The present invention relates to surfaces, and more particularly to surfaces which can be used to prevent or reduce the activation of complement.

Complement and its activation are now well known, and are described for example in Roitt et al , "Immunology", Fourth Edition, published by Churchill-Livingston (London) . The activity ascribed to complement (C ) depends upon the operation of nine protein components (Cl to C9) acting in sequence, of which the first consists of three major sub-fractions termed Clq, Clr and Cls. Complement can be activated by the classical or alternative pathway, both of which will now be briefly described.

In the classical pathway, antibody binds to Cl, whose Cls subunit acquires esterase activity and brings about the activation and transfer to sites on the membrane or immune complex of first C4 and then C2. This complex has "C3-convertase" activity and splits C3 in solution to produce a small peptide fragment C3a and a residual molecule C3b, which have quite distinct functions. C3a has anaphylatoxin activity and plays no further part in the complement amplification cascade. C3b is membrane bound and can cause immune adherence of the antigen- antibody-C3b complex, so facilitating subsequent phagocytosis.

In the alternative pathway, the C3 convertase activity is performed by C3bB, whose activation can be triggered by extrinsic agents, e.g. particular microbial polysaccharides such as endotoxin, acting independently of antibody. The convertase is formed by the action of

Factor D on a complex of C3b and Factor B. This forms a positive feedback loop, in which the product of C3 breakdown (C3b) helps form more of the cleavage enzyme.

In both the classical and alternative pathways, the C3b level is maintained by the action of a C3b mactivator (Factor I) . C3b readily combines with Factor H to form a complex which is broken down by Factor I and loses its haemolytic and immune adherence properties.

The classical and alternative pathways are common after the C3 stage. C5 is split to give C5a and C5b fragments. C5a has anaphylatoxin activity and gives rise to chemotaxis of polymorphs. C5b binds as a complex with C6 and C7 to form a thermostable site on the membrane which recruits the final components C8 and C9 to generate the membrane attack complex (MAC) . This is an annular structure inserted into the membrane and projecting from it, which forms a transmembrane channel fully permeable to electrolytes and water. Due to the high internal colloid osmotic pressure, there is a net influx of sodium ions and water, leading to cell lysis.

It has been known for many years that, when human blood is pumped extracorporeally through, for example, a kidney dialysis membrane or a heart lung machine, and then returned to the patient, complement is activated via the alternative pathway, which produces significant morbidity in the patient. This morbidity results, at least in part, from the release of histamine by complement components C3a and C5a which causes vasodilation, brochopasm, a drop in blood pressure and which can be fatal. Manufacturers of devices used in extra-corporeal circulatory systems have adopted a number of strategies to reduce complement activation. Early attempts to

improve blood compatibility involved ionically binding heparin to a surface of the extra-corporeal circulatory system which contained blood. Although improved performance was sometimes observed it was attributed to heparin leaching from the material. Subsequent attempts involved other methods of coupling heparin to surfaces, generally using covalent bonds, for example end point attachment of heparin (Carmeda Bio Active Surface) . The performance of heparin in preventing thrombosis depends on the method by which it is bound to a biomaterial and the type and length of the spacers used in the coupling procedure. Physically or chemically incorporating other agents such as phospholipids, fibrinolytic enzymes, streptokinase or urokinase or prostaglandins or altering polymer hydrophilicity also reduces thrombogenecity. An additional approach involves biological modification of a polymer by protein adsorption or seeding with endothelial cells (including genetically altered cells producing tissue plasminogen activator) .

However none of the above methods are entirely satisfactory, for reasons of efficacy, expense, complexity or otherwise. There therefore exists a need to develop alternative ways of reducing morbidity due to complement activation.

According to the present invention there is provided a surface bound to an homologous complement restriction factor (HCRF) , said surface not being a surface to which said homologous complement restriction factor binds in vivo .

The homologous complement restriction factor is immobilised by said surface but retains the ability to prevent or reduce the activation of complement. This is

useful in preventing ^" or reducing morbidity in patients.

Thus a medical apparatus can advantageously be provided with such a surface bound to a HCRF. The term "medical apparatus" when used herein includes equipment and components thereof adapted for use in medicine. The medical apparatus may be used for transporting or treating whole blood or some other fluid containing complement (e.g. serum) . It may be provided in sterile form. The surface may be present at a part of the apparatus which contacts said fluid.

The medical apparatus may provide an extra-corporeal circulatory system, e.g. it may be a heart lung machine or a kidney dialysis machine.

The medical apparatus may be a conduit (e.g. cardiac by¬ pass tubing) , a valve, a membrane, a pump, an oxygenator, a catheter or cannula, or a fluid reservoir (e.g. for the storage of blood in a blood bank) .

It may be a prosthesis (e.g. an artificial heart, a pace¬ maker, a stent, a vascular graft, an artificial joint, etc. ) .

The medical apparatus may also be a piece of surgical equipment (e.g. suture material) .

It should however be appreciate that the present invention is not limited to medical apparatuses. It can be used to provide any article likely to contact blood or another complement containing fluid with a surface capable of preventing or reducing the activation of complement .

For example a medical dressing, drape or swab may be provided with such a surface.

The present invention can also be used in diagnosis. For example, an HCRF could be immobilised (e.g. by linking it covalently to a resin, such as CNBr-activated sepharose 4B (available from Pharmacia) and used to diagnose whether or not a patient has an abnormality which prevents the complement system from working normally or from being properly regulated. In the case of recombinant DAF, this could be used to measure C3b and C4b levels in order to diagnose patients suffering from paroxysmal nocturnal hemoglobinuria (PNH) . For example, serum from patients suffering from PNH could be run down a DAF or MCP column. C3b and C4b could be eluted using citric acid pH 4.0-6.0 or 100 mM Triethylamine. The amounts of these proteins could then be compared to the amounts in normal serum using an ELISA for the two proteins .

A diagnostic kit is therefore with the scope of the present invention. The kit may include instructions for use.

The present invention can also be used to purify substances which bind to HCRFs (e.g. by providing a resin as discussed above) . Thus, it can be used to purify complement components e.g. C3b and C4b. It can also be used to screen for substances which have similar structures to complement components. Such substances may themselves act as inhibitors of complement activity by competing with complement components. The substances can then be incubated in the presence of complement and assayed to see if they are indeed effective in inhibiting complement activity (e.g. by using a cell lysis assay) .

Inhibitors of complement activity are useful as therapeutic agents since, for example, they can be used to reduce the likelihood of complement-mediated rejection of xenografts .

The HCRF used to bind to the surface of an article may be naturally occurring or may be artificial (i.e. not found in nature) . One or more such HCRFs may be bound to such a surface. Preferred HCRFs are those which can prevent or reduce activation of complement via the alternative pathway. Such HCRFs include DAF, MCP, CR1 and CD59.

Reagents and kits for testing for complement activation are known in the art and are available from Quidel Corporation of 10165 McKellar Court, San Diego, CA 92121, USA (for example) . The techniques used are disclosed in publications available from Quidel Corporation entitled "Rapid Testing of Biomaterials for Complement Activation using In Vi tro Complement Immunoassays" and "An Algorithm for Complement Biocompatibility Testing" . Certain of these techniques can enable complement activation which occurs via the alternative pathway to be distinguished from activation which occurs via the classical pathway. In particular, an assay for the presences of Bb in a test sample can be performed. The presence of Bb is an indication of complement activation via the alternative pathway rather than the classical pathway.

Preferred HCRFs for use in the present invention can therefore reduce the level of Factor Bb in a test sample by reducing complement activation via the alternative pathway.

Alternatively an assay for the presence of C3a may be used if it is desired to assay the total degree of

complement activation (irrespective of the pathway by which activation occurs) . One such assay is the Quidel"" C3a Enzyme Immunoassay. This is based upon detection of C3a-des Arg, which is a cleavage product of C3a and which is longer lived in serum than C3a.

Although particular naturally occurring HCRFs are discussed above, it will be appreciated by the skilled person that the present invention is not limited to naturally occurring HCRFs.

Thus the present invention includes molecules which are variants of naturally occurring HCRFs, having one or more amino acid insertions, deletions or substitutions relative to such HCRFs, but which still have activity in preventing or reducing the activation of complement via the alternative pathway.

Such variants may have a substantial degree of amino acid sequence identity with the sequence of naturally occurring HCRFs (e.g. at least 50, 75, 90 or at least 95% sequence identity) . Where there is a high degree of sequence identity, a variant may have only a few amino acid differences from the amino acid sequence of a naturally occurring HCRF (e.g. less than 10 differences or less than 5 differences) .

Two embodiments of the present invention are discussed below. It is important to note that the present invention is not limited to these embodiments.

First Embodiment of the Present Invention

In a first embodiment of the present invention the HCRF desirably comprises a hydrophobic structure which can be

used to anchor the ^" HCRF in a cell membrane. Most desirably it comprises phosphatidylinositol (PI) . The structure and function of PI membrane anchors is discussed by Caras et al (Nature 325:545-549 (1987)) , by Davitz et al (J. Exp . Med. 163:1150-1161 (1986)) and by Medof et al (Biochemistry 25:6740-6747 (1986)) . Since phosphatidylinositol is normally found bound to an oligosaccharide in membrane anchors, the term "GPI anchor" (glycophosphatidylinositol anchor) is often used interchangeably with the term "PI anchor". The term "PI anchor" when used herein therefore includes GPI anchors.

One test which can be used to detect the presence of a PI anchor is its cleavage by phosphatidylinositol-specific phospholipase C (PIPLC) . This cleavage allows a part of the molecule to which the PI anchor is attached to be released. The presence of the PI anchor can be determined using FACS analysis. After digestion with PIPLC, no shift in the fluorescence should be seen as the protein has been removed from the cell surface. This assay is explained in Davies and Morgan {Biochem J. 295:889-896 (1993)) , where insect Sf9 cells expressing CD59 were incubated with PIPLC.

An HCRF molecule which does not normally comprises a PI anchor can be produced which has one. This can be done by altering a DNA molecule (or RNA molecule) which encodes an HCRF so that the altered version encodes a chimaeric molecule having a PI anchor at its C-terminus.

Thus chimaeric HCRF molecules can be used in the present invention.

Of course, some HCRF molecules already include PI anchors and therefore it is not necessary to provide these HCRFs

with PI anchors. These molecules include DAF, CD59 and variants thereof.

The HCRF (preferably comprising a PI region) can be bound to a surface by any appropriate method, providing that the HCRF when bound still retains activity in preventing or reducing the activation of complement. This can be assayed as discussed supra .

In a preferred method of the first embodiment of the present invention a liquid comprising an HCRF is incubated in the presence of a surface to which the HCRF is to be bound. This is done for sufficient time and at a temperature and a pH which allow binding to occur.

The temperature is preferably from 1 to 37°C, but it may be higher than this (e.g. up to 40°C) .

The pH may be, for example, from pH 6 to pH 9, and is preferably from pH 7 to pH 8. In order to keep the pH within a given range the liquid may be buffered. For example it may be phosphate buffered saline (PBS) .

The incubation period can depend upon the temperature, the pH, the surface and the HCRF. However, it can be determined by the skilled person using reasonable trial and error. Often an incubation period of over an hour will be used (e.g. an over-night incubation) .

Preferably a surface to which the HCRF is to be bound and a liquid containing the HCRF to be bound to the surface are both provided substantially free of detergent. Thus if an HCRF is initially provided in a relatively impure composition comprising sufficient detergent to prevent binding of the HCRF to a target surface, further

purification should ^" be performed, e.g. by using an immunoaffinity column to bind the HCRF, followed by an elution step and a neutralisation step.

In the case of DAF, for example, (which is often initially provided in a composition which also comprises the detergent Nonidet P-40 (Shell Chemicals)) elution from an anti-DAF immunoaffinity column can be used to bind to the DAF and the detergent can be washed away. Elution of DAF from the column can then be achieved using 100 mM triethylamine. Fractions can then be neutralised with 0.5 M TRIS.

Anti-DAF monoclonal antibodies which can be used in an immunoaffinity column are now widely available since a number of hybridoma cell lines producing such antibodies have been established. The skilled person can also produce his/her own hybridoma cell line to provide antibodies against a desired HCRF using the well known Kohler and Milstein technique or variations thereof.

Second Embodiment of the Present Invention

In a second embodiment of the present invention a surface bound to an HCRF via a covalent linkage is provided.

This may be achieved by using a cross-linking agent. The cross-linking agent may join the HCRF to the surface in any particular manner provided that the HCRF retains at least some HCRF activity.

For example the cross-linker may bind to a region of the

HCRF which is glycosylated. This can be achieved by oxidising a carbohydrate moiety present on the HCRF to provide an aldehyde group, which is then covalently

linked to the cross-linker (e.g. by reacting with a hydrazide group present on the cross-linker to form a hydrazone bond) .

The cross-linker may be used to covalently bind with a surface in any desired manner. This may be done before or after it is bound to the HCRF. For example it may be provided with a photo-reactive group which allows a covalent linkage to be formed with said surface.

A preferred cross-linker for use in the present invention comprises a photo-reactive group and a group capable of forming a covalent bond when reacted with an aldehyde group (e.g. a hydrazide group) .

Any suitable photo-reactive groups can be used. For example photoazides can be used. Here a benzene ring may be substituted with a variety of groups (in addition to an -N ₃ group) in order to vary its properties. For example an -N0 ₂ group can be used to modify sensitivity to UV light. Various molecules comprising photo-reactive groups which are known cross-linkers are shown in Table 1 overleaf:

A cross-linker which has been used by the present inventors is ABH (p-Azidobenzoyl hydrazide) . This is obtainable from Pierce of 3747 N. Meridian Road, PO Box 117, Rockford, IL 61105, USA; together with instructions in respect of its use in cross-linking.

Of course if cross-linkers with photo-reactive groups are used then care should be taken to ensure that the photo- reactive groups do not react with surfaces until this is desired. This can be done by keeping such cross-linkers in dark conditions until a photochemical reaction is desired.

Photo-reactive groups can be used to react with a wide range of materials. For example, they can be used to react with polyurethane, polythene, pσlyvinylchloride (PVC) , cellulose acetate, polysulphone and paralyne.

Furthermore, photo-reactive groups could be used to link HCRFs to surfaces already modified with heparin or a hydrogel .

The first embodiment of the present invention will now be described by way of example only, with reference to the accompanying drawings, wherein:

FIGURE 1 shows the cDNA sequence for human GPI anchored DAF.

The amino acid sequence and corresponding cDNA sequence for mature GPI anchored DAF is located between the points B and C indicated on the Figure.

(The part of the cDNA sequence between B and C which is indicated in dark type was originally thought by one group to correspond to an intron which could be spliced out to provide a fra eshift. However this is not now believed to be the case.)

The amino acid sequence located between points A and B is a 34 ammo acid signal peptide (which is indicated in dark type) .

FIGURE IA provides ammo ac d and cDNA sequence data in respect of MCP.

FIGURE 2 is a map of the vector pAcHLT-A. This vector can be obtamed from PharMingen, 10975 Torreyana Road, San Diego, CA92121, USA and is available under the catalogue number 21467P.

FIGURE 3 illustrates the expression of different forms of DAF using a Baculovirus expression system.

FIGURE 4 illustrates how the "His Tag" purification system works.

FIGURE 5 illustrates the use of Imidazole in eluting from DAF Ni-NTA resin.

FIGURE 6 illustrates the effect of increasing DAF concentration on the haemolysis of sheep CH50 erythrocytes .

FIGURE 7 provides ELISA results in respect of Cobe tubing coated with different forms of DAF at 4°C.

FIGURE 8 shows the effect of increasing DAF concentration on the coating of Cobe tubing at 4°C.

FIGURE 9 indicates that both DAF and MCP can be expressed in GPI anchored form in the insect cell line Sf9.

EXAMPLE 1

Expression of DAF using the Baculovirus System and subsequent tests of i ts functionali ty and ability to bind to plastic cardiac bypass tubing

This example discloses how to:

1) express GPI anchored and soluble forms of DAF using the baculovirus expression system (disclosed in Biochem J

295: 889-896 (1993)) ; and

2) use these expressed proteins to coat extra-corporeal bypass tubing, oxygenators and dialysis membranes; the rationale being to prevent complement activation when human blood passes through these systems during, for example, cardiac bypass.

Methods and Resul ts

The following scheme of work was followed:

1) Isolate the cDNAs for GPI anchored DAF and soluble forms of DAF and then ligate them into an appropriate expression vector.

2) Express the recombinant forms of DAF in the insect cell line SF9.

3) Purify the expressed forms of DAF.

4) Test the functionality of the different forms.

5) Test the ability of the different forms of DAF to bind to extracorporeal bypass tubing.

1) Isolation of cDNAs and ligation into a Baculovirus Expression Vector

EcoRI sites were added to the 5' and 3' ends of the cDNA sequence for the GPI anchored form of DAF. This ligated into the vector pAcHLT-B, at the EcoRI site in the multiple cloning site (MCS) .

A soluble form of DAF was prepared by mutating the codon for Glycine 320 to a stop codon using PCR. EcoRI sites were placed at the 5' and 3' ends of the sequence and it was ligated into the vector pAcHLT-A, at the EcoRI site in the MCS (see Figure 2) .

Finally normal human soluble DAF cDNA was isolated using reverse transcriptase PCR from RNA prepared from human lung tissue. This cDNA was again ligated into pAcHLT-A. Therefore, three forms of DAF cDNA were prepared and ligated into the pAcHLT plasmids.

• Human GPI anchored DAF

• Mutated soluble DAF

• Normal Human soluble DAF.

The amino acid sequence of Human GPI anchored DAF is provided in Figure 1 (see the discussion of Figure 1 at page 10) .

2) Expression of DAF in the insect cell line Sf9

It was decided to use the baculovirus system for the expression of DAF for the following reasons;

• It is a eukaryotic expression system • Post-translational modifications of the expressed

protein such as ^"* 0- and N-glycosylation should be the same as in humans . • Functionally active complement proteins have already been expressed using the baculovirus system. • The system gives high levels of expression.

To achieve expression of DAF the insect cell line Sf9 which originates from the fall armyworm ( Spodoptera frugiperda) , was co-infected with the cDNAs for the three forms of DAF. Three days post infection the cell supernatants were tested for the presence of DAF using ELISA. The results are shown in Figure 3 and it can be seen that expression of the three forms of DAF was obtained.

3) Purification of DAF from Sf9 cells

The "His tag" system was used although other purification techniques could have been used. (For example, an immunoaffinity column comprising anti-DAF monoclonal antibodies covalently bound to CNBr-activated Sepharose AB (Pharmacia) would provide a good purification system which may have advantages over the "His tag" system) .

In the His tag system, six histidine residues are added to the N-terminus of a protein when it is expressed in the pAcHLT plasmids. These histidine residues will bind to a nickel ion containing resin (called Nickel Nitrilo triacetic acid agarose or Ni-NTA, for short) .

Ni-NTA has an extremely high affinity for 6xHis residues. The Ni ²⁺ ion has six co-ordination sites, four of which interact with the NTA ligand leaving two sites for the binding of the 6xHis tag (see Figure 4) .

The binding affinity ^" is approximately Kd = IO ^"13 , which is higher than for most antibody/antigen or enzyme/substrate interactions. Elution of the tagged protein is achieved by competition with imidazole, which binds to the Ni-NTA resin and displaces the tagged protein (see Figure 5) .

OO

Hist-dm*

The three forms of DAF were successfully purified using

Ni-NTA agarose.

When assayed using a Lowry type protein assay, the yields of protein were as follows:

GPI Anchored DAF = 520 ug/ml (Total = 3 mis from

4xl0 ⁷ Sf9 cells) Mutated soluble DAF = 460 ug/ml (Total = 3 mis from

4xl0 ⁷ Sf9 cells) Norm Human soluble

DAF = 460 ug/ml (Total = 3 mis from

4xl0 ⁷ Sf9 cells)

4) Functional assays of the expressed forms of DAF

In order to determine whether the expressed forms of DAF were functional i.e. inhibit complement activation, a haemolysis assay was devised. If the proteins were functional then haemolysis would be inhibited. To measure haemolysis sheep CH50 erythrocytes were used.

The cells were sensitised to complement by incubation with IgM. After incubation with IgM the cells were washed and then incubated with increasing concentrations of recombinant DAF for 30 mins at 30°C, as described in Moran et al ( Journal of Immunology 149:1736-1743 (1992)) .

Human serum was added as a source of complement (classical pathway) , and the erythrocytes and serum were then incubated at 37°C for 35 min. Haemolysis was measured by determining the absorbance at 415nM. GPI anchored DAF purified from human packed erythrocytes was used as a control . The results are shown in Figure 6.

From Figure 6 it can be seen that the recombinant GPI anchored DAF reduced haemolysis significantly compared to the soluble forms.

5) Coating of the different forms of DAF onto Cobe ^* extracorporeal bypass tubing ( ^* Cobe is a trade mark of Cobe Laboratories Ltd. , Athena III, Olympus Business Park, Quedgeley, Gloucester, England, GL2 6NF) .

After proving that the proteins were functional the next challenge was to see if they would bind to plastic bypass tubing.

In order to show this on a small scale, discs of the tubing were cut out using a paper punch and tested. The discs were around 5 mm in diameter. 5 discs were then placed in a solution of each of the different forms of DAF (100 ug/ml in PBS) in Eppendorf tubes. 5 discs were also placed in only PBS and the same number in a solution of GPI anchored DAF plus 0.1% NP40 (a detergent which causes the anchors to coalesce in micelle) , as controls.

The discs were incubated and mixed overnight at 4°C.

After incubation the discs were blocked with 2% BSA for lh. They were then washed with PBS+0.1% Tween 20 five times (Tween 20 is a trade mark) , each wash was removed by aspiration. A solution of a biotinylated anti DAF antibody was then added and the discs incubated at 4°C for lh. The discs were then washed as described previously and then a solution of streptavidin conjugated horseradish peroxidase was added and the discs incubated at 4°C for 30 min. Again the discs were washed as described previously and after the final aspiration each disc was transferred to a separate well in a 96 well plat. A detection solution containing o-phenyldiamine was then added and the colour change was allowed to proceed for 5 min. IM sulphuric acid was added to quench the reaction. The results are shown in Figure 7 in histogram form.

O-phenylenediamine dihydrochloride (OPD) was used since it is cleaved by horseradish peroxidase in the presence of hydrogen peroxide to give a yellow precipitate. The reaction can be quenched by adding IM sulphuric acid (as above) . The OPD used was obtained in tablet form from SIGMA.

From the results it can be seen that the GPI anchored DAF binds very well to the tubing while the soluble forms do not. Furthermore, the presence of the detergent NP40 abolishes the effect, showing that the binding is mediated through the GPI anchor.

To quantitate further these results discs of Cobe tubing were incubated with a range of concentrations of GPI anchored DAF.

From Figure 8 it can ^" be seen that saturation of the discs with GPI anchored DAF is achieved around 50 ug/ml of protein.

EXAMPLE 2

It was decided to use FACS analysis techniques to determine whether Sf9 cells which were coinfected with the cDNAs for GPI anchored DAF and an artificially produced form of MCP having a GPI anchor (sometimes referred to as "membrane-bound MCP") . This artificially produced form of MCP is referred to both as CD46-GPI and as MCP-PI in The Journal of Virology, p7891-7899, Dec 1994, (Varior-Krishnan et al ) . The "Materials and Methods" section of this article describes the structure of this artificial form of MCP and also how it was obtained. Further detail is provided in an article by Lublin and Coyne entitled "Phospholipid-anchored and Transmembrane Versions of Either Decay-accelerating Factor or Membrane Cofactor Protein Show Equal Efficiency in Protection from Complement-mediated Cell Damage" (J. Exp . Med . 174:35-44 (1991))

The results are shown in Figure 9 together with results for an Sf9 control. These results indicate the expression of both DAF and MCP.

24 (2/2)

The second embodiment of the present invention will now be described by was of example only, with reference to the accompanying drawings, wherein:

FIGURE 10 shows a comparison of the binding of DAF

(90μg/ml) COBE™ tubing in the presence and absence of DAF.

FIGURE 11 shows a comparison of average control and absorbed DAF circuits.

FIGURE 12 shows a comparison of average control and DAF photolink circuits.

25 ^" EXAMPLE 3

Using the chemical cross-lin inσ reagent p-AzidoJbeazorl hydrazide (ABH) to link DAF onto COBE™ tubing

Introduction

ABH or p-Azidobenzoyl hydrazide (manufactured by Pierce, 3747 N. Meridian Road, P.O. Box 117, Rockford, EL 61105, USA) is a crosslinker that binds carbohydrates at one end and reacts photochemically at the other end. The chemical reaction occurs with the modification of vicinal hydroxyl groups on carbohydrates, in the presence of sodium merα-periodate (NaI0 ) in the dark, converting them to reactive aldehyde groups. These reactive aldehyde groups then react with the hydrazide group on the crosslinker to form hydrazone bonds. Subsequently, the photoreactive group on the crosslinker inserts non-specifically into neighbouring molecules when irradiated with UV or visible hght'.

ABH p-Azidobenzoyl hydrazide MW 177.17

Sugar moieties ofthe protein are oxidized to generate aldehydes that can react with the crosslinker. When the glycoprotein of interest is an antibody, it is advantageous to conjugate in a manner which will maintain its immunological activity. By using a crosslinker that binds to the Fc portion ofthe molecule away from its antigen binding site, a divalent immunologically active immunoglobulin is formed ² . Mild oxidation of an immunoglobulin with sodium periodate will open the ring structure, and produce reactive aldehydes on the carbohydrate moieties of the Fc portion. These will then bind to free amino group (-NH,) found in ABH. Sialic acid residues on proteins can be specifically oxidized with periodate under carefully controlled conditions. At 1 mM NaI0 ₄ , and a temperature of 0°C, the reaction is restricted to sialic acid residues ³ .

In order to link DAF to COBE tubing using ABH the followmg points had to be considered. As DAF was generated using the baculovirus system there may be no sialic acids present on the protein, as Sf cells have been found not to add on this sugar residue'- ³ . Therefore, more " harsh" oxidation conditions were required to incoφorate the ABH; lOmM Sodium periodate at room temperature ¹ .

26

Protocol for Conjugation of GPI DAF to COBE tubing

Materials required

A). Sodium acetate buffer: lOOmM sodium acetate buffer, pH 5.5

B). Sodium merα-periodate solution: 200mM sodium mera-periodate (NaI0 ₄ ).

Prepare fresh in sodium acetate buffer or water. Keep in dark.

C). ABH stock solution: 50 mM in dimethyl sulphoxide (DMSO).

D). PBS: 20mM phosphate buffer, 150mM NaCl, pH 7.4.

E). Glycerol.

Method 1: Testing whether ABH will bind GPI DAF to COBE tubing

Perform steps 1-8 in the dark

1). 15 ml of GPI DAF + 0.01% NP40 (90ug/ml) were dialysed for 72hrs in lOOmM sodium acetate buffer, pH 5.5.

2). ABH was added to 10ml of PBS to give a final concentration of ImM.

3). The ABH solution was pipetted into a length of COBE tubing (10cm). 5 mis of solution was required to fill each length. A length of tubing was filled with just

PBS to act as a no ABH control

4). The lengths of tubing were then wrapped in aluminium foil and incubated in the dark at 37°C for 2h.

5). While the tubing was incubating with ABH, 1 ml of 200mM sodium mefa- periodate was added to the GPI DAF to give a final concentration of lOmM. The oxidation was carried out for lh at room temp. lOmM sodium mera-periodate in lOOmM sodium acetate buffer, pH 5.5 was used as a no DAF control.

6). After lh glycerol was added to a final concentration of 15mM and the solution incubated for 5 min at 0°C to stop the oxidation.

7). After the 2h incubation the tubing containing the ABH was flashed three times with a UV Hght. The tubing was left for five minutes then the solution was removed.

8). The DAF solution was added to the tubing and this was then incubated at room temperature for 1 h

9). This solution was removed and the tubing washed twice with PBS and then it was ready for testing for the presence of DAF using the ELISA method described previously.

Method 2: Large scale binding to test COBE circuit and determination of complement activation

Perform steps 1-8 in the dark

1). 80 ml of GPI DAF 0.01% NP40 (40ug/ml) were dialysed for 72hrs in lOOmM sodium acetate buffer. pH 5.5.

2) ABH was added tn 80ml of PRS to give a final concentration of ImM

27

3). The ABH solution was pipeited into two test COBE circuits which were lm in length. 35 mis of solution was required to fill each circuit. 4). The circuits were then wrapped in aluminium foil and incubated in the dark at 37°C for 2h.

5). While the tubing was incubating with ABH, 4 ml of 200mM sodium meta- ⁵ periodate was added to the GPI DAF to give a final concentration of lOmM. The oxidation was carried out for lh at room temp.

6). After lh glycerol was added to a final concentration of 15mM and the solution incubated for 5 min at 0°C to stop the oxidation. 7). After the 2h incubation the tubing containing the ABH was flashed three l o times with a UV light. The tubing was left for five minutes then the solution was removed.

8). The DAF solution was added to the tubing and this was then incubated at room temperature for 1 h

9). This solution was removed and the tubing washed twice with PBS and then it

15 was ready for testmg with human blood.

10). Heparinsed blood was (35ml) was added to the coated circuits and the control circuit.

12). The blood was pumped around the circuits using a peristaltic pump and the circuits kept at 37°C using a water bath.

13). Samples (0.5 ml) were taken at defined time intervals (0,1,2,4,6,8,10,15,20,

20 30,45,60, and 90 min) and tested for C3a-des-arg as a measure of complement activation.

14). This was carried out using Quidel™ C3a EIA (discussed supra).

Results

25

ABH was found to link GPI DAF onto COBE™ tubing (see Figure 10).

An ABH linked DAF circuit was then compared with a control circuit (no DAF), and one which had been incubated with a DAF solution for lh at 37°. When 30 samples were tested from these circuits using a Quidel™ kit, the absorbed DAF circuit did not differ significantly from the control circuit (Fig 11) but there was a significant difference between the photolinked circuit and the control circuit (Fig 12). All these results are a mean of 2.

References

1) Das, M., and Fox, C.F., (1979) , Chemical crosslinking in biology, Ann. Rev. Biophys . and Bioeng.

8, .65-196

2) 0' Shannessy, D.J., and Quarles, R.H., (1985) , Specific conjugation reactions of the oligosaccharide moieties of immunoglobulins, J. Applied Biochem . 7, 347- 355.

3) 0' Shannessy, D.J., Voorstad, P.J., and Quarles, R.H., (1984) , A novel procedure for labelling immunoglobulins by conjugation to oligosaccharide moieties, Immunol . Lett . 8, 273-277.

4) Davies, A., and Morgan, B.P., (1993) , Biochem . J. 295, 889-896.

5) Letson, C.S., and Sodetz, J.M., (1996) , Expression and characterization of recombinant human C8 _/ S, Mol . Immunol . 33, supplement 1, 11.