility = distance t r a v e l l e d by tracking dye The R^ , values f o r the proteins of known molecular weights are given i n Table 3.1. The r e l a t i v e m o b i l i t i e s were plotted against the known molecular weights expressed on a semi-logarithmic scale. The plot i n Figure 3.3 indicates a l i n e a r r e l a t i o n s h i p and provides the molecular weight c a l i b r a t i o n f o r the gels. The SDS-PAGE rod gels of the stock unlabel led BSA and r a d i o l a b e l l e d BSA along with the molecular weight standards are shown i n Figure 3.4. A densitometric scan of the unlabel led BSA and the r e l a t i v e mobility versus cpm f o r r a d i o l a b e l l e d BSA are given i n Figure 3.5. The unlabel led BSA showed bands with apparent molecular weights of 60,000, 133,000 and 214,000 corresponding to R^ , values of 0.540, 0.371 and 0.270. The R^ , values of the r a d i o l a b e l l e d BSA at 0.539, 0.371 and 0.264 likewise imply molecular weights of approximately 60,000, 134,000 and 220,000. The more intense band with a molecular weight of approximately 60,000 corresponds to the BSA monomer while the less intense bands at 134K and 220K presumably correspond 38 Table 3.1 Molecular weight assignments f o r the protein standards used on the 3.75% SDS-PAGE gels. P r o t e i n Molecular weight R f Thyroglobulin 330,000 0. 165 F e r r i t i n (half unit) 220,000 0. 311 Phosphorylase b 94,000 0. 420 Albumin (BSA, HMWK) 67,000 0. 495 Albumin (BSA, LMWK) 67,000 0. 506 Catalase 60,000 0. 540 Ovalbumin 43,000 0. 607 Lactate dehydrogenase 36,000 0. 653 Carbonic anhydrase 30,000 0. 702 Trypsin i n h i b i t o r 20,000 0. 771 F e r r i t i n 18,500 0. 786 oc-Lactalbumin 14,400 0. 831 to the dimer and trimer. The 220K band i s less intense than the 134K band. The BSA molecular weight determined from the gels i s lower than the actual value of 66,000. This may have been due to the wide bands of stained p r o t e i n i n the standards. Since the R^ . values f o r both the l a b e l l e d and the r a d i o l a b e l l e d BSA are s i m i l a r i t i s concluded that r a d i o l a b e l 1 i n g does not e f f e c t the mobility of the protein. The s p e c i f i c a c t i v i t y f o r the oligomeric species have been ca l c u l a t e d to be approximately the same as the monomer. The nature of the BSA dissolved i n solutions at a v a r i e t y of pH values 39 100 1 -I 1 1 1 1 1 1 1 1 1 1 0.0 0.2 0.4 0.6 0.8 1.0 Rf Fig. 3.3. The molecular weight on a semi—log scale is plotted against the relative mobility (Rf) for a variety of SDS—protein complexes run on 3.75* gels. 40 Molecular we i ght 94,000 67,000 14,400 A B C D Fig. 3.4. SDS-PAGE rod gels stained with coomassie blue A low molecular weight standards B high molecular weight standards C unlabel led stock BSA D r a d i o l a b e l l e d BSA 41 1.0 E c 0.8 -LO cn LO -•+-> O C D 0.6 -O C o — _Q t O 0.4 -to D -> -\—> O 0.2 -C D 0.0 I 1 r I1 I1 I1 I1 1 Ii , TD \ • , , i 0.0 0.2 0.4 0.6 0.8 Relative mobility 1.0 M . 5 T 2.5 2.0 h0.5 0.0 X E Q_ O Fig. 3.5. The mobility of BSA in 3.75* SDS-PAGE gels: ( ) densitometric scan of BSA. The protein was coomassie blue stained and scanned at 595 nm. TD = tracking dye. ( ) relative mobility vs cpm for radiolabelled BSA. 42 was investigated by electrophoresis. The samples of the gels p l o t t e d i n Figures 3.6a-d were from the stock solutions in the adsorption experiments. The amounts of monomer and polymer i n the BSA samples were c a l c u l a t e d from the r a d i o a c t i v i t y of the gel s l i c e s and the data i s presented i n Table 3.2. It i s apparent from the plo t s that as the pH increases there i s a s i g n i f i c a n t decrease i n the amount of monomer present. Table 3.2 The e f f e c t of pH on the amount of BSA polymer present i n the stock solutions. The amount of polymer i s represented as a percentage of the t o t a l p r o t e i n ± error. The error was calculated from the a c t i v i t y ; see Appendix 2. pH % monomer % dimer % trimer % tetramer % pentamer 4. 4 83. 36 + 0. 72 11. 42 + 0. 21 2. 68 + 0. 13 1. 31 + 0. 08 7. 4 83. 16 + 0. 48 11. 87 + 0. 14 2. 93 + 0. 08 1. 54 + 0. 08 9. 4 71. 61 + 0. 45 19. 42 + 0. 20 5. 17 + 0. 10 1. 94 + 0. 06 1. 05 ± 0. 06 12. 0 52. 96 + 0. 31 25. 53 + 0. 22 10. 92 + 0. 14 4. 95 + 0. 10 2. 01 ± 0. 05 It i s of in t e r e s t to e s t a b l i s h i f the BSA i s a l t e r e d upon adsorption to the polyethylene tubing. An adsorption experiment was run at pH 7.4 and SDS-PAGE was c a r r i e d out on, ( i ) the stock r a d i o l a b e l l e d BSA i . e . , the r a d i o l a b e l l e d BSA before adsorption, ( i i ) the r a d i o l a b e l l e d BSA pumped out of the tubing during input of the BSA, ( i i i ) the BSA pumped out of the tubing a f t e r the four hour e q u i l i b r a t i o n time and (iv) the BSA washed o f f the tubing using hot basic SDS. Plots of the gels are shown i n Figures 43 0.4 0.0 0.2 0.4 0.6 0.8 1.0 Relative mobility Fig. 3.6a. Relative mobility vs relative radioactivity of BSA, pH 7.4. 0.0 0.2 0.4 0.6 0.8 1.0 Relative mobility Fig. 3.6b. Relative mobility vs relative radioactivity of BSA, pH 4.4. 44 0.0 0.2 0.4 0.6 0.8 1.0 Relative mobility Fig. 3.6c. Relative mobility vs relative radioactivity of BSA, pH 9.4. 0.15 Relative mobility Fig. 3.6d. Relative mobility vs relative radioactivity of BSA, pH 12.0. 45 3.7a-d. The amount of monomer and polymer i n each sample i s given i n Table 3.3. Table 3.3 The amount of BSA polymer present i n various samples represented as a percentage of the t o t a l protein ± error. Sample % monomer % dimer % trimer % tetramer ( i ) 77. 91 + 0. 54 14. 83 + 0. 19 3. 47 + 0. 10 1. 86 + 0. 08 ( i i ) 78. 21 + 0. 66 14. 21 + 0. 23 3. 51 + 0. 12 1. 54 + 0. 10 ( i i i ) 79. 87 + 0. 48 14. 15 + 0. 16 3. 34 + 0. 08 1. 41 + 0. 06 (iv) 65. 91 + 0. 60 17. 68 + 0. 26 9. 45 + 0. 19 The data i n Table 3.3 shows that there i s no difference between samples i , i i and i i i but that the BSA washed o f f the tubing (iv) contains less monomer. The bands i n Figure 3.7d are wider since a larger sample volume (60 ul) was used to obtain a s i g n i f i c a n t level of r a d i o a c t i v i t y . 20 u l samples were loaded on the other gels. Assuming that the BSA washed o f f the tubing i s representative of the BSA adsorbed then one may conclude that less monomer i s adsorbed. It i s seen that more of the higher molecular weight dimer and trimer are p r e f e r e n t i a l l y bound. The r a d i o a c t i v i t y at higher R^ , values may be due to low molecular weight fragments r e s u l t i n g from the harsh treatment with hot basic SDS. It was possible to remove more than 90% of the adsorbed BSA with hot basic SDS as indicated by counting the tubing f o r r a d i o a c t i v i t y following the washing procedure. 46 0.6 0.5 H > 0.4 H •4-» o I 0.3 H o -*-» I 0.1 H 0.0 A 0.0 0.2 0.4 0.6 0.8 1.0 Relative mobility Fig. 3.7a. Relative mobility vs relative radioactivity of sample (i). 0.4 ^0.3 H *> o 1 0.2 H 0> > 0.1 H or 0.0 0.0 0.2 0.4 0.6 0.8 1.0 Relative mobility Fig. 3.7b. Relative mobility vs relative radioactivity of sample (ii). 47 0.4 0.0 0.2 0.4 0.6 0.8 1.0 Relative mobility Fig. 3.7c. Relative mobility vs relative radioactivity of sample (iii). 0.20 *> '•+-> o I 0.10 D > or: 0.00 T 0.0 0.2 0.4 0.6 0.8 1.0 Relative mobility Fig. 3.7d. Relative mobility vs relative radioactivity of sample (iv). 48 3. 4 Adsorption Experiments The surface concentration (weight/unit area) of albumin was ca lcula ted by comparing the s p e c i f i c a c t i v i t y (cpm/ug) of BSA with the a c t i v i t y of a known surface area and i s computed using equation 3.3 SA 4 (3.3) where 2 T = surface concentration (ug/cm ) C = counts per minute from the tubing (cpm) SA = s p e c i f i c a c t i v i t y (cpm/ug) 2 4 = surface area (cm ) 3.4.1 Surface concentration as a function of the r a t i o of r a d i o l a b e l l e d to unlabel led BSA As a preliminary to studying adsorption i t was necessary to e s t a b l i s h whether the r e l a t i v e amounts of l a b e l l e d and unlabelled p r o t e i n had any e f f e c t on the adsorption. The adsorption experiments were c a r r i e d out using a s e r i e s of s o l u t i o n s with a concentration of 0.5 mg/ml but the r a t i o of l a b e l l e d to unlabelled protein was varied. The e q u i l i b r a t i o n time was 4 o hours at 23 C. The r e s u l t s are given i n Table 3.4, (assuming the molecular weight of BSA to be 66,000). The r e s u l t s indicate that the surface concentration i s e s s e n t i a l l y independent of the r a t i o of l a b e l l e d to unlabelled protein. Therefore one can conclude that i o d i n a t i o n at a tracer 125 l e v e l of less than one I atom per protein molecule does not a f f e c t the proteins' a f f i n i t y f o r the polyethylene surface. 49 TABLE 3.4 The e f f e c t of l a b e l l e d BSA content on the adsorption to polyethylene from a 0.5 mg/ml so l u t i o n , (see Appendix 2 f o r the c a l c u l a t i o n of AD. Moles of unlabel led BSA Surface concentration ( D 125 2 per mole of I-BSA (ug/cm ± AD 668 0.197 ± 0.019 804 0.223 ± 0.021 1352 0.215 ± 0.022 1682 0.215 ± 0.022 1721 0.196 ± 0.023 2350 0.192 ± 0.018 2470 0.202 ± 0.020 mean ± error 0.206 ± 0.050 3.4.2 Surface concentration as a function of time It was important to determine the time f o r the surface concentration to reach a steady state so that a su i t a b l e time could be chosen f o r the adsorption study. Using the technique from the l i t e r a t u r e of counting the tubing f o l l o w i n g washout of protein by buffer, adsorption runs were c a r r i e d out using various e q u i l i b r a t i o n times. The time curve f o r a 0.19 mg/ml BSA s o l u t i o n at 23°C i s given i n Figure 3.8. The k i n e t i c s of adsorption was in agreement with e a r l i e r reports (Brash and Davidson, 1976). It was found 2 that the surface concentration reaches a value of approximately 0. 14 fxg/cm within 2 hours and remains constant over a period of up to 8 hours. 50 0.16 en CM 0.14 H E P 0.12 cn =1 0.10 c o £ 0.08 H -4-< c c 0.06 o o g 0.04 o t co 0.02 0.00 i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — r 0 5 10 15 Time (hours) 1 1 1 1 1 1 1 r -20 25 Fig. 3.9. Time dependence for the adsorption of BSA on polyethylene at a solution concentration of 0.19 mg/ml. 3.4.3 Adsorption isotherm The BSA-polyethylene system was investigated by determining the quantity of protein adsorbed as a function of s o l u t i o n concentration. The surface concentration was c a l c u l a t e d d i r e c t l y by c u t t i n g up the tubing and gamma counting. Since the tubing had been rinsed p r i o r to counting, t h i s value must represent a minimum amount adsorbed. Any loosely bound pr o t e i n would have been removed during the r i n s i n g period. The adsorption isotherm o f o r albumin at 23 C over a concentration range of 0 to 2.7 mg/ml i s shown i n Figure 3.9, the upper l i m i t being about 6% of the value in blood. The e r r o r i s c a l c u l a t e d from the various components in equation 3.3, see Appendix 2. The surface concentration increases with concentration asymptotically 2 — 12 2 u n t i l a plateau value of approximately 0.20 fig/cm , (3 x 10 M/cm ). The surface concentration i s s i m i l a r to the values reported previously (Brash and Davidson, 1976; Morrisey and Stromberg, 1974). This behaviour indicates a l i m i t e d capacity of the surface f o r adsorption. This general behaviour has been observed with other types of macromolecules at a s o l i d - s o l u t i o n interface (Silberberg, 1962) and i s in general t y p i c a l of a l l p r o t e i n - p l a s t i c systems. The surface concentration quoted i s an average taken from the 5 m tube length. The p l o t s i n Figures 3.lOa-h show the surface concentration against tube s e c t i o n f o r BSA adsorbed from r a d i o l a b e l l e d BSA solutions of varying concentrations. Tube se c t i o n 1 i s the end of the tube at the f r a c t i o n c o l l e c t o r while tube s e c t i o n 25 i s the end of the tubing connected to the syringe. Adsorption from a s o l u t i o n concentration of 8 ug/ml i s not uniform along the tube. This may be due to a concentration gradient along 52 0.25 CO E 0.20 o cn c 0.15 o § 0.10 c o o I 0.05 D CO H * 0.00 0.0 i—i—i—i—|—i—i—i—i—|—i—i—i—i—i—i—i—i—i—|—i—i—i—i—|—i—r 0.5 1.0 1.5 2.0 BSA concentration (mg/ml) 2.5 Fig. 3.9. Adsorption isotherm for albumin on polyethylene at 23°C. o \ 00 a. G O {ti L C to o c o u o c o o 169 and r a d i a l d i f f u s i o n predominates. Taylor has obtained an approximate s o l u t i o n to the problem where longit u d i n a l molecular d i f f u s i o n has been neglected and where the r a d i a l d i f f u s i o n i s rapid. The longitudinal transfer i s due to convection. This w i l l be the case when a dissolved material of uniform concentration C 0 i s allowed to enter a pipe at a uniform rate. At time t = 0 the p o s i t i o n of the invading material i s given by x = 0, where x represents the distance from the entrance of the tube. The pipe i s f i l l e d with solvent only, concentration C = 0. The s o l u t i o n to t h i s problem i s given by 91 c/c„ = c/c„ = 1 • i e r f 2 2 1 - i erf 2 2 2v^kt x. 2/ kt ^ (x 1<0) (x 1>0) (A3.2) (A3.3) where erf z = 2 e dz (A3.4) x« = x - -u_t i 2 0 (A3.5) and k = 2 2 a u 0 192D (A3.6) In t h i s l i m i t the concentration i s constant across any cross s e c t i o n due to rapid r a d i a l d i f f u s i o n . Using Taylor's approximation i t was possible to c a l c u l a t e the length of the mixing zone f o r the displacement of a protein s o l u t i o n by a buffer i n a polyethylene tube. The relevant constants used i n the c a l c u l a t i o n were l i s t e d on page 91. The t h e o r e t i c a l d i s t r i b u t i o n of the concentration was ca l c u l a t e d f o r the displacement of BSA i n a polyethylene tube. The plot of C/C 0 against distance i s shown i n Figure A3.1. The plot shows that at C/C Q = 0.5, x = 500 cm, i . e . , the tube has been f i l l e d with one tube volume of the d i s p l a c i n g buffer. When x i s 700 cm, C/C D = 0.004 and 99.6% of the BSA has been displaced by the buffer. This length of 700 cm can be used to c a l c u l a t e the amount of protein displaced and f i n a l l y the amount of protein adsorbed. 92 250 350 450 550 650 750 x (cm) Fig. A3.1. C / C 0 vs x for the miscible displacement of BSA with buffer in a capillary tube. The volume contained i n a tube of diameter 0.038 cm and length 700 cm i. e . , volume used to displace 99.6% of the protein, i s 794 u l . From the displacement of the protein i n an adsorption experiment the number of counts, hence the amount of BSA, i n 794 ul i s determined. The surface concentration i s ca l c u l a t e d from the following equation SA d where T = surface concentration C t = t o t a l counts i n tube during the adsorption experiment C 0 = cpm output i n a volume of 794 pi SA = s p e c i f i c a c t i v i t y d = t o t a l surface area The surface concentration c a l c u l a t e d i n t h i s way w i l l represent the amount adsorbed i n equilibrium with the bathing solution. If anything i t w i l l underestimate the true equilibrium value since some rapodly desorbing material may appear in the displaced desorblng s o l u t i o n , thus increasing C 0 and reducing ( C - C 0 ) . 94 APPENDIX 4 STATISTICAL ANALYSIS A plot of C (bulk concentration) against C /T (bulk b b concentration/surface concentration per unit area,) appears to be best f i t by two li n e a r regions of d i f f e r e n t slopes. A li n e a r regression equation was calculated f o r each set of data. The question to ask i s , are the slopes of these l i n e s s i g n i f i c a n t l y d i f f e r e n t or are they estimating the same population? The n u l l hypothesis H0: 1 t = 1 2 w i l l be tested. The n u l l hypothesis i s a statement of no difference. In t h i s case we are t e s t i n g the e q u a l i t y of two l i n e a r regression l i n e s (Zar, 1984). For a simple l i n e a r regression Y = a + bX (A4.1) using the method of least squares the slope or regression c o e f f i c i e n t i s given by (A4.2) and a = Y - bX (A4.3) where the crossproducts sum of the deviations from the mean i s given as 95 Exy = E(X - X)(Y - Y) = ZXY - ££_±I (A4.4) X and Y are the mean values of X and Y respectively, and n i s the number of samples. x i s the de v i a t i o n of a X value from the mean of a l l X's and the sum of squares i s given as Ex 2 = S(X - X) = SX 2 - (A4.5) n The student's t test was used to test the eq u a l i t y of two regression l i n e s . The test s t a t i s t i c i s mean difference b t - b 2 t = = (A4. 6) standard error of mean difference s-b ± - b 2 If the test s t a t i s t i c , t, i s greater than some c r i t i c a l value the hypothesis H n i s rejected and the alternate hypothesis, H , 1, * 1_ accepted. The c r i t i c a l value of t depends on the degrees of freedom {v) and the le v e l of s i g n i f i c a n c e (a). For a two-tailed t test H 0 w i l l be rejected i f Itl > t (A4.7) 1 1 a(2),u The standard er r o r of the differe n c e between the regression c o e f f i c i e n t s of sample 1 and sample 2 i s 96 s- -b - b 1 2 1 Y - x J p + L Y x J p (A4.8: ( Z x 2 ) 1 ( Z x 2 ) 2 [ s 2 x] i s the pooled residual mean square and denotes the varience of the Y coordinate a f t e r taking into account the dependence of Y on the X coordinate. The pooled r e s i d u a l mean square i s given by (residual S S ) ± + (residual S S ) 2 [ s 2 ) = (A4.9) p (residual DF) ± + (residual DF) 2 where SS = sum of squares DF = degree of freedom 2 ( Z X y ; Z residual SS = Zy - (A4.10] 2 Zx residual DF = n - 2 (A4.ll) Zy 2 i s the sum of squares of the dif f e r n c e between Y and the mean Y, and i s given as Xy 2 = Z(Y - Y) = ZY 2 - i?XL (A4. 12) n In the comparison of the l i n e a r regression l i n e s sample 1 r e f e r s to the l i n e obtained from c u t t i n g up the tubing and sample 2 i s that c a l c u l a t e d from the depletion of t o t a l counts. 97 A4. 1 Testing the difference between the two regression l i n e s H0= 1± = 1 2 H : 1. * 1. For sample 1: For sample 2: ZX = 13433.03 ZX = 13433.03 ZY = 73031.48 ZY = 69946.35 The X and Y values were f i r s t divided by 1000. Zx 2 = 19.62382 2 Zy = 517.7767 Zxy = 100.4844 n = 23 a = 184.661 b = slope = 5.1205 re s i d u a l SS = 3.2426 re s i d u a l DF = 21 Zx = 19.62382 Zy 2 = 467.0201 Zxy = 95.40137 n = 23 a = 197.134 b = slope = 4.8614 residual SS = 3.2255 residual DF = 21 f s 2 ] = 0 . 1 5 4 0 s- - =0.1253 b - b 1 2 t = 2.0675 Reject H_ i f It I 2= t ° 0 1 1 Ct(2),l> The c r i t i c a l value of t f o r v = 42 and a s i g n i f i c a n c e l e v e l a = 0.05 ( i . e . , 5%) i s taken from tables and t = 2.018 0. 05(2) ,42 H 0 i s rejected, therefore one can conclude that there i s a s i g n i f i c a n t d i f f e r e n c e between the two regression l i n e s . 98 If there i s a c o r r e l a t i o n between sample 1 and sample 2 as in the case when the surface concentration i s ca l c u l a t e d by c u t t i n g up the tubing and by depletion of the t o t a l r a d i o a c t i v i t y , a paired-sample t test can be used. This two-tailed t test c a l c u l a t e s a t value by using the difference, d , between the two samples. The mean, variance, standard deviation and standard e r r o r are calculated using the differ e n c e betwwen the samples and n i s the number of differences. A4.2 The paired-sample test f o r the adsorption isotherm data H 0: p 1 = u 2 HA: a± * u 2 In t h i s case the alternate hypothesis i s given by H u±> u when sample 1 r e f e r s to the surface concentration (T) c a l c u l a t e d from the depletion of the t o t a l r a d i o a c t i v i t y and sample 2 to that from the rinsed tubing. From the r e s u l l t s i n Table 3.6 Zd = 0.125 Ed 2 = 1.911 x 10"3 n = 23 d = Zd/n = 5.343 x 10 - 3 v = 23 - 1 = 22 SS = I d 2 - = 1. 232 x 10"3 n 2 SS -s variance = s = = 5.60 x 10 d v standard d e v i a t i o n = s =