AU2006316039B2

AU2006316039B2 - A method of chromatography using semi-synthetic heparin ligands

Info

Publication number: AU2006316039B2
Application number: AU2006316039A
Authority: AU
Inventors: Mathias Alterman; Nicolas Thevenin
Original assignee: GE Healthcare Bio Sciences AB
Current assignee: Cytiva Sweden AB
Priority date: 2005-11-21
Filing date: 2006-11-15
Publication date: 2011-08-18
Anticipated expiration: 2026-11-15
Also published as: EP1951394A1; EP1951394A4; WO2007058592A1; AU2006316039A1

Abstract

The present invention relates to a method of isolating heparin-binding target compounds by liquid chromatography, which method comprises to provide a mobile phase comprising the target compound(s); to contact the mobile phase with a separation matrix comprising semi-synthetic heparin ligands to adsorb the target compounds to the matrix; and, optionally, to recover one or more target compounds by selective elution. The invention is useful to purify proteins such as Factor Xa; Heparin Cofactor II (HCII); and Antithrombin III (ATIII).

Description

WO 2007/058592 PCT/SE2006/001297 A method of chromatography using semi-synthetic heparin ligands. Technical field The present invention relates to isolation and purification of heparin-binding compounds, 5 such as coagulation factors. More specifically, the invention relates to a method of chromatography, wherein a novel chromatography ligand is used. The invention also embraces a method of preparing a separation matrix that comprises such ligands; a chromatography column comprising the separation matrix and the use thereof. 10 Background In the biotech industry, recent scientific advances such as the mapping of the human ge nome and improved computerised instruments such as high through put scanning ma chines provide a steady and rapidly increasing flow of new information. However, once the DNA sequence of an active compound such as a drug candidate or a diagnostic tool 15 has been determined, much work will still remain to be done. Firstly, a suitable expres sion system must be designed, including a vector capable of carrying the DNA sequence and a host cell capable of expressing the DNA at sufficiently high levels. Secondly, to obtain a quality which is acceptable for medical purposes, an efficient purification scheme is required, which needs to be adapted to the specific contaminants present in 20 each expression system. For example, if animal host cells are used, residues such as traces of virus or prions will constitute a potential risk of serious contamination. For safety reasons, the approving authorities in most countries have set levels of acceptable contamination levels which are either very low or zero. 25 Due to its versatility and sensitivity to the target compounds, chromatography is in volved as at least one step in many currently used biotech purification schemes. The term chromatography embraces a family of closely related separation methods, which are all based on the principle that two mutually immiscible phases are brought into contact. More specifically, the target compound is introduced into a mobile phase, which is con 30 tacted with a stationary phase. The target compound will then undergo a series of inter actions between the stationary and mobile phases as it is being carried through the sys- WO 2007/058592 2 PCT/SE2006/001297 tem by the mobile phase. The interactions exploit differences in the physical or chemical properties of the components of the sample. The stationary phase in chromatography is comprised of a solid carrier to which ligands, which are functional groups capable of interaction with the target compound, have been coupled. Consequently, the ligands will impart to the carrier the ability to effect the sepa ration, identification, and/or purification of molecules of interest. Liquid chromatogra phy methods are commonly named after the interaction principle utilised to separate compounds. For example, ion exchange chromatography is based on charge-charge in teractions; hydrophobic interaction chromatography (HIC) utilises hydrophobic interac tions; and affinity chromatography is based on specific biological affinities. Many biotech products are proteins or at least protein-based compounds. As is well known, depending on their isoelectric point, proteins are chargeable at certain pH values and can hence be purified using ion exchange chromatography. Other chromatographic techniques, such as HIC, thiophilic chromatography and immobilised metal affinity chromatography (IMAC) are also useful to this end. However, a chromatography tech nique which due to its outstanding specificity is extensively used for protein purification is affinity chromatography, wherein a mechanism sometimes denoted lock/key interac tion is utilised. Thus, in affinity chromatography, a highly specific biological affinity between two compounds is utilised, such as presented between an antibody and its anti gen; between an enzyme and its receptor; and between other ligand and receptor pairs, such as biotin/avidin. A large number of biotech products are currently purified by chromatography using as ligand their native receptor. For example, antibody products may be purified by affinity chromatography using their respective antigen as a ligand immobilised to a suitable carrier. When biotech products for use in the medical or diagnostic field are purified, the purity obtained in the final product will be of utmost importance. In chromatography, a well recognised risk of impurities involves ligand leakage or other contamination due to the nature of the ligand. In affinity chromatography processes, ligand leakage i.e. release in WO 2007/058592 PCT/SE2006/001297 the elution step not only of the target compound but also of ligands or ligand residues may occurs due to the presence of peptide bonds in the ligands, which will consequently be sensitive to certain elution conditions that cause peptide bond breakage in turn result ing in a contaminated end product. Further, if it does occur, ligand leakage will often constitute a more serious problem in affinity chromatography than in other chroma tographic modes, since the leaked affinity ligand, or even a fragment thereof, may have retained its native biological function. Needless to say, such biological functions are un desired and hence regarded a contamination of the end product. Furthermore, if the pro tein ligand has been obtained by extraction directly from an animal source, the affinity matrix may contain residues of biological substances normally or exceptionally present in the animal tissues such as viruses or prions. In practise, the problems of ligand leakage and contamination are commonly overcome by introducing further downstream purification step(s) that target the 'potentially leaked ligand or contaminant. However, trace amounts may still be present, and as indicated above, the drug approval authorities these days apply extremely strict limits. An alterna tive way of overcoming the problem is to avoid altogether the use of ligands that involve such risk of contamination. For example, organic compounds or smaller synthetic pep tides will often be preferred options to protein ligands. In the case of animal-extracted protein ligands, an alternative solution is to provide a recombinant expression system, such as a bacterial host cell system which is known to be safe. However, expressing a complex protein in a bacterial host will commonly result in a product that differs slightly from the original, for example in terms of folding and glycosylation levels. With regard to the choice of the optimal purification scheme of a biotech product, it is well known that some proteins are more difficult than others to purify. For example, DNA-binding proteins form an extremely diverse class of proteins sharing a single char acteristic, namely their ability to bind to DNA. To enable a more selective purification, they can be produced as fusion proteins comprising a tag sequence fused to that coding for the protein of interest. However, tagging is a time-consuming and costly technology, which since it always involves the addition of a foreign material to the target protein will WO 2007/058592 PCT/SE2006/001297 include potential risks associated to that material. However, due to their ability to bind DNA, group specific affinity chromatography using heparin as a ligand has proved to be an alternative way to go. Heparin, which is a highly sulphated glycosaminoglycan, has also been shown to have the ability to bind a wide range of other biomolecules, and therefore heparin affinity chromatography is also used to purify other proteins of medical interest such as coagulation factors. Heparin occurs in nature as a mixture of chains consisting of repeating disaccharide units formed by an uronic acid (L-iduronic acid or D-glucuronic acid) and by an amino sugar (glucosamine), joined by a-1--> 4 or P-1--+ 4 bonds. The uronic acid unit may be sul phated in position 2 and the glucosamine unit is N-acetylated or N-sulphated and 6-0 sulphated. Heparin is a polydisperse copolymer with a molecular weight ranging from about 3,000 to about 30,000 Dalton. Casu et al (Carbohydrate Letters, Volume 1, pp. 107-114 (1994): "Biologically Active, Heparan Sulfate-like Species by Combined Chemical and Enzymatic modification of the Escherichia coli Polysaccharide K5") relates to the simulation of the heparin structure and biological activity by chemical sulfation of other polysaccharides obtained by ex pression in E.coli. In this article, the interest in heparin for its anticoagulant and anti thrombotic properties in therapy is discussed. It is shown that by formation of non sulfated polysaccharide chains, covalently bound to a protein core, and by subsequent modifications thereof, an in vitro inhibition of activated Factor X by antithrombin simi lar to that of beef mucosal heparan sulfate could be obtained. However, compared to commercial heparin, the anticoagulant strength was consistently lower. WO 01/02597 (Petrucci et al.) relates to a process of preparing the polysaccharides K4 and K5, which are responsible for extra-intestinal infections. The polysaccharides are obtained from Escherichia coli by fermentation, wherein an aqueous medium compris ing defatted soy flour, mineral salts and glucose is used as the culture medium. The ad vantage of the disclosed process is stated to reside in the yields obtained. Since then, further modified K4 and K5 polysaccharides have been suggested, see e.g. WO WO 2007/058592 PCT/SE2006/001297 02/50125 and WO 02/068477. Use of the modified K5 polysaccharide in pharmaceutical preparations has also been disclosed, see e.g. WO 02/083155 (angiogenesis) and WO 03/011307 (HIV). Heparin affinity ligands are used commercially in separation matrices available from a number of sources, such as HeparinTM Sepharose (GE Healthcare, Uppsala, Sweden); Affi-GelTM (Bio-Rad); and Immobilised Heparin (Pierce). However, said commercial heparin ligand products all contain heparin obtained by extraction from an animal source, also known as extractive heparins, such as porcine heparin. WO 95/05400 (Minnesota Mining and Manufacturing Company) relates to an improved method of providing heparin functional affinity supports, wherein a heparin functional polymer that comprises biologically active heparin is bonded to a hydrazide reactive group formed from reaction of hydrazine with an azlactone-functional polymer. The stated advantages of the affinity support so prepared are a high binding efficiency, a sta ble coupling linkage and the mild reaction conditions used in its preparation. The term "heparin functional polymer" as used in WO 95/05400 includes natural heparins i.e. heparins derived from animal sources, various chemical modifications thereof, and syn thetic heparin-like molecules. However, as discussed above, there are inherent disadvan tages of using heparin of animal origin due to the risks of transferring animal residues such as viruses or prions to the final purified product. In addition, even though no meth ods are disclosed in WO 95/05400 for preparing fully synthetic heparin molecules, con sidering the currently available technologies, it can be assumed that even though it may be desired to avoid the above-mentioned risks, chemical synthesis of heparin in the labo ratory would be very time-consuming and hence costly. Thus, there is still a need in this field of improved methods of providing chromatogra phy matrices useful for the separation and/or isolation of heparin-binding compounds, such as coagulation factors and DNA-binding proteins.

WO 2007/058592 PCT/SE2006/001297 Short description of the present invention In one aspect, the present invention relates to a method of separating and/or isolating one or more heparin-binding compounds which method does not leave any traces of animal origin in the purified product. This can be achieved by a method as defined in the ap pended claims, which comprises to contact the heparin-binding compound(s) with a separation matrix comprising ligands of semi-synthetic origin. Further aspects and advantages of the present invention will appear from the detailed description and claims that follow. Definitions The term "semi-synthetic heparin" means herein heparin which has been obtained by expression in a bacterial host system and subsequent chemical and/or enzymatic modifi cation including sulfation. The term "eluent" means a liquid capable of releasing one or more adsorbed compounds from a separation matrix. Eluents conventionally used in liquid chromatography are de fined by conditions such as pH and/or conductivity. The term "K5 polysaccharide" or "polysaccharide K5" refers to the polysaccharide com posed of glucuronic acid and N-acetylglucosamine described by W.F. Vann et al., Eur. J. Biochim. 1981, 43, 51-134. The term "heparin-binding compound" means herein any compound, molecule or other entity capable of binding to heparin and semi-synthetic heparin with a binding strength that allows its separation by liquid chromatography. Detailed description of the invention Thus, in a first aspect, the present invention relates to a method of separating and/or iso lating at least one heparin-binding target compound from other component(s) of a liquid, which method comprises (a) providing a mobile phase comprising the target compound; WO 2007/058592 7 PCT/SE2006/001297 (b) contacting the mobile phase with a separation matrix comprising semi-synthetic heparin ligands to adsorb the target compound(s) to the matrix; and, optionally, (c) recovering one or more target compounds by contacting an eluent with the separa tion matrix. The liquid from which one or more target compounds are to be separated may be any liquid, and is commonly the liquid wherein the target compound(s) have been manufac tured. Thus, in one embodiment of the present method, the liquid is a cell lysate or the supernatant of a fermentation broth. The cell lysate may have been clarified on before hand, such as by simple filtration or a preceding step of chromatography. Alternatively, the liquid is an unclarified cell lysate. Thus, the present method may be a capture step, wherein the majority of the target compounds are isolated. In an alternative embodiment, the present method is an intermediate or polishing step, in which case the mobile phase will include an eluate or flow-through phase originating from a preceding purification step. In an advantageous embodiment, the separation matrix comprising semi-synthetic hepa rin ligands is provided as described in more detail below, in the context of the second aspect of the invention. It is understood that the term "semi-synthetic heparin ligands" embraces semi-synthetic ligands comprising heparin and functional fragments thereof; as well as fusion proteins comprising any of the above. Thus, the present semi-synthetic heparin ligands are characterised by being able to bind substantially the same target compounds as extractive heparin. In an advantageous embodiment, the semi-synthetic heparin ligands used in the present method exhibit binding capacities substantially equivalent to those of porcine extractive heparin. To provide a suitable mobile phase, the liquid comprising the target compound(s) is commonly combined with an appropriate buffer wherein conditions such as pH and/or conductivity are controlled. Buffers suitable to this end are as a general rule the buffers conventionally recommended as mobile phases for commercial separation matrices com prising extractive heparin ligands. Thus, in one embodiment, the mobile phase comprises WO 2007/058592 PCT/SE2006/001297 a buffer selected from the group consisting of Tris-HC1, EDTA, mercaptoethanol and glycerol. In one embodiment, the pH of the mobile phase, as it is contacted with the separation matrix, is neutral to weakly alkaline, such as in the range of 7-14. In an ad vantageous embodiment, the pH of the mobile phase is about 8. Even though the present method will commonly utilise conventional conditions for affin ity chromatography, the skilled person will easily understand that since the semi synthetic heparin ligands provided according to the invention are polyanionic com pounds, they are equally useful in ion-exchange chromatography applications. Thus, in a first embodiment, the present method is an affinity chromatography process utilising the specific biological affinity between target and ligand in the adsorption step. In an alter native embodiment, the method is an ion-exchange process for the purification of nega tively charged target compounds wherein charge-charge interactions are utilised in the adsorption step. The separation matrix may be provided in the form of porous or non-porous particles, such as substantially spherical beads. Such particles may have an average size in the range of 10-500 ptm, such as in the range of 50-150 ptm. In an advantageous embodi ment, the particles exhibit an average particle size of about 90 pm. In this context, the particle sizes are given as the median particle size of a cumulative volume distribution. However, alternative formats of the separation matrix are also embraced by the present invention. For example, the separation matrix may be irregularly shaped particles, such as crushed agarose; a monolith; a membrane; a chip; a surface; a capillary; or the like. Suitable materials for making the separation matrix will be discussed in more detail be low in the context of the second aspect of the invention. To effect the contacting of step (b), the separation matrix may be provided in a chroma tography column, commonly as a packed or fluidised bed of particles. In one embodi ment, step (b) is carried out by passing the mobile phase across the separation matrix. In an advantageous embodiment, the flow rate of the mobile phase across the separation WO 2007/058592 PCT/SE2006/001297 matrix is in the range of 50-400 cm/h. This embodiment preferably utilises a column comprising a packed bed of essentially spherical particles. As will be apparent to the skilled person in this field, step (c) will be included in the pre sent method if the desired product is one or more purified target compounds, while step (c) will be omitted if the desired product is a purified liquid which is substantially devoid of one or more target compounds. Thus, in one embodiment, the present method com prises a step of recovering target compound(s) by adding an eluent to the separation ma trix. In this context, an eluent is understood to be any liquid capable of releasing the ad sorbed compound(s) from the matrix. In one embodiment, the salt concentration is higher in the eluent than in the mobile phase, which means that the target compound is eluted by salt addition, in other words, by an increase in conductivity. This elution mode is preferably carried out by adding a gradient of changing salt concentration i.e. an eluent wherein the salt concentration gradually increases. A linear gradient is commonly used, such as from 1- 100% of 1.OM of a common salt such as NaCl. Alternatively, a step-wise salt gradient is used in the present eluent. In addition to the details above, general well known principles commonly applied in liq uid chromatography may be utilised in working the present invention. For a review of chromatography methods useful for protein purification, see e.g. Protein Purification Principles, High Resolution Methods and Applications (J.-C. Janson and L. Ryd6n, 1989 VCH Publishers, Inc.). The target compound may be any heparin-binding compound, cell or molecule. In one embodiment, the heparin-binding compound is a coagulation factor or a DNA-binding protein. In this context, it is understood that the terms "DNA-binding protein" and "co agulation factor" include in addition to the full protein any functional fragment or fusion protein thereof, which retains the original binding properties. Thus, the heparin-binding compound may be selected from the group consisting of restriction endonucleases; pro tein synthesis initiation factors and proteins; seine protease inhibitors, such as anti thrombin; enzymes, such as lipases; growth factors; and lipoproteins. In an alternative C.\NRPonblDCC\KXG\3727129 I.DOC-30AM/21 I - 10 embodiment, the heparin-binding compound is selected from the group consisting of serum coagulation proteins. In a second aspect, the present invention relates to a method of preparing a separation 5 matrix by chemical and enzymatic modification of heparin expressed in E. coli. The present method comprises to provide semi-synthetic heparin and coupling said heparin to a carrier. In one embodiment, the method of preparing a separation matrix comprising semi synthetic heparin ligands which method comprises i) expression of K5 polysaccharide in bacterial host cells; 10 ii) N-deacetylation of GlcNAc residues of the K5 polysaccharide so obtained, and N-sulfation of resulting amino groups; iii) C5 epimerisation of D-GIcA to L-iduronic acid; iv) sulfation of the epimerised product to provide semi-synthetic heparin ligands; and 15 v) coupling of the ligands from step (d) to a carrier. In another aspect, the present invention relates to a method of preparing a separation matrix comprising semi-synthetic heparin ligands by coupling of the semi-synthetic heparin ligands to a carrier, wherein the semi-synthetic heparin ligands are provided by 20 (i) expression of K5 polysaccharide in bacterial host cells; (ii) N-deacetylation of GIcNAc residues of the K5 polysaccharide so obtained, and N-sulfation of resulting amino groups; (iii) C5 epimerisation of D-GIcA to L-iduronic acid; and (iv) sulfation of the epimerised product to provide semi-synthetic heparin 25 ligands, and wherein the coupling is performed by reductive amination. The K5 polysaccharide of step (i) is composed of substantially equimolecular quantities of glucuronic acid and N-acetylglucosamine, which make up the alternate linear repetitive unit 4-p-glucuronil-1,4,a-N-acetylglucosamine. In step (i), a suitable bacterial host cell is 30 fermented by any well known method, such as submerged culture, to produce the K5 C \NRPonblDCC\KXG\727129_1 DOC-3A0620 I - 1OA polysaccharide. The prokaryotic cells are preferably bacterial cells, such as Escherichia coli. In an advantageous embodiment, the host cells secrete the polysaccharide product into the culture medium. Ecoli strains suitable for production of polysaccharide K5 are obtainable from public collections, such as ATCC [American Type Culture Collection] or 5 DSM [Deutsche Sammlung von Microorganismen]. In one embodiment, the present method uses strain Ecoli 010:K5:H4, ATCC number 23506, for providing the K5 polysaccharide. The K5 polysaccharide is preferably purified according to well known methods. A suitable starting amount may be in the range of 5-15 g, such as about 10 g. For a review of expression of E.coli to produce K5 polysaccharide, see WO 01/02597, 10 which is hereby included herein via reference.

WO 2007/058592 11 PCT/SE2006/001297 In the first step of chemical modification, the K5 polysaccharide is, after appropriate pu rification and conditioning, subjected to N-deacetylation and subsequent N-sulfation, which are carried out by methods known per se. In an advantageous embodiment, the GlcNAc residues of the polysaccharide K5 are N-deacetylated by adding sodium hydrox ide and reacting at increased temperature such as 40-80 0 C, for an appropriate period of time. The solution is allowed to cool to ambient temperature, and preferably neutralised by adding acid such as hydrochloric acid. The subsequent N-sulfation of resulting amino groups is provided by adding a suitable sulfating agent such as pyridine sulfur trioxide or trimethylamine sulfur trioxide and re acting for an appropriate period of time at a slightly increased temperature. Salt is then preferably removed by using any well known technique for desalting a solution. The C5 epimerisation, which is a step of enzymatic treatment, converts D-GlcA to L iduronic acid. Such epimerisation may be performed in solution or with immobilised enzyme. The enzyme glucuronosyl C5 epimerase, commonly known as C5 epimerase, is commercially available and methods for epimerisation are well known in this field. In one embodiment, the sulfation of step (iv) comprises an O-oversulfation carried out by converting the C5 epimerised K5 polysaccharide into a ternary or quaternary salt thereof and treating the salt with an 0-sulfation agent according to well known methods. In an advantageous embodiment, the sulfation of step (iv) also comprises a selective 0 desulfation carried out according to any well known method. This may e.g. be provided by passing the 0-oversulfated product through a cationic exchange resin IR 120 H+; wa shing of the resin with about 3 volumes of deionised water; and neutralisation with pyri dine. In this context, selective 0-desulfation means that the sulfate groups in position 6 of the glucosamine are eliminated first, then the sulfate groups in position 3 and 2 of the uronic acid and finally the sulfate group in position 3 of the amino sugar. In a specific embodiment, the present method also comprises a step of 6-0-sulfation of the sulfated product from the preceding step. Such 6-0 sulfation may be carried out as WO 2007/058592 12 PCT/SE2006/001297 discussed above for O-sulfation. In the most advantageous embodiment, the so obtained product is subjected to N-sulfation. The semi-synthetic heparin ligands so produced are advantageously purified according to well known methods, such as diafiltration. To obtain ligands of the desired molecular weight, the product is then depolymerised to obtain a suitably sized ligand which still retains its biological activity. In one embodiment, the semi-synthetic heparin ligand which is immobilised in the subsequent step exhibits a molecular weight of 10,000 25,000 Dalton, such as about 20,000 Dalton. Further details regarding steps (ii)-(iv) of the present method are found e.g. in WO 02/50125. In step (v), the carrier to which the semi-synthetic heparin ligands according to the in vention are immobilised may be any suitable porous or non-porous carrier material commonly used in the purification of proteins. In one embodiment, the carrier is com prised of a cross-linked carbohydrate material, such as agarose, agar, cellulose, starch, pectin, dextran, chitosan, konjac, carrageenan, gellan, alginate etc. The carrier can easily be prepared according to standard methods, such as inverse suspension gelation (S Hjerten: Biochim Biophys Acta 79(2), 393-398 (1964). Alternatively, the carrier is a commercially available product, such as SepharoseTM FF (GE Healthcare, Uppsala, Sweden). In an advantageous embodiment of the present invention, the carrier is a cross linked polysaccharide, such as agarose. In a specific embodiment, the agarose has been prepared and/or modified to present an improved rigidity in order to withstand high flow rates, see e.g. US 6,602,990 (Berg). Alternatively, the carrier used in the present method is comprised of cross-linked syn thetic polymers, such as styrene or styrene derivatives, divinylbenzene, acrylamides, acrylate esters, methacrylate esters, vinyl esters, vinyl amides etc. Carriers made from such polymers are easily produced according to standard methods, see e.g. "Styrene based polymer supports developed by suspension polymerization" (R Arshady: Chimica WO 2007/058592 13 PCT/SE2006/001297 e L'Industria 70(9), 70-75 (1988)). Alternatively, the carrier is a commercially available product, such as SourceTM (GE Healthcare, Uppsala, Sweden). However, in this em bodiment, the surface of the carrier is preferably modified to increase its hydrophilicity, usually by converting the majority of the exposed residual double bonds to hydroxyl groups, before allylation and coupling of the semi-synthetic heparin ligands. In a specific embodiment, the semi-synthetic heparin ligands are inmobilised to the car rier via extenders, or a coating polymer layer. Such extenders, also known as "flexible arms", may be any well known organic or synthetic polymers. Thus, the carrier may e.g. be coated with dextran, to provide a hydrophilic nature to the support, to which the semi synthetic heparin ligands are immobilised according to well known methods in this field. For a review of ligand coupling techniques, see e.g. Hermanson et al in "Immobilised Affinity Ligand Techniques", Academic Press Inc. (1992). Finally, the coupling of the 0-sulfated products from step (iv) may be achieved by im mobilisation to a carrier using any well known method. However, since chromatography matrices are commonly washed and/or regenerated using alkali such as sodium hydrox ide, the coupling of the semi-synthetic heparin ligands obtained should preferably be alkali stable for an extended period of time. Thus, in an advantageous embodiment, in step (v), the product from the preceding step is coupled to the carrier using reductive amination, as disclosed e.g. in Hermanson et al in "Immobilised Affinity Ligand Tech niques", Academic Press Inc. (1992), pp. 69-79, which is hereby included herein via ref erence. In a specific embodiment, the reducing agent is selected from the group consist ing of NaBH 4 and NaCNBH 3 . In a third aspect, the present invention relates to a separation matrix, which comprises a carrier to which ligands have been coupled, wherein the ligands are bacterially ex pressed, chemically and enzymatically modified heparin. Thus, this aspect of the inven tion is a separation matrix, which comprises semi-synthetic heparin ligands coupled to a carrier. The carrier may be as discussed above in the context of the first and second as pects of the invention. In a specific embodiment, the separation matrix according to the WO 2007/058592 14 PCT/SE2006/001297 invention comprises chromatography ligands prepared by expression of K5 polysaccha ride in bacterial host cells; deacetylation; sulfation of resulting amino groups, C5 epimer isation of D-GlcA to L-iduronic acid; and O-sulfation of C2 of IdoA and at C6 of gluco samine units. In one embodiment, the semi-synthetic heparin ligands have been coupled to the carrier via amine linkages. In one embodiment, the present matrix has been prepared by the method defined in any one of claims 12-21. Thus, any of the details of the separation matrix discussed above in relation to the first and second aspects of the invention may apply to this aspect. The present invention also encompasses a chromatography column comprising a separa tion matrix as described above. In an advantageous embodiment, the column is made from any conventional material, such as biocompatible plastic, e.g. polypropylene, steel, such as stainless steel, or glass. The column may be of a size suitable for laboratory scale or large-scale purification. In a specific embodiment, the column according to the inven tion is provided with luer adaptors, tubing connectors, and domed nuts. The separation matrix may be packed in the column, or provided as a fluidised bed. Finally, an additional aspect of the invention is the use of a separation matrix, which comprises a carrier to which semi-synthetic heparin ligands have been coupled, wherein the semi-synthetic heparin ligands comprises bacterially expressed, chemically and en zymatically modified heparin, in liquid chromatography. In one embodiment of the present use, the semi-synthetic heparin ligands are coupled via amine linkages to the carrier. In an advantageous embodiment, the present use enables a heparin-binding target molecule selected from the group consisting of Factor Xa; Hepa rin Cofactor II (HCII); and Antitbrombin III (ATIII) to be separated and/or isolated. Such purified heparin-binding targets compounds are useful e.g. in the medical or diag nostic field.

WO 2007/058592 15 PCT/SE2006/001297 The present invention also encompasses a kit which comprises, in separate compart ments, a chromatography column as described above; at least one buffer; and written instructions for purification of heparin-binding target compounds. The kit according to the invention may be used e.g. in the medical or diagnostic field. EXPERIMENTAL PART The present examples are provided for illustrative purposes only, and should not be con strued as limiting the present invention as defined by the appended claims. Example 1: Manufacture of a separation matrix according to the invention The K5 polysaccharide was prepared and modified chemically and enzymatically essen tially as described in Casu et al (Carbohydrate Letters, Volume 1, pp. 107-114 (1994): "Biologically Active, Heparan Sulfate-like Species by Combined Chemical and Enzy matic modification of the Escherichia coli Polysaccharide K5"). As carriers, pre-activated agarose gel particles (article number 17-3092-09 Amino SepharoseTM, GE Healthcare, Uppsala, Sweden) containing highly cross-linked 6% aga rose were used. The K5 polysaccharide was depolymerised to a molecular weight of 20,000 and coupled to the pre-activated carriers via reductive amination. More specifically, 10 g of the pre activated gel above was first drained, then washed with water and buffer. It was sucked up and transferred to a vial fitted with screw cap. After having shaken the vial, reducing buffer comprising NaCNBH 3 and K5 polysaccharide were added. The reaction mixture was poured onto a glass filter funnel and the functionalised gel so obtained was washed, sucked dry and transferred to an ethanol-containing storage solution. Three different levels of semi-synthetic ligand were tried in order to cover a broad ligand density, since nothing was known concerning the activity of this ligand compared to the conventionally used natural heparin.

WO 2007/058592 16 PCT/SE2006/001297 Level A: (U1154040A): 200 mg of K5 ligand in reaction solution Level B: (U1154040B): 250 mg of K5 ligand in reaction solution Level C: (U1154040C): 300 mg of K5 ligand in reaction solution Example 2: Separation of anti-thrombin from bovine plasma Chemicals Tris(hydroxymethyl)-aminomethane p.a. Trisodium citrate p.a. Sodium chloride p.a. Hydrochloric acid p.a. Ethanol 99.5% (v/v) Bovine plasma (obtained from SVA-Bro, Sweden) Apparatus The separation matrix prepared as described in Example 1 was tested on an HR 5/5 col umn equipped with one LCC-500 control unit, two P-500 pumps, UV-1 monitor (280 nm, HR-10 cell), one MV-7 motor valve and four MV-8 motor valves. One 100 mL su perloop, two 10 mL measuring flasks for each column. The absorbance was measured on an UV-Vis spectrophotometer. Buffers Buffer A: 0.1 M Tris, 0.01 M trisodium citrate, 0.225 M NaCl adjusted to pH 7.4 with 50% (v/v) hydrochloric acid. Buffer B: 0.1 M Tris, 0.01 M trisodium citrate, 0.33 M NaCl adjusted to pH 7.4 with 50% (v/v) hydrochloric acid. Buffer C: 0.1 M Tris, 0.01 M trisodium citrate, 2.0 M NaCl adjusted to pH 7.4 with 50% (v/v) hydrochloric acid.

WO 2007/058592 17 PCT/SE2006/001297 All buffers were filtered with 0.45 gm filter (Millipore) before use. Sample Bovine plasma from five different cows was pooled and frozen in fractions of 45 mL. Two frozen samples were thawn and filtrated through 0.45 Im filter (Millipore SVHV 01015) with a peristaltic pump. 80.0 mL of the filtrated plasma was then mixed with 40.0 mL buffer A and poured into a superloop (100 mL, HR 16/50 column tube) Chromatography Flow rate: 0.50 mL/min Test temperature: 20-26 0 C The column and superloop were connected to the FPLC System. The column was equili brated with 5 mL of buffer A. 45 mL sample was then injected with the superloop. Sub sequent washings were performed with 40 mL buffer A and 15 mL buffer B. The anti thrombin (AT III) was eluted from the column with 9 mL buffer C. The eluate was col lected in a 10 mL measuring flask which was then filled to the mark with buffer C. The absorbance was measured at 280 nm with buffer C as a blank. Evaluation The binding capacity for AT III is calculated from the formulas listed below. Formula I: (F x V x A 280 )/ a = X Formula II: X / (t x h x r2Y Formula III : Y x f= C Where the symbols stands for the following quantities: F* = (AT III capacity plasma x AT III capacity plasma 1 ) / (AT III capacity plasma 1 x AT III capacity plasma > F plasmaa) x AT III capacity plasma 1 / AT III capacity plasma 2 WO 2007/058592 PCT/SE2006/001297 V = eluate volume (10 mL9

A

280 = absorbance at 280 nm a = absorbance coefficient X = mg desorbed AT III H = bed height R = column radius (HR 5/5: 0.25 cm) Y = mg AT III / mL packed gel F = packing factor (mL drained gel / mL packed gel: 0.934) C = binding capacity (mg AT III / mL drained gel) * To eliminate the variation in result depending on the different AT II content of bovine plasma a correction factor is used. The factor F is calculated from the results on a refer ence gel tested with the former plasma plasmaa) and a new batch of plasma (plasma 2 ). The factor is 1.00 for the first plasma batch plasmaao. Results Elemental analysis Gel ID Test 1: S (mg/kg) Test 2: S (mg/kg) Average: S (mg/kg) U 1154040A 6090 6090 6090 U 1154040B 8840 8820 8830 U1154040C 8210 8260 8235 U1154040D (li- 148000 139000 143500 gand) WO 2007/058592 19 PCT/SE2006/001297 Results for AT III binding capacity (mg AT III/mL drained gel) Gel ID Column 1 Column 2 Average U 1154040A 4.00 3.86 3.93 U 1154040B 5.80 5.60 5.70 U1154040C 6.75 5.70 6.22 Reference gel: He- 2.56 2.63 2.49 parin SepharoseTM 6FF (293578) The values found for the commercial product are about 3.0 mg/mL gel with a maximum of 10% difference, usually less than 5%. The protein capacity for the three prototypes with the semi-synthetic K5 polysaccharide as ligand showed higher capacity: The gel U1 54040A presents a capacity 30% higher than the commercial product, while U1 154040B and U1 154040C have an AT III capac ity of 5.7 and 7.2 mg/mL gel respectively corresponding to more than double the capac ity compared to the reference. Thus, the prototypes with immobilized semi-synthetic heparin present all a higher AT III binding capacity, with more than double the value for gel U1 154040C.

CNRPonbDCC\KXG\1727129.1 DOC-3/06/201 I - 19A The reference in this specification to any prior publication (or information derived from it), or to any matter which is known, is not, and should not be taken as an acknowledgment or admission or any form of suggestion that that prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of 5 endeavour to which this specification relates. Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or step or group of integers or steps 10 but not the exclusion of any other integer or step or group of integers or steps.

Claims

1. A method of preparing a separation matrix comprising semi-synthetic heparin ligands by coupling of the semi-synthetic heparin ligands to a carrier, wherein the semi synthetic heparin ligands are provided by (i) expression of K5 polysaccharide in bacterial host cells; (ii) N-deacetylation of GlcNAc residues of the K5 polysaccharide so obtained, and N-sulfation of resulting amino groups; (iii) C5 epimerisation of D-GIcA to L-iduronic acid; and (iv) sulfation of the epimerised product to provide semi-synthetic heparin ligands, and wherein the coupling is performed by reductive amination.

2. A method according to claim 1, wherein in step (ii), the N-deacetylation is provided by adding sodium hydroxide.

3. A method according to any one of claims 1 or 2, wherein the C5 epimerisation is carried out enzymatically by adding glucuronosyl C5 epimerase.

4. A method according to any one of claims 1-3, wherein the sulfation of step (iv) comprises O-oversulfation and selective O-desulfation.

5. A method according to any one of claims 1-4, which also comprises 6-0-sulfation and N-sulfation of the product obtained from step (iv).

6. A method according to any one of claims 1-5, wherein the carrier comprises essentially spherical, porous particles.

7. A method according to claim 6, wherein the porous particles are agarose beads.

8. A method according to any one of claims 1-7, wherein the semi-synthetic heparin ligands have a molecular weight in the range of 10,000-25,000 Dalton. C:\NRPortbI\DCC\KXG\3727129_ DOC-3106/201I - 21

9. A method according to claim 8, wherein the semi-synthetic heparin ligands have a molecular weight of about 20,000 Dalton.

10. A method of separating and/or isolating at least one heparin-binding target compound from other component(s) of a liquid, which method comprises (a) providing a mobile phase comprising the target compound(s); and (b) contacting the mobile phase with a separation matrix comprising semi synthetic heparin ligands to adsorb the target compound(s) to the matrix, wherein the separation matrix comprising semi-synthetic heparin ligands is provided according to any one of claims 1 - 9.

11. A method according to claim 10, further comprising (c) recovering one or more target compounds by contacting an eluent with the separation matrix.

12. A method according to any one of claims 1 to 11, substantially as hereinbefore described.