Oligo- and polysaccharides are chains composed of saccharide units. A recent review, which can not be comprehensive, because of the fast expanding knowledge, has been written by A. Varki (Varki, A. (1993: "Biological roles of oligosaccharides: all the theories are correct", Glycobiology, 3: 97 - 130).

Because of the fundamental role of oligosaccharides in patholog- ical processes the scope of recent research is to develop carbo- hydrate based diagnostics, drugs, and biomaterials. Therefore, there is a great demand to develop a technology, which permits to produce economically oligosaccharides and/or polysaccharides with well characterized molecular weights and structures.

Other biological polymer families, like peptides and proteins, oligonucleotides and polynucleotides are well studied and the synthesis and isolation of oligonucleotides and polypeptides are well developed.

Oligo- and polysaccharides are the least well studied class of biological macromolecules. Numerous classical techniques for the synthesis of carbohydrates have been developed, but these techniques suffer the difficulty of requiring selective protec- tion and deprotection. Organic synthesis of oligosaccharides is further hampered by the lability of many glycosidic bounds, difficulties in achieving regioselective sugar coupling and generally low synthetic yields. These difficulties, together with the difficulties of isolating and purifying carbohydrates has made this area of chemistry a most demanding one.

The use of enzymes as catalysts for the synthesis of carbohy- drates has been demonstrated. Until recently, the major limiting factor to the use of enzymes in carbohydrate synthesis was the limited availability of the broad range of enzymes required to accomplish carbohydrate synthesis. Developments in genetic en- gineering of enzymes will make enzymes more available. At pre- sent a number of glycosyltransferases and other enzymes needed in carbohydrate synthesis are available, but are still expensive (Thiem, J. (1995): "Application of Enzymes in synthetic carbohy- drate chemistry", FEMS Microbiology Reviews. 16: 193 - 211; Wong C.-H. (1992): "Engineering enzymes for chemoenzymatic synthe- sis", 10: 337 - 3a1) Many techniques for the isolation of oligosaccharides are focussed on the identification and isolation of oligosaccharides with complex, branched molecular structures or special binding properties. Carbohydrate binding proteins, e.g. specific anti- carbohydrate antibodies or immobilized lectins, which recognize specific sugar sequences located on glycoproteins, are used to isolate and identify glycoproteins from a complex mixture of cell-surface carbohydrate ligands (Smith D.F. (1981): "Glycoli- pids of cell surfaces: Isolation of specific sugar sequences using carbohydrate-binding proteins", Archives of Biological; Smith, D. F.(1982): "Isolation of specific sugar sequences using carbohydrate-binding proteins", Methods in Enzymology, 83: 241 - 248) . Other examples include the isolation of oligosaccharides with special growth factor binding properties. For this purpose heparan sulfate was enzymatically cleaved, and bound to hepato- cyte growth factor (HGF) after pretreatment on a chromatographic gel permeation column. Elution of the HGF-bound material leads to a set of oligosaccharides (6-20 saccharide units) with homogeneous binding properties to HGF (Lyon, M. Gallagher, J.T.

(1994): "Heparan Sulfate oligosaccharides having hepatocyte growth factor binding affinity", WO-A-94/21689). On the other hand immobilized oligo- and polysaccharides have been used to isolate the corresponding binding proteins.

Linear oligosaccharides are currently isolated in a semi-prepa- rative or preparative scale by degrading large polysaccharides by ultra sonic treatment, enzymatic digestion or by other means.

The mixture of oligo- and remaining polysaccharides is then loaded on a chromatographic gel permeation column and fractions containing the oligosaccharides are collected. For example heparin octasaccharides were cleaved by a heparitinase to yield a hexasaccharide which were further purified by gel permeation chromatography (Choay, S.A., Lormeau, J.-C. (1981): "Oligosac- charides chaines courtes posoedant des proprietes biologiques, leur preparation et leur application en tant que medicaments", EP-A-0 064 452 A). A more complex method was described later, comprising a modification step at the reducing end of the saccharides which gave opportunity to insert a functional group at this site. This opens up the possibility for a better purifi- cation (for example on an ion exchanger matrix) either during or after the enzymatic cleavage. The molecular size of the saccharides prepared this way is exclusively determined by the specifity of the used enzymes. In accordance with this sac- charides prepared by this method are composed of one to six saccharide units (Maruo, S. (1991): "Process for producing saccharide" WO-A-93/10256).

Laboratory methods exist to analyze the sugar composition of larger oligosaccharides (in general up to 25 saccharide units) by different degrading techniques (Jackson, P. (1994): "High resolution polyacrylamide gel electrophoresis of fluorophore- labeled reducing saccharides", Methods in Enzymology, 230: 251 - 265).

Another method for the preparation of oligo- and polysaccharides is the enzymatic synthesis of the polysaccharide using the corresponding polysaccharide synthase and the activated nucleo- side sugars. It was demonstrated that hyaluronan could be synthesized using the streptococcal hyaluronan synthase and UDP-N-acetyl- glucosamine and UDP-glucuronic acid including the regeneration of the UDP-sugars (Lansing, M., Prehm, P., O Regan, M., Martini, I. (1994)) Italian Patent Application, Nr PD94000042; De Luca, C.; Lansing, M., Martini, I., Crescenzi, F., Shen, G.-J., O Regan, M., Wong, C-H. (1995): "Enzymatic synthesis of hyaluronic acid with regeneration of sugar nucleo- tides", Journal of the American Chemical Society, 117: 5869 - 5870). Even with this methological approach there is the need of additional purification of the synthesized products because of unwanted saccharide chain terminations during synthesis. This purification step is necessary for removing byproducts, enzymes and oligosaccharides of unwanted chain length.

The drawback of all these methods is that besides some type of very small oligosaccharides (up to 6 saccharide units) only classes of oligosaccharides with polydisperse molecular weights can be isolated according to the characteristics of the used purification method. Especially with larger oligosaccharides, in general larger than 20 saccharide units, oligosaccharides cannot be resolved using conventional chromatographic methods.

Even in high sophisticated detection systems like a described hyaluronan quantification and detection assay there is no possibility to differentiate between different classes of oligosaccharides (Ghosh, P. (1988): "Method for the Detection and Quantification of Hyaluronan" WO-A-90/07121).

Accordingly, in light of their potential uses and their diffi- culty or the impossibility to obtain oligosaccharides with well characterized molecular weights, there exists a need for a general isolation method for the production of oligosaccharides in an efficient, cost effective, and generally applicable manner.

Because of their pathophysiological role, oligo- and/or poly- saccharides are an important diagnostic tool in order to detect disorders. Furthermore, they are therapeutically useable sub- stances of high therapeutic value.

It is an object of the present invention to provide oligosaccha- rides and polysaccharides with well defined molecular weights, and saccharide compositions which comprise oligo- and/or poly- saccharides.

It is another object of this invention to provide a wide variety of saccharide compositions, including those not found in nature.

It is still another object to provide a process for obtaining an affinity chromatographic media useful in isolating and preparing oligo- and polysaccharides.

These and other objects are achieved by the present invention, which provides in principle a modified affinity chromatographic technique with recombinant binding molecules which allow speci- fic binding of oligo- and polysaccharides and special trimming and elution steps for preparing oligosaccharides, polysacchari- des and other saccharide compositions with a novel grade of homogenity in molecular size and molecular weight.

According to the invention, the objects are attained by a method for the preparation of oligo- and/or polysaccharides providing substances having essentially homogenous length of their carbo- hydrate chain and/or their respective molecular weight. These oligo- and/or polysaccharides, also addressed as carbohydrate polymers, can easily be derived from samples in which precursors are present. Such precursors are more or less heterogenous with respect to the length of the carbohydrate chain or their respect tive molecular weight or do not have the appropriate chain length. The method according to the invention comprises the steps of: immobilizing precursors of the oligo- and/or poly- saccharides on a surface having at least one domain for specific interact ion with at least parts of the precursors of the oligo- and/or polysaccharides in the sample. The sample is contacted with said surface and forms a complex between the one or more domains and the precursors of the oligo- and polysaccharides.

In order to obtain the appropriate fraction of oligo- and/or polysaccharides, which are essentially homogenous with respect to chain length and/or molecular weight the complex is treated with an agent for cleaving carbohydrate chains selectively at the site of the domains. Due to binding of the oligosaccharides to the one or more domains, they remain immobilized. These parts of the precursor molecules projecting beyond said at least one domain are cleaved and the remaining parts are immobilized (at least transiently) on the domain(s).

One aspect of the invention is a method for the preparation of oligo- and/or polysaccharides, which are essentially homogenous with respect to the length of a carbohydrate chain and/or the respective molecular weight from a sample having precursors of the oligo- and/or polysaccharide to be prepared, comprising oligo- and/or polysaccharides which are heterogenous with respect to the length of the carbohydrate chain and/or their respective molecular weight or do not have the appropriate length or weight comprising the steps of: - immobilizing precursors of the oligo- and/or polysaccharides with a carbohydrate chain present in a sample, by reversible binding the oligo- and/or polysaccharides on a surface having at least one domain for a specific interaction with said oligo- and/or polysaccharides, the at least one domain being capable of interacting at least with parts of the precursor of the oligo- and/or polysaccharide chain by contacting the sample with said surface and forming a complex between at least one domain and the precursor of the oligo- and/or polysaccharides, - treating said complex with an agent for cleaving the carbohy- drate chain selectively at the site of the domains for cleava- ge of those parts of the oligo- and/or polysaccharide chain projecting beyond said at least one domain, - removing the cleaved product and release the oligo- and/or polysaccharides bound to the domain, said oligo- and/or polysaccharides having an essentially homogenous length of the carbohydrate chain or an essentially homogenous molecular weight.

The length of the oligo- and/or polysaccharides obtainable by the method of the invention depends on the extension of the one or more domains. If the carbohydrate binding domain is relative- ly small, then the oligo- and/or polysaccharides obtained have a rather short carbohydrate chain length. When the oligo- and/or polysaccharides with longer chain length are prepared, it is preferred to either arrange binding domains in series, which means close to each other and preferably almost equidistant or to provide the surface of the support with at least two binding domains having a defined distance from each other.

If the oligo- and/or polysaccharides to be bound are longer than the distance spanning by one or more domains, at least at one end they are extending the area of the domains forming exten- sions. The agent for cleaving the carbohydrate chains selective- ly cuts such extensions, so that a uniform chain length, defined by the distance spanning by one or more domains is obtained.

Not bound material is washed away with an appropriate washing agent. After treatment with an appropriate agent the oligo- and/or polysaccharides bound by one or more domains having an essentially homogenous length of carbohydrate chain or essen- tially homogeneous molecular weight removed from the domains in particular by elution. The removal can take place by various conditions, for example by changing the pH or the ionic strength of the medium. See also figure 1.

Figure 1 is a schematic presentation of the purification of oligosaccharides. 1) Two carbohydrate binding domains (CBD) are immobilized onto a solid support binding a polysaccharide chain; 2) the bound polysaccharide is treated with an enzyme, which cleaves the polysaccharide at the ends of the CBS while the bound polysaccharide is protected by the CES against the cleav- ing activity of the enzyme; 3) unbound material, saccharides and the enzyme are washed away and 4) finally, the oligosaccha- ride can be eluted.

The not bound material washed away can be brought into contact with another surface having a shorter domain for specific inter- action with at least parts of the non-bound material. Again, if present extensions are removed by treatment with an agent for cleaving the carbohydrate chains selectively and after washing and eluting with appropriate agents, bound oligo- and/or polysaccharide having an essentially homogeneous molecular weight are eluted. By repeating binding, trimming, washing and elution using surfaces having step by step shorter domains it is possible to isolate a set of carbohydrate chains with stepped chain length e.g. of gradually lower molecular weight.

The process involves, in another aspect, the preparation of an affinity chromatographic medium, which consists of at least one, in particular at least two types of carbohydrate binding unit(s) arranged in series and immobilized to a solid support or to provide the surface of the support with a plurality of at least one type of binding domain having a defined distance from another binding domain.

From a sample having poly- and/or oligosaccharides, the poly- and/or oligosaccharides can be isolated by means of the affinity chromatographic medium. To achieve this affinity chromatographic medium is brought into contact with the mixture containing oligo- and/or polysaccharides or oligo- and polysaccharide compositions. Under appropriate conditions the affinity chro- matographic medium binds oligo- and/or polysaccharides with a molecular weight corresponding to the number of carbohydrate binding units linked in series. After binding and washing the affinity chromatographic medium under appropriate conditions oligosaccharides with a defined molecular weight can be isola- ted.

According to the present invention, the term "domain" means a structural element preferably constructed by a peptide or protein which can bind a certain arrangement of saccharides particular in a linear oligo- and/or polysaccharide chain having a defined extension. The structural elements may be separated by spacers which can be a protein chain, as well, having no affinity to the carbohydrate or can be another polymeric struc- ture. A domain may represent one or a plurality of a carbohydra- te binding site(s) of one specific kind. If the domain is only one kind of a carbohydrate binding site, at least two domains must be arranged together. Preferably, the domains may be parts of carbohydrate binding proteins, which bind to oligo- or polysaccharides. These domains bind to a defined number (m) of sugars. There are carbohydrate binding domains (CBD) which bind to hexa- (m=6), hepta- (m=7), octa- (m=8) etc. saccharides. By <BR> <BR> connecting binding domains in series (CUD) n or by attaching an array of functional units at a defined distance from each other on a support surface it is possible to bind saccharides with n x m sugars. By choosing appropriate reaction and process conditions, it is possible to isolate oligo- and/or polysac- charides with n x m saccharide units. This technique allows the preparation of oligo- and polysaccharides with exactly defined molecular size and molecular weight. Especially for the prepara- tion of polysaccharides exceeding 20 saccharide units this method can generate an up to now unknown grade of purity.

Saccharide compositions with a well defined number of sugar units, prepared according to the invention find wide utility in diagnostics, therapeutics, and pharmacological applications.

In particular, the carbohydrate binding units are produced as recombinant proteins.

Preferably, the surface comprising one or more domains is a solid support. In particular, the support may be essentially constructed from an inorganic or organic polymer matrix having covalently, ionically or physically bound said one or more domains, which are preferably linked via a spacer. The chemical structure of a spacer is well known to the skilled person. The spacer serves for keeping the domain in a certain distance from the support. Particularly, the spacer is a protein having no affinity to the carbohydrate or can be another polymeric struc- ture, as well.

The agent for cleaving the carbohydrate chain selectively is an enzyme such as an endo- or exoglycosidase or the like.

Relevant samples to be treated according the method of the invention are biological samples containing hyaluronic acid, heparin, heparan, chitin, chitosan as well as polymannuronic acid.

Subject matter of the present invention is also an oligo- and/or a polysaccharide obtainable by the method outlined above. the oligo- and/or a polysaccharide according to the invention are preferably derived from hyaluronic acid and comprising a termi- nal N-Ac-glucosamine-moiety or glucuronic acid-moiety. Such carbohydrate chain having an essentially defined and essentially homogenous carbohydrate chain length or molecular weight are of significance since their physiological effectivity is related to their length. Different fractions of hyaluronic acid being different in their chain length are interacting with different receptors. This leads to a different physiological response involved with the respective receptor upon binding of a certain oligosaccharide (Montesano, R., Kumar, S., Orci, L., Pepper, M.S. (1996) Synergistic effect of hyaluronan oligosaccharides and vascular endothelial growth factor on angiogenesis in vitro.

Laboratory Investigation, 75, 249 - 262; Ziegler, J. (1996) Hyaluronan seeps into cancer treatment trials. Journal of the National Cancer Institute, 88, 97 - 99.).

Polysaccharides, and in particular sulfated low molecular weight polysaccharides have been found to inhibit the replication of some enveloped viruses, e.g. human immunodeficiency virus (HIV), herpes virus (HSV) or cytomegalovirus (CMV). During research on these saccharides it has been shown that the polysaccharide preparations, obtained by the methods available, are not homoge- nous with respect to their molecular weight. To study the interaction with the proteins of the virus there is a demand for oligo- and polysaccharides with defined molecular weight.

(De Clercq, E. (1992) Anti HIV activity of sulfated polysaccha- rides. In: "Carbohydrates and Carbohydrate Polymers", Yalpani, M (ed) ATL Press, Mt Prospect, Il, USA, 87 - 101.) For example1 oligo- and/or polysaccharides are used as vehicles in drug delivery systems. Liposomes coated with polysaccharides are water soluble and stable against phospholipases. In addi- tion, the chain length of the polysaccharide controls the cha- racteristics of liposomes, e.g. half-life of a liposome and the release of the drug from the liposome (Kawaguchi, Y., Matsukawa, K., Gama, Y., Ishigami, Y. (1992) The effect of polysaccharide chain-length in coating liposomes with partial palmitoyl hyalu- ronates, Carbohydrate Research, 18, 139 - 142).

Therefore, the oligo- and/or polysaccharides obtainable by the present invention are useful tools in form of pharmaceutical compositions. The pharmaceutical composition may contain besides the oligo- and/or polysaccharides of the present inven- tion inert carriers or adjuvents or they can be formulated together with other pharmacologically effective substances in order to support the respective effects.

The kind of administration depends on the type of disease or disorder to be cured. Principally, the compositions of the invention can be applied in any particular manner known in the pharmaceutical art. Depending on the type of administration, special formulations or galenic preparations are to be pre- ferred. Those are well known to the skilled person.

Subject of the present invention are also compounds having one or more domains which are capable of interacting with at least parts of precursors of the oligo- and/or polysaccharide chain.

The compound of the invention comprises a peptide or protein with affinity to said oligo- and/or polysaccharide chain. The definition of the term "domain" has already been given herein above.

According to the invention also a method is provided for detec- tion of oligo- and/or polysaccharides in a sample by reversible binding the oligo- and/or polysaccharides on a surface precoated with recombinant oligo- and/or polysaccharide binding-proteins having at least one domain for a specific interaction with said oligo- and/or polysaccharide eluting the non or weakly bound oligo- and/or polysaccharides to separate special sets of molecules in dependance of their molecular size comprising further the step of detecting these sets of oligo- or polysac- charide from the sample bound to the surface either after elution of the bound oligo- or polysaccharide from the surface or by detecting the oligo- or polysaccharide bound on the surface or by detecting the remaining free binding domains.

A basic method of the present invention employs also the method for detection of oligo- and/or polysaccharides with specific molecular weight classes. For the detection of oligo- and/or polysaccharides with a specific molecular weight class the compound with at least two domains for the specific interaction with said oligo- and/or polysaccharide is immobilized on a solid support. If there is a sample in which a class of molecular weight of said oligo- and/or polysaccharide should be detected, or the concentration should be determined, the sample is brought in contact with said immobilized compound. After interaction of the oligo- and/or polysaccharide with said immobilized compound unbound material is removed. Furtheron, oligo- and/or polysaccharides weakly bound to said immobilized compound are removed by treating them with increasing salt concentrations or any other method capable for removing weakly bound oligo- and/or polysaccharides. The said immobilized compounds, now free of any oligo- and/or polysaccharide are detected by bringing them into contact with a labelled oligo- and/or polysaccharides, which allows a specific detection reaction. A specific detection reaction can be an enzymatic reaction, a fluorescence or a radiofluorographic reaction or any other reaction, capable indicating the presence of said free immobilized compound.

By comparing the reaction of the sample containing an unknown concentration of a molecular weight class of said oligo- and/or polysaccharide with the total amount of said oligo- and/or polysaccharide in the sample and with a known concentration of said oligo- and/or polysaccharide the fraction of the molecular weight class can be determined.

In a preferred embodiment, the carbohydrate binding units are immobilized by attachment to a solid support and the oligo- and/or polysaccharides to be contacted therewith are added thereto.

Preferred procedures for immobilization of carbohydrate binding units include immobilization of the carbohydrate binding units amino groups onto solid support oxirane groups or onto cyanogen bromide activated "Sepharose". Other immobilization procedures may be used, such as those described by Scouten (Scouten, W.H.

(1987) A survey of enzyme coupling techniques. Methods in Enzymology, 135, 30 - 65) or others).

It will be appreciated that impairment of the binding site of the carbohydrate binding unit (domain) due to immobilization should be avoided. During the immobilization process the carbo- hydrate binding unit (domain) may be protected by the carbohy- drate which binds carbohydrate binding unit (domain). For exam- ple, hyaluronan carbohydrate binding units (domains) may be protected with hyaluronan during the immobilization. In this way, contaminating proteins, if present, are not protected in any way during the immobilization.

The immobilized carbohydrate binding unit (domain) can be used as an affinity chromatographic medium.

The present invention is further illustrated by way of the following examples.

A) Reagents and Molecular Cloninq DNA-Manipulations, transformation and cultivation of the appro- priate host strains, expression and the isolation of the recom- binant products were carried out using standard methodology as described by Sambrook, J.et al. (1989) and are known to those skilled in the art. DNA modifying reactions were performed according to the manufactors instructions.

The choice of the expression vector depends on the host strain used. Host strains are preferably microorganisms, e.g. a proka- ryont or a yeast. Preferable, bacterial strains, like Escheri- chia coli strains and appropriate expression vectors, are used and are known by those skilled in the art.

B) Isolation of Recombinant Proteins For the production of recombinant proteins host strains are transformed with the appropriate expression vector and grown in an appropriate medium. Cells are harvested by centrifugation, resuspended in an appropriate buffer and the cells are lysed by high-pressure homogenization (French-Pressure Cell). The lysate is centrifugated to remove remaining particles.

Various methods purifying recombinant proteins which can be used in accordance with the invention have been published. The recombinant proteins may be isolated by conventional chromato- graphic techniques or by its binding properties to the specific saccharide. In actual practice of the invention the carbohydrate binding unit is expressed as an recombinant protein with an peptide tag which allows the purification by affinity chromatog- raphy Homogenity and purity of the isolated protein was analyzed by SDS-PAGE (Laemmli, 1970) C) Affinitychromatoqraphic Medium In a preferred embodiment, the carbohydrate binding units are immobilized by attachment to a solid support and the oligo- and/or polysaccharides to be contacted therewith are added thereto.

Example 1 Cloning a hyaluronan two domain binding protein The expression vector constructed is schematically drawn in figure 2: Plasmid prPHA-20 for the expression of a two domain hyaluronan binding protein. bla, tetR, ori, tetp/O, ompA, G1-B, tlpp denote the b-lactamase gene, the tet repressor gene, the origin of replication from the pUC family of plasmids, the tet promotor/operator sequence, the signal sequence of the outer membrane protein A, the globular 1 B region of aggrecan, and the terminator of transcription.

Fragments of the gene coding for a hyaluronan binding protein (Doege, K.J., Sasaki, M., Kimura, T., Yamada, Y. (1991) Complete coding sequence and deduced primary structure of the human cartilage aggregating proteoglycan, aggrecan. Journal of Biolo- gical Chemistry, 266, 894 - 902) is precisely fused to the bacterial outer membrane protein (OmpA) signal sequence encoded by the pASK75 plasmid (Biometra GmbH, Göttingen, Germany).

The DNA- sequence coding for the Gl-B hyaluronan binding domain of aggrecan was amplified by PCR using following primers (StuI- primer 5'acagcgaggcctccctggaaagtcgtg3 , SstI- primer: 3 ccctcgagctcctcggcgaagcag5 , BamHI-primer: 5 acagc ggatccaccctggaagtcg3', PstI- primer: 5 ctcacctgcagtctcctcggc gaag3'). Template cDNA is prepared from RNA extracted from human cartilage. RNA is isolated according to Adams et al. (Adams, M.E., Huang, D.Q., Yao, L.Y., Sandell, L.J. (1992) Extraction and isolation of mRNA from adult articular cartilage. Analytical siochemistry, 202, 89 - 95) . cDNA synthesis was performed according the PCR Application Manual from Boehringer Mannheim (Boehringer Mannheim GmbH, Mannheim, Germany, 1995) The ampli- fied DNA was cloned in pASK75 using the restriction sites StuI, Ssti and BamHI and PstI creating a plasmid expressing two G1-s domains separated by a small peptide spacer.

StuI-Primer 1: The PCR product amplified using the primers StuI and SstT was ligated to pASK75 digested with StuI and SstI resulting in the plasmid prPHA-10. The plasmid prPHA-10 is brought in E. coli JM110 by electroporation and is prepared using standard plasmid purification protocols. The second PCR product using the primers BamHI and PstI is ligated to prHA-10 digested with BamHI and PstI resulting in the plasmid prPHA-20, which is propagated in E. coli JM110.

Figure 3 shows coding sequence of the G1-B domain of human aggrecan and primers for PCR amplification and cloning in pASK75 whereas figure 4 shows the coding sequence for the cloned duplicate G1-s domain starting from the RBS of the tetO/ of prHA-20 Example 2 Isolation of a recombinant protein A preculture is obtained by inoculating 5 ml LE medium contain- ing 100 Hg/ml ampicillin with a single colony of E. coli JM110 harbouring the plasmid prPHA-20. 250 ml LE medium containing 100 g/ml ampicillin is inoculated with 2.5 ml of the preculture and shaken (175 rpm) at 25or. When the culture reaches an optical density at 600 nm of 0.5 25 l of anhydrotetrazyklin (2 mg/ml) is added and shaking is continued for 2 hours. All further manipulations are done at 40C. E. coli JM110 cells expressing the recombinant rPHA-20 protein are harvested by centrifugation (4,500 x g, 20 min) . After discarding the super- natant the cells are resuspended in 10 ml lysis buffer (100 mM Tris/HCl, 1 mM EDTA, pH 8.0) and incubated for 30 min on ice to isolate the proteins from the periplasmic space. Speroblatsa and cell debris are removed by centrifugation at 18,000 x g for 20 minutes at 40C. After ultracentrifugation at 100,000 x g for 1 hr at 4"C the cleared supernatant is brought to 500 mM NaCl and is loaded on a column (d: 16 mm) containing 10 ml Ni2+NTA agarose resin (Qiagen GmbH, Heiden, Germany). The resin is washed 500 ml phosphate buffer I (20 mM NaHPG4, 500 mM NaCl, pH 7.8) and phosphate buffer II (20 mM NaHPO4, 500 mM NaCl, pH 6) at a flow rate of 0,5 ml/min. The recombinant rPHA-20 protein is eluted from the resin using imidazole buffer (20 mM NaHPOd, 500 mM NaCl, 300 mM imidazole, pH 6) at a flow rate of 0.5 ml/min. Fraction containing the rPHA-20 protein were pooled together, dialyzed for 16 hrs against 5 1 (20 mM NaPPO4, pH 7.5) and concentrated by ultrafiltration using an Amicon ultrafiltra- tion apparatus and a membrane with molecular weight cut-off of 30,000 kDalton.

Example 3 Preparing a affinity chromatographic matrix The isolated rPHA20 protein (10 mg/ml) is mixed with hyaluronic acid, partially degraded by ultrasonic treatment (final concen- tration 0,01 mg hyaluronic acid/ml) in phosphate buffer (20 mM NaPPO4, pH 7.5) prior to immobilization to protect the carbohy- drate binding domains. Immobilization is carried out in the same phosphate buffer. The protected protein is added to dry Euper- git-C (100 mg protein/g beads) and the mixture is left to stand at room temperature with occasional mixing to ensure adequate infiltration. The extend of immobilization is monitored by loss of protein from the supernatant. A solution of protein without beads is used as control. After the immobilization is judged to be complete (90t loss of protein from the supernatant) the beads are washed with washing buffer (20 mM NaHPG4, 500 mM NaCl, pH 7.5) to remove unbound or loosely bound protein. Washing is continued until the absorbance at 280 nm was less than 0.1. The beads are stored at 40C in storage buffer(20 mM NaHP04, 0.01 NaN3, pH 7.5) until required.

Example 4.

Isolation of hyaluronan oligosaccharides with 20 sugar units A chromatographic column (1.6 cm x 5 cm) is packed with the Eupergit-C beads to which the recombinant rPHA-20 protein recognizing 20 sugar units is immobilized (rPHA20-beads). The column is loaded with a digested preparation of hyaluronic acid (0.1 M phosphate buffer, pH 7.5; 0.25 ml/min) . After washing the column with phosphate buffer to remove unbound material the column is incubated with hyaluronoglucosaminidase (E.C.

3.2 1.35) in the same buffer for 30 min at room temperature.

Again the column is washed with phosphate buffer and the column is washed with a gradient from 0 to 2 M NaCl in phosphate buffer to remove unspecific bound material. Oligosaccharides are eluted from the column with 4 M guanidine chloride in phosphate buffer.

The samples are dialyzed again 0.1 M NaCl to remove guanidine chloride and are lyophilized.

Example 5 Determination of the fraction of hyaluronic acid oligosacchari- des smaller than 20 saccharide unites in a hyaluronic acid sample The wells of a conventional microwell plate (e.g. Nune GmbH, Wiesbaden, Germany) are coated with the above described modified byaluronan binding proteins (rPHA-20) . Therefore, the wells of the microwell plate were incubated for 30 min at 370C with 10 Ug/ml of rPHA-20 in coating buffer (for example: 100 mM Tris/HCl, pH 7.4; 50 mM NaCl). After incubation the wells were washed three times with coating buffer containing 0.1W Tween 20. Finally, the wells were incubated with 3% (w/v) BSA in coating buffer for 30 min at 370C and washed three times with coating buffer containing 0.1t Tween.

In lane 1 of the microwell plate 50 yl of standard solutions of HA (1 Hg/ml, 0.1 yg/ml etc.) in buffer I (0.1 M glycine, 10 mM MgCl2, pH 7.5) are pipetted. To the well 1A only buffer I is pipetted which serves as a blank value. A serial dilution of a sample with unknown concentration of HA is prepared in buffer I and 50 jil of each dilution step is pipetted in the corresponding lanes 2 and 3, Lane 2 serves for the determination of the total amount of HA, lane 3 for the determination of the fraction of small oligosaccharides. After incubation for 30 min the well were washed with buffer I and land 2 is washed additio- nally with buffer I containing an appropriate salt concentration of NaCl preferable 2 M NaCl, to elute oligosaccharides smaller than 20 saccharide units. For the detection of the remaining free binding sites HA conjugated to alkaline phosphatase in buffer I is added to the well of the microwell plate. After incubation for 30 min at room temperature the whole solution is removed from the wells and 50 l of buffer II (0.1 M glycine, 10 mM MgCl2, pH 10) and 50 M1 of the detection buffer III (15.2 mM p-nitrophenyl phosphate) is added to each well. After in- cubation for 30 min at 370C to each well 100 ijl 0.2 N NaOH is added and mixed. The absorbance at 405 nm is recorded using a microwell plate reader. By comparing the absorbance of the standards with the absorbance of the sample, the concentration of hyaluronic acid and the fraction of oligosaccharides smaller than 20 saccharide units in the sample can be calculated.

SEQUENCE LISTING (1) GENERAL INFORMATION: (i) APPLICANT: (A) NAME: Manfred Lansing (B) STREET: Duelmenerstr. 155 (C) CITY: Dorsten-Wulfen (E) COUNTRY: Germany (F) POSTAL CODE (ZIP): 46286 (A) NAME: Jochen Uhlenkueken (B) STREET: Paulstr. 22 (C) CITY: Muenster (E) COUNTRY: Germany (F) POSTAL CODE (ZIP) 48151 (A) NAME: Gerd Schmidt (B) STREET: Heinrich-Hertz-Str. 13 (C) CITY: Recklinghausen (E) COUNTRY: Germany (F) POSTAL CODE (ZIP): 45657 (ii) TITLE OF INVENTION: Method for reversible immobilizing oligo- and/or polysaccharides (iii) NUMBER OF SEQUENCES: 9 (iv) COMPUTER READABLE FORM: (A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS (D) SOFTWARE: PatentIn Release #1.0, Version #1.30 (EPO) (2) INFORMATION FOR SEQ ID NO: 1: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 348 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1: GACAGCGAGG CCACCCTGGA AGTCGTGGTG AAAGGCATCG TGTTCCATTA CAGAGCCATC 60 TCTACACGCT ACACCCTCGA CTTTGACAGG GCGCAGCGGG CCTGCCTGCA GAACAGTGCC 120 ATCATTGCCA CGCCTGAGCA GCTGCAGGCC GCCTACGAAG ACGGCTTCCA CCAGTGTGAC 180 GCCGGCTGGC TGGCTGACCA GACTGTCAGA TACCCCATCC ACACTCCCCG GGAAGGCTGC 240 TATGGAGACA AGGATGAGTT TCCTGGTGTG AGGACGTATG GCATCCGAGA CACCAACGAG 300 ACCTATGATG TGTACTGCTT CGCCGAGGAG ATGGAGGGTG AGGTCTTT 348 (2) INFORMATION FOR SEQ ID NO: 2: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 116 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2: Asp Ser Glu Ala Thr Leu Glu Val Val Val Lys Gly Ile Val Phe His 1 5 10 15 Tyr Arg Ala Ile Ser Thr Arg Tyr Thr Leu Asp Phe Asp Arg Ala Gln 20 25 30 Arg Ala Cys Leu Gln Met Ser Ala Ile Ile Ala Thr Pro Glu Gln Leu 35 40 45 Gln Ala Ala TyrtGlu Asp Gly Phe His Gln Cys Asp Ala Gly Trp Leu 50 55 60 Ala Asp Gln Thr Val Arg Tyr Pro Ile His Thr Pro Arg Glu Gly Cys 65 70 75 80 Tyr Gly Asp Lys Asp Glu Phe Pro Gly Val Arg Thr Tyr Gly Ile Arg 85 90 95 Asp Thr Asn Glu Thr Tyr Asp Val Tyr Cys Phe Ala Glu Glu Met Glu 100 105 110 Gly Glu Val Phe 115 (2) INFORMATION FOR SEQ ID NO: 3: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 24 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: other nucleic acid (A) DESCRIPTION: /desc = "Primer" (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3: ACAGCGGATC CACCCTGGAA GTCG 24 (2) INFORMATION FOR SEQ ID NO: 4: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 24 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: other nucleic acid (A) DESCRIPTION: /desc = "Primer" (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4: ACAGCGAGGC CTCCCTGGAA GTCG 24 (2) INFORMATION FOR SEQ ID NO: 5: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 25 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: other nucleic acid (A) DESCRIPTION: /desc = "Primer" (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5: CTCACCTGCA GTCTCCTCGG CGAAG 25 (2) INFORMATION FOR SEQ ID NO: 6: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 23 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: other nucleic acid (A) DESCRIPTION: /desc = "Primer" (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6: CCCTCGAGCT CCTCGGCGAA GCA 23 (2) INFORMATION FOR SEQ ID NO: 7: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 770 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7: GAGGGCAAAA AATGAAAAAG ACAGCTATCG CGATTGCAGT GGCACTGGCT GGTTTCGCTA 60 CCGTAGCGCA GGCCTCCCTG GAAGTCGTGG TGAAAGGCAT CGTGTTCCAT TACAGAGCCA 120 TCTCTACACG CTACACCCTC GACTTTGACA GGGCGCAGCG GGCCTGCCTG CAGAACAGTG 180 CCATCATTGC CACGCCTGAG CAGCTGCAGG CCGCCTACGA AGACGGCTTC CACCAGTGTG 240 ACGCCGGCTG GCTGGCTGAC CAGACTGTCA GATACCCCAT CCACACTCCC CGGGAAGGCT 300 GCTATGGAGA CAAGGATGAG TTTCCTGGTG TGAGGACGTA TGGCATCCGA GACACCAACG 360 AGACCTATGA TGTGTACTGC TTCGCCGAGG AGCTCGGTAC CCGGGGATCC ACCCTGGAAG 420 TCGTGGTGAA AGGCATCGTG TTCCATTACA GAGCCATCTC TACACGCTAC ACCCTCGACT 480 TTGACAGGGC GCAGCGGGCC TGCCTGCAGA ACAGTGCCAT CATTGCCACG CCTGAGCAGC 540 TGCAGGCCGC CTACGAAGAC GGCTTCCACC AGTGTGACGC CGGCTGGCTG GCTGACCAGA 600 CTGTCAGATA CCCCATCCAC ACTCCCCGGG AAGGCTGCTA TGGAGACAAG GATGAGTTTC 660 CTGGTGTGAG GACGTATGGC ATCCGAGACA CCAACGAGAC CTATGATGTG TACTGCTTCG 720 CCGAGGAGAC TGCAGGCAGC GCTTGGCGTC ACCCGCAGTT CGGTGGTTAA 770 (2) INFORMATION FOR SEQ ID NO: 8: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 759 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: cDNA (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8: ATGAAAAAGA CAGCTATCGC GATTGCAGTG GCACTGGCTG GTTTCGCTAC CGTAGCGCAG 60 GCCTCCCTGG AAGTCGTGGT GAAAGGCATC GTGTTCCATT ACAGAGCCAT CTCTACACGC 120 TACACCCTCG ACTTTGACAG GGCGCAGCGG GCCTGCCTGC AGAACAGTGC CATCATTGCC 180 ACGCCTGAGC AGCTGCAGGC CGCCTACGAA GACGGCTTCC ACCAGTGTGA CGCCGGCTGG 240 CTGGCTGACC AGACTGTCAG ATACCCCATC CACACTCCCC GGGAAGGCTG CTATGGAGAC 300 AAGGATGAGT TTCCTGGTGT GAGGACGTAT GGCATCCGAG ACACCAACGA GACCTATGAT 360 GTGTACTGCT TCGCCGAGGA GCTCGGTACC CGGGGATCCA CCCTGGAAGT CGTGGTGAAA 420 GGCATCGTGT TCCATTACAG AGCCATCTCT ACACGCTACA CCCTCGACTT TGACAGGGCG 480 CAGCGGGCCT GCCTGCAGAA CAGTGCCATC ATTGCCACGC CTGAGCAGCT GCAGGCCGCC 540 TACGAAGACG GCTTCCACCA GTGTGACGCC GGCTGGCTGG CTGACCAGAC TGTCAGATAC 600 CCCATCCACA CTCCCCGGGA AGGCTGCTAT GGAGACAAGG ATGAGTTTCC TGGTGTGAGG 660 ACGTATGGCA TCCGAGACAC CAACGAGACC TATGATGTGT ACTGCTTCGC CGAGGAGACT 720 GCAGGCAGCG CTTGGCGTCA CCCGCAGTTC GGTGGTTAA 759 (2) INFORMATION FOR SEQ ID NO: 9: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 252 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: unknown (D) TOPOLOGY: unknown (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: NO (1) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9: Met Lys Lys Thr Ala Ile Ala Ile Ala Val Ala Leu Ala Gly Phe Ala 1 5 10 15 Thr Val Ala Gln Ala Glu Leu Glu Val Val Val Lys Gly Ile Val Phe 20 25 30 His Tyr Arg Ala Ile Ser Thr Arg Tyr Thr Leu Asp Phe Asp Arg Ala 35 40 45 Gin Arg Ala Cys Leu Gln Asn Ser Ala Ile Ile Ala Thr Pro Glu Gln 50 55 60 Leu Gln Ala Ala Tyr Glu Asp Gly Phe His Gln Cys Asp Ala Gly Trp 65 70 75 80 Leu Ala Asp Gln Thr Val Arg Tyr Pro Ile His Thr Pro Arg Glu Gly 85 90 95 Cys Tyr Gly Asp Lys Asp Glu Phe Pro Gly Val Arg Thr Tyr Gly Ile 100 105 110 Arg Asp Thr Asn Glu Thr Tyr Asp Val Tyr Cys Phe Ala Glu Glu Leu 115 120 125 Gly Thr Arg Gly Glu Thr Leu Glu Val Val Val Lys Gly Ile Val Phe 130 135 140 His Tyr Arg Ala Ile Ser Thr Arg Tyr Thr Leu Asp Phe Asp Arg Ala 145 150 155 160 Gln Arg Ala Cys Leu Gln Asn Ser Ala Ile Ile Ala Thr Pro Glu Gln 165 170 175 Leu Gln Ala Ala Tyr Glu Asp Gly Phe His Gln Cys Asp Ala Gly Trp 180 185 190 Leu Ala Asp Gln Thr Val Arg Tyr Pro Ile His Thr Pro Arg Glu Gly 195 200 205 Cys Tyr Gly Asp Lys Asp Glu Phe Pro Gly Val Arg Thr Tyr Gly Ile 210 215 220 Arg Asp Thr Asn Glu Thr Tyr Asp Val Tyr Cys Phe Ala Glu Glu Val 225 230 235 240 Gin Gly Ser Ala Trp Arg His Pro Gln Phe Gly Gly 245 250