EP1406638A2 - Macrophage receptor agonist or antagonist - Google Patents

Abstract

A substance capable of direct interaction with Dectin-1, said substance not inhibiting the binding of T cells to cells expressing Dectin-1, is useful because Dectin-1 has now been identified as the therapeutically effective β glucan receptor.

Description

MACROPHAGE RECEPTOR

FIELD OF THE INVENTION

The present invention relates to the macrophage β-glucan receptor.

The macrophage β-glucan receptor is an innate pattern recognition receptor involved in recognition of β-glucans and the phagocytosis of Candida albicans, for example.

In general, β-l,3-D-glucans are biological response modifiers with potent effects on the immune system (Williams, D. L., et al, Clinical Immunotherapy 5, 392-399 (1996)), including anti- tumour (Ross, G. D., et al, Immunopharmacology 42, 61-74 (1999)) and anti-infective activities, protecting against fungal (Williams, D. L., et al, J Reticuloendothel Soc 23, 479-490 (1978)), bacterial (Kokoshis, P. L., et al, Science 199, 1340-2 (1978)), viral (Itoh, W., Mediators of Inflammation 6, 267-269 (1997)) and protozoal infections (Cook, J. A., et al, Infect Immun. 37, 1261-9 (1982)).

Despite the therapeutic potential of the β-glucan polymers, little is known about the mechanisms underlying their immunomodulatory abilities. The macrophage β-glucan receptor is thought to play a crucial role in mediating these responses (Czop, J. K., Pathol. Immunopathol. Res. 5, 286-96 (1986)), but it has defied all efforts at identification.

The existence of a macrophage β-glucan receptor was originally identified over twenty years ago, and was described as a phagocytic receptor for particulate activators of the alternative complement pathway (Czop, J. K., et al, J. Immunol 120, 1132-8 (1978)). Receptors with similar functions were subsequently identified on a variety of other leukocytes, including neutrophils and natural killer cells (Williams, D. L., Mediators of Inflammation 6, 247-250 (1997)). Although receptors, such as complement receptor 3 (CR3) (Ross, G. D., Cain, J. A. & Lachmann, P. J., J. Immunol 134, 3307-15 (1985)) and lactosylceramide (CDwl7) (Zimmerman, J. W. et al, J. Biol. Chem. 273, 22014-20 (1998)), have been implicated in the binding of β-glucans, the identity of the macrophage β-glucan receptor has remained elusive. The immunomodulatory activities of β-glucans are thought to stem from their ability to activate leukocytes, resulting in the stimulation of phagocytic activity and in the production of reactive oxygen intermediates, inflammatory mediators and cytokines (Williams, D. L., et al, Clinical Immunotherapy 5, 392-399 (1996); Czop, J. K., Pathol Immunopathol Res 5, 286-96 (1986)). Although the molecular mechanisms underlying these effects are unclear, cellular responses to zymosan and yeast pathogens have been shown to require at least two receptors; a phagocytic receptor and the signalling Toll-like receptors (Underbill, D. M. et al. Nature 401, 811-5 (1999)). Toll-like receptors sample the phagosome and trigger an inflammatory response, mediated in part through NF-κB activation (Underhill, D. M. et al. Nature 401, 811-5 (1999)), while ligand binding to the macrophage β-glucan receptor has been shown to result in NF-κB activation (Battle, J. et al. Biochem Biophys Res Commun 249, 499-504 (1998)).

US-A-6,046,158 (Ariizumi et al.) discloses murine molecules Dectin-1 and Dectin-2, and their possible relationships with human molecules. No reference is made to macrophage β-glucan receptors.

Human Dectins are disclosed in WO 01/68848, although no activity is associated with the molecules.

US-A-5,504,079 (Jamas et al.) discloses ways of using modified yeast cell wall glucans to treat patients who are at risk of infection.

US-A-5,057,503 (Czop et al.) discloses heptaglycosides which can interact β-glucan receptors, but without identifying the receptors.

By using a novel screening method, we have now surprisingly found that Dectin-1 has identity with the macrophage β-glucan receptor.

Accordingly, with this knowledge, it is now possible to accurately target the macrophage β-glucan receptor and to exploit its interactions with other molecules and entities. SUMMARY OF THE INVENTION

Thus, in a first aspect, the present invention provides a substance capable of direct interaction with Dectin-1, and which inhibits the binding of zymosan to Dectin-1.

The nature of the inhibition of zymosan binding may be competitive or non-competitive, and this will generally depend on where the substance binds. If it binds, or interacts, with Dectin-1 at the β-glucan binding site, then it is likely that the nature of the inhibition will be competitive. If the binding, or interaction, is in the proximity of the β-glucan binding site, then the inhibition is more likely to be non-competitive.

In an alternative aspect, the present invention provides a substance capable of direct interaction with Dectin-1, said substance not inhibiting the binding of murine T cells to cells expressing murine Dectin-1.

Dectin-1 has now been found to display all of the essential characteristics previously ascribed to the macrophage β-glucan receptor. It recognises a range of glucan polymers possessing β-1,3 and/or β-1,6 linkages; it has a sensitivity to trypsin; is independent of metal ions for activity; and has an ability to mediate the binding and phagocytosis of zymosan and yeast particles, including the opportunistic pathogen, Candida albicans. Dectin-1 (Ariizumi, K. et al. J. Biol. Chem. 275, 20157-67 (2000)) has previously been identified on a specific class of leukocytes - dendritic cells - and is associated with the binding and activation of T cells in mice.

Substances of the present invention may be anything that binds, or otherwise interacts, with Dectin-1, and it will be appreciated that such substances will generally have a protein or saccharide structure. Human isoforms of Dectin-1 are indicated in nucleotide sequences SEQ ID NO'S 1 and 3, encoding for the polypeptide sequences SEQ ID NO'S 2 and 4 respectively. The isoforms of SEQ ID NO. 1, encoding the polypeptide of SEQ ID NO. 2, possess no stalk region. The nucleotide sequence of murine Dectin 1 is indicated in SEQ ID NO. 5, encoding the polypeptide of SEQ ID NO. 6. These sequences are also available at www.ncbi.nhn.nih.gov and have the accession numbers (human dectin-1 isoforms) AF313468 and AF313469, respectively, and AF262985 (mouse Dectin 1). However, the present invention extends to any Dectins having β glucan receptor activity, such as those disclosed in the art prior to our recognition that Dectin-1 was the β glucan receptor. DETAILED DESCRITION OF THE INVENTION

Binding to Dectin-1 may serve to activate or block receptor activity, such as by agonism, antagonism, or simple steric hindrance. We have established that Dectin-1 is a mediator of the activity of β glucans, characteristically leading to increased levels of TNF (Tumour Necrosis Factor), a standard marker for the β glucan receptor, i a preferred embodiment, compounds of the present invention are capable of causing activation, or partial activation, of macrophages through their interaction with Dectin-1. Partial activation, or partial stimulation, effectively serves to put the macrophage on notice, so that a secondary insult, or aggravated insult, which otherwise might not have served to activate the macrophage, is sufficient to effect activation. For example, soluble β glucans may suffice to prime the macrophages, with a bacterial insult then resulting in macrophage activation. Antagonists and blockers may typically be used to reduce the effect of activators, or to prevent or reduce anticipated activation.

Proteinaceous substances of the present invention comprise amino acid residues, whether naturally occurring or synthetic and, in addition, may comprise further side chains or intra-chain residues which, themselves, may be tailored to interact with Dectin-1. It is particularly preferred that substances of the invention are not derived from S. cerevisiae.

Proteinaceous substances capable of interaction with Dectin-1 may comprise antibodies, for example. Such antibodies are preferably targeted to the extracellular domain of Dectin-1 and, particularly, to the C-type lectin fold. Depending on the intended use of such antibodies, their specificity may either be for a unique area of the lectin fold away from the β-glucan binding region, or may be the β-glucan binding region, or its environs.

Antibodies which bind the C-type lectin fold may be used, for example, to simply detect the presence of Dectin-1, or cells expressing Dectin-1. hi certain circumstances, binding such areas of Dectin-1 may also serve to block the glucan binding site, and such possibilities will be discussed below in respect of antibodies binding this site.

If it is desired to establish whether a tissue expresses Dectin-1, then it maybe appropriate to label the antibody in a manner such that, after application and removal of unbound antibody, bound antibody can be visualised. Such an effect may be achieved by tagging the antibody with an enzyme which can subsequently react with a predetermined substrate for visualisation, or the antibody may simply be bound to, for example, a fluorescent marker.

Antibodies which bind Dectin-1 may also be bound to a column, for example, in order to separate cells expressing Dectin-1 from cells which do not. This may be advantageous, for example, if it is desired to generate a culture of a patient's macrophages which, in turn, may be useful in the treatment of certain conditions in the patient. For example, if the patient is about to undergo an immune comprising regimen, then isolating Dectin-1 expressing cells may be useful in the prophylaxis or treatment of any opportunistic infections or conditions arising as complications.

It will also be appreciated that such antibodies can be used to isolate Dectin-1 itself, or in combination with other substances, such as membrane fragments, from preparations comprising such.

Antibodies binding the β-glucan binding site may be used in the manner described above, but may also be used to either activate or block the activity of the β-glucan receptor. Whether an antibody activates or blocks the receptor will depend on the idiotype of the antibody. In general, however, such antibodies will block the activity of the β-glucan receptor, unless the binding region directly binds the glucan binding site.

In general, it is advantageous to activate macrophages, as they are particularly active in fighting infections, including viral, bacterial, protozoal, parasitic and cancerous cells. However, unnecessarily activated macrophages can cause other problems and, in particular, are associated with conditions such as asthma, hay fever and allergic dermatitis.

Accordingly, antibodies which block the β-glucan receptor may be useful in the treatment of conditions associated with activated macrophages, especially chronically activated macrophages, and may be administered by any suitable means, such as inhalant spray, injection or topical administration, such as lotions, ointments or creams.

Antibodies may be obtained by any suitable means, such as by introduction of a suitable human antigen into the mouse or rabbit, for example, generating a monoclonal antibody, then genetically modifying the result to obtain humanised antibody. Suitable antibodies binding to Dectin-1 can be made by techniques available in the art. The antibodies can be chimaeric or humanised, and may take the form of a fragment, notably a binding domain of the antibody.

An antibody against Dectin-1 may be produced using methods which are generally known in the art. In particular, purified Dectin-1 polypeptide (also simply referred to as Dectin-1 herein) may be used to produce antibodies or to screen libraries of pharmaceutical agents to identify those which specifically bind Dectin-1. Antibodies to Dectin-1 may also be generated using methods that are well known in the art. Such antibodies may include, but are not limited to, polyclonal, monoclonal, chimeric, and single chain antibodies, Fab fragments, and fragments produced by a Fab expression library.

For the production of polyclonal antibodies, various hosts including goats, rabbits, rats, mice, humans, and others may be immunised by injection with Dectin-1 or with any fragment or oligopeptide thereof which has immunogenic properties. Rats and mice are preferred hosts for downstream applications involving monoclonal antibody production. Depending on the host species, various adjuvants may be used to increase immunological response. Such adjuvants include, but are not limited to, Freund's, mineral gels, such as aluminium hydroxide, and surface active substances such as lysolecithin, pluronic polyols, polyanions, peptides, oil emulsions, EXH, and dinitrophenol. Among adjuvants used in humans, BCG (bacilli Calmette-Guerin) and Corynebacterium parvum are especially preferable. The methods for antibody production and analysis are described in Harlow, E. and Lane, D. (1988; Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor NY).

It is preferred that the oligopeptides, peptides, or fragments used to induce antibodies to Dectin-1 have an amino acid sequence consisting of at least about 5 arnino acids, and, more preferably, of at least about 14 amino acids. It is also preferable that these oligopeptides, peptides, or fragments are identical to a portion of the amino acid sequence of the natural protein and contain the entire amino acid sequence of a small, naturally occurring molecule. Short stretches of Dectin-1 amino acids may be fused with those of another protein, such as KLH, and antibodies to the chimeric molecule may be produced.

Monoclonal antibodies to Dectin-1 may be prepared using any technique which provides for the production of antibody molecules by continuous cell lines in culture. These include, but are not limited to, the hybridoma technique, the human B-cell hybridoma technique, and the EBV- hybridoma technique (Kohler, G. et al. (1975) Nature 256:495-497; Kozbor, D. et al. (1985) J. Immunol. Methods 81:31-42; Cote, RJ. et al. (1983) Proc. Natl. Acad. Sci. 80:2026-2030; and Cole, S.P. et al. (1984) Mol. Cell Biol. 62:109-120, respectively).

In addition, techniques developed for the production of "chimaeric antibodies," such as the splicing of mouse antibody genes to human antibody genes to obtain a molecule with appropriate antigen specificity and biological activity, can be used (Morrison, S.L. et al. (1984) Proc. Natl. Acad. Sci. 81:6851-6855- Neuberger, M.S. et al. (1984) Nature 312:604-608; and Takeda, S. et al. (1985) Nature 314:452-454). Alternatively, techniques described for the production of single chain antibodies may be adapted, using methods known in the art, to produce Dectin-1 -specific single chain antibodies. Antibodies with related specificity, but of distinct idiotypic composition, may be generated by chain shuffling from random combinatorial immunoglobulin libraries (Burton D.R. (1991) Proc. Nat. Acad. Sci. 88:10134-10137).

Antibodies may also be produced by inducing in vivo production in the lymphocyte population or by screening immunoglobulin libraries or panels of highly specific binding reagents as disclosed in the literature (Orlandi, R. et al. (1989) Proc. Natl. Acad. Sci. 86: 3833-3837; Winter, G. et al. (1991) Nature 349:293-299).

Antibody fragments which contain specific binding sites for Dectin-1 may also be generated. For example, such fragments include, but are not limited to, F(ab')2 fragments produced by pepsin digestion of the antibody molecule and Fab fragments generated by reducing the disulfide bridges of the F(ab')2 fragments. Alternatively, Fab expression libraries may be constructed to allow rapid and easy identification of monoclonal Fab fragments with the desired specificity (Huse, W.D. et al. (1989) Science 246:1275-1281).

Various methods such as Scatchard analysis in conjunction with radioimmunoassay techniques maybe used to assess the affinity of antibodies for Dectin-1. Affinity is expressed as an association constant, K_a, which is defined as the molar concentration of Dectin-1 -antibody complex divided by the molar concentrations of free antigen and free antibody under equilibrium conditions. The K_a determined for a preparation of polyclonal antibodies, which are heterogeneous in their affinities for multiple Dectin-1 epitopes, represents the average affinity, or avidity, of the antibodies for Dectin-1. The K_a determined for a preparation of monoclonal antibodies, which are monospecific for a particular Dectin-1 epitope, represents a true measure of affinity. High-affinity antibody preparations with K_a, ranging from about 10⁹ to 10¹² L/mole are preferred for use in immunoassays in which the Dectin-1-antibody complex must withstand rigorous manipulations. Low-affinity antibody preparations with K_a, ranging from about 10 to 10 L/mole are preferred for use in immunopurification and similar procedures which ultimately require dissociation of Dectin-1, preferably in active form, from the antibody (Catty, D. (1988) Antibodies, Volume 1: A Practical Approach, IRL Press, Washington D.C.; Liddell,. J. E. and Cryer, A. (1991) A Practical Guide to Monoclonal Antibodies, John Wiley & Sons, New York, NY).

The titre and avidity of polyclonal antibody preparations may be further evaluated to determine the quality and suitability of such preparations for certain downstream applications. For example, a polyclonal antibody preparation containing at least 1-2 mg specific antibody/ml, preferably 5-10 mg specific antibody/ml, is preferred for use in procedures requiring precipitation of Dectin-1 -antibody complexes. Procedures for evaluating antibody specificity, titre, and avidity, and guidelines for antibody quality and usage in various applications, are generally available (Catty, supra; Coligan, supra).

The polypeptides of the invention or their fragments or analogues thereof, or cells expressing them, can also be used as immunogens to produce antibodies immunospecific for Dectin-1 polypeptides. The term "immunospecific" means that the antibodies have substantially greater affinity for the polypeptides of the invention than their affinity for other related polypeptides in the prior art.

Antibodies generated against Dectin-1 polypeptides can be obtained by administering the polypeptides or epitope-bearing fragments, analogues or cells to an animal, preferably a non- human, using routine protocols. For preparation of monoclonal antibodies, any technique which provides antibodies produced by continuous cell line cultures can be used. Examples include the hybridoma technique (Kohler, G. and Milstein, C, Nature (1975) 256:495-497), the trioma technique, the human B-cell hybridoma technique (Kozbor et al, Immunology Today (1983) 4:72) and the EBV-hybridoma technique (Cole et al, MONOCLONAL ANTIBODIES AND CANCER THERAPY pp. 77-96. Alan R. Liss, Inc., 1985).

Techniques for the production of single chain antibodies (US-A-4,946,778) can also be adapted to produce single chain antibodies to polypeptides of this invention. Also, transgenic mice, or other organisms including other mammals, may be used to express humanised antibodies. Saccharides which bind the β-glucan receptor will generally target the binding site itself. Previous work (Czop, supra) with the β-glucan receptor, before its identification, indicates that the minimum length for a binding saccharide is a heptaglycoside. We have established that the saccharide linking must be β-1, 3 or β-1, 6. There appears to be no preference for when the linking is β-1, 3 or β-1, 6, and it is not necessary for any one binding area of a polysaccharide to comprise all β-1, 3 or all β-1, 6 linkages.

In addition, the monomeric hexose types making up the oligomers are important, the primary hexose recognised by the β-glucan receptor apparently being glucose, but at least galactose may also serve as a useful monomer. Furthermore, simple heptaose (a heptaglycoside) does not activate the β-glucan receptor, and it is generally preferred that the saccharide be at least a decamer, although octameric and nonameric saccharides are also envisaged. Indeed, the invention envisages saccharides having up to several hundred hexose residues, although such molecules tend to be unwieldy and unnecessary, but may be preferred when obtained from natural sources, such as zymosan.

Compounds of the present invention are preferred to have macrophage activation properties. Without being bound by theory, it is believed that multimerisation of Dectin-1 is necessary in order to activate the macrophages. In this respect, although laminarin binds Dectin-1 strongly, it does not appear able to activate macrophages. By contrast, glucan phosphate is larger, and is capable of partially activating macrophages, while zymosan is larger still, and can completely activate macrophages, all via the β-glucan receptor, or Dectin-1, as we have established it to be.

Thus, preferred saccharide lengths, and lengths of substances which mimic them, are in excess of the laminarin polymeric number. The chain length is more preferably approaching that of glucan phosphate, or greater, where partial activation is desired. Where full activation is desired, then even greater chain lengths, in the region of those observed for zymosan, are preferred. Accordingly, preferred compounds may contain between about 15 and 100 monomeric residues but, more preferably, contain between about 70 and 500 residues, or more.

It is generally preferred to use substances whose synthesis is closely controlled. These may either be completely synthetic, in the case of saccharides, for example, or obtained from engineered cells, such as in the case of peptides and especially antibodies. Substances such as zymosan may then be excluded. It will be appreciated that the present invention does not extend to known compounds per se, although novel uses therefor are contemplated. Thus, zymosan, laminarin and glucan phosphate, for example, are not claimed, per se.

The substance of the invention may be part of a larger saccharide, such as starch or other polysaccharide, which may be chosen to break down in situ, or which may be chosen to resist catabolism. In the former case, where the substance of the invention is a saccharide, the β-1, 3 and β-1, 6 linkages tend to provide a substantial degree of resistance to catabolism, so that just the core substance may be left after ingestion, or other form of administration.

The present invention extends to branched oligo- and poly- saccharides comprising saccharide units of the invention, as well as to other molecules which are equivalent thereto, or which carry said units or their equivalent. For example, a suitable carrier protein may be glycosylated with units of the invention, or their equivalent.

In accordance with the present invention, substances which interact with the β-glucan receptor include substances which comprise an oligosaccharide chain comprising in excess of 7 hexose sub-units, each sub-unit linked to the adjacent sub-unit by a β-1, 3 or β-1, 6 linkage.

It will be appreciated that substances comprising such oligosaccharides may, themselves, simply be saccharides, or, for example, may be proteinaceous materials comprising such oligosaccharides. As used herein, reference to saccharides as substances of the invention incorporates reference to any other substance of the invention, unless otherwise apparent.

Although the minimum octameric saccharide or larger oligosaccharide may be incorporated directly into a protein structure, it is generally preferred that proteins comprising such saccharides are linked thereto at recognised glycosylation sites. Suitable glycosylation sites are well known in the art.

Depending on the nature of the protein, glycosylation may be effected by expression in suitable cells, or may be effected synthetically, after the protein has been isolated. Alternatively, the proteinaceous material may itself be entirely synthetically manufactured, and may be little more than an oligopeptide, for example. Substances which interact with the β-glucan receptor may be readily identified. For example, in one simple assay, a preparation of Dectin-1 and the substance to be assayed can be mixed, and binding of the proposed ligand determined by interference with antibody binding, for example.

More preferably, it is preferred to provide substances which interact with membrane bound Dectin-1, and this may be achieved either by whole cell preparations, such as NIHT3 cells expressing Dectin-1, or by synthetic membranes, such as liposomes. The complexity of any such preparation will depend on what the substance is being assayed for, and suitable activities for the assay include simple binding to the Dectin-1 molecule, up to levels of phagocytic stimulation, h the latter case, whole cell preparations may be necessary.

AΆ advantage of the present invention is that the absolute structure of the β-glucan receptor is now known, and the receptor itself is readily prepared and isolated, making it a simple matter of routine to identify, prepare and isolate substances which interact with the receptor.

In particular, substances which activate the receptor are useful in many circumstances and, in general, where it is desired to stimulate the immune response to infection.

Activation of the β-glucan receptor serves to stimulate the proliferation of macrophages, and can also provide long term protection against, for example, E. coli or Staphylococcus aureus.

Such substances have been found to provide particularly good activity in combination with antibiotics for example, where the effect appears to be synergistic, and they can also serve as excellent adjuvants.

Other indications for macrophage stimulants of the invention include anti-cancer preparations, as well as serving as antioxidants, in their own right, haematopoietic stimulants and an anti-ageing agents. Further uses include:

-prophylactics against diseases as mentioned above, especially for post operative patients and those with depressed immune systems;

-increased wound healing;

-treatment of Candida and other fungal related infections;

-for the treatment of transplantation rejection; and -protection against radiation induced damage.

In particular, it is preferred that substances of the invention capable of interacting with Dectin-1 are ligands for the glucan binding site. Most preferred are peptide mimetics or other peptide substances, which comprise an oligosaccharide chain having in excess of seven β-1, 3 or β- 1, 6 linked hexose sub-units therein. The oligosaccharide may be singly present on such peptide compounds, but is preferably present in multiple sites on the peptide. The peptide may be engineered to comprise as many glycosylation sites as desired, and the number of such sites is preferably at least five, and more preferably at least 10, and may be anywhere up to fifty, or more.

The advantage of such compounds primarily lies in the fact that they are not as susceptible to catabolic influences in the body as simple oligosaccharides. hi addition, it is possible to tailor such substances to have other activities so that, for example, an anti-cancer antibody having one or more β-glucan receptor ligands thereon is envisaged by the present invention.

In general, the present invention envisages other agonists and antagonists of the β-glucan receptor binding site, and these may be readily determined, as described above. Such substance are not necessarily limited to proteinaceous or saccharide substances, and will be readily apparent to those skilled in the art.

Substances for administration to the human or animal body may be prepared in any suitable manner, and will generally comprise a suitable vehicle for a substance. The preparation should be therapeutically or pharmaceutically acceptable and may comprise any other suitable ingredients, such as substance to render the preparation isotonic in the case of injection, for example, or other suitable excipients, emollients, flavouring substances, thickeners, sweeteners, adjuvants or other desirable or pharmaceutically effective agents.

The preparations of the invention for administration to the human or animal body, preferably the human body, may be formulated for administration by any suitable route, including oral administration, such as tablets, capsules and delayed or sustained release devices, dermal administration, such as by patches, lotions, ointments or creams, suppositories, sprays for inhalation, and injections, such as subcutaneous, intraperitoneal, intramuscular and intravenous injections. In order to identify the molecular nature of the macrophage β-glucan receptor, a method based on the inherent ability of receptors to recognise their respective cognate ligands was devised. This involved the use of retrovirally-generated stable cDNA expression libraries which could be screened using fluorescently labelled ligand-bearing particles, for example. Subsequently, a single receptor which bound zymosan was isolated. The DNA sequence obtained identified the receptor as Dectin-1, a small type-II membrane receptor (~28kDa) possessing an extracellular C-type lectin- like domain fold and a cytoplasmic domain having an immunoreceptor tyrosine-based activation motif (ITAM) (Ariizumi, K. et al. J Biol Chem 275, 20157-67 (2000)).

Thus, it will be appreciated that the present invention provides means for assaying for substances of the invention, comprising the use of Dectin-1, or a substance having the β glucan binding properties thereof, and assaying for binding with a test substance. The Dectin-1 of this method may be presented in any suitable manner, such as embedded in a membrane. The membrane may typically be artificial, although this is not essential, but provides for less variability. Substances which meet the desired binding criteria, such as agonism or blocking, for example, may then be selected.

Substances that bind Dectin-1 may then be further classified, for example, according to whether they activate, or partially activate, macrophages, for instance.

Previously, Dectin-1 had only been reported on dendritic cells. However, re-examination of Dectin-1 distribution showed that Dectin-1 was expressed, not only on dendritic cells, but also by all macrophage populations, as well as monocytes and neutrophils. In fact, Dectin-1 was found to be expressed in many more tissues than previously reported, with the highest expression levels in the liver, lung and thymus.

Dectin-1 functions as a pattern recognition receptor, recognising a variety of fungal and plant β-1,3 and β-l,6-linked glucans. Laminarin and glucan phosphate, a structurally defined immunologically active β-glucan, bind strongly, both having previously been shown to bind to the β-glucan receptor on monocytes and macrophages.

Zymosan binding by Dectin-1 transfectants is trypsin sensitive, a well documented characteristic of the β-glucan receptor. The β-glucan receptor is involved in the uptake of soluble β-glucans and the phagocytosis of unopsonised zymosan. Dectin-1 is capable of mediating the actin-dependent phagocytosis of zymosan and is enriched to zymosan phagosomes. Truncating the cytoplasmic tail inhibits or prevents phagocytosis of zymosan. Tagging of Dectin-1 also shows that it has a vesicular distribution in the cell, consistent with the observation that the β-glucan receptor is a recirculating receptor and present in a mobilisable intracellular pool.

Cells expressing Dectin-1 bind the β-glucan expressing organisms, Saccharomyces cerevisiae and C. albicans, for example, in a β-glucan dependent fashion.

The C-type lectin domain fold possessed by the β-glucan receptor is similar to those of the natural killer (NK) cell C-type lectin-like receptors which typically recognise protein ligands (Weis, W. I., et al, Immunol Rev 163, 19-34 (1998)), although this interaction may also involve the recognition of carbohydrates (Parham, P., Curr Biol 10, R195-7 (2000)). Dectin-1 and the NK cell C-type lectin domains lack the residues known to be involved in calcium co-ordination, which are required for carbohydrate binding in the classical Ca²⁺ dependent C-type lectins (Weis, W. I, et al, Immunol Rev 163, 19-34 (1998)). This is consistent with the fact that the zymosan binding ability of Dectin-1 is independent of metal ions.

The lectin binding site on the macrophage β-glucan receptor is distinct from the site known to bind T-lymphocytes (Ariizumi, K. et al, J Biol Chem 275, 20157-67 (2000)), suggesting that at least murine Dectin-1 is capable of recognising both endogenous and exogenous ligands. In humans, the macrophage β-glucan receptor/ Dectin-1 (the two terms are used interchangeably herein), interacts with about 1 in 4 T cell subsets, although the significance of this is unknown.

Thus, Dectin-1 appears to have two ligand binding sites; one which recognises an endogenous ligand on T-cells, which may be a protein, and a second binding site for exogenous carbohydrate ligands.

A human homologue of Dectin-1 was recently deposited in Genbank (accession number AY009090) and the genomic sequence partially covered by the human genome project (contig accession number NT_009704). Although similar to the murine receptor, this human homologue lacks an extracellular stalk region and does not contain any N-linked glycosylation sites. The predicted size of this receptor (22kD), lack of glycosylation sites and potential tyrosine phosphorylation motif in the cytoplasmic tail corresponds closely to a 20kD subunit found in a large complex described by Czop and Kay (J Exp Med 173, 1511-20 (1991)). in anti-antiidiotype investigations of the β-glucan receptor. These investigations did not succeed in identifying the receptor, and the line of investigation was not followed up.

According to the genomic sequence, the human homologue is located on chromosome 12pl3.3, a region syntenic with mouse chromosome 6, and places both the mouse and human receptors in the NK gene complex of C-type lectin-like receptors (Sobanov, Y. etal., hnmunogenetics 49, 99-105 (1999)).

Brief description of the Figures

Figure 1. Dectin-1 binds zymosan particles and is expressed on macrophages and dendritic cells, a, The binding of labelled zymosan (green) particles to actin (red) stained NIH3T3 cells expressing Dectin-1 can be clearly visualised by fluorescence microscopy, b, Cartoon structure of Dectin-1 showing the predicted glycosylation sites (lollipop structures), the C-type lectin-like domain (CL) and an ITAM sequence (Y) in the cytoplasmic tail (Ariizumi, K. et al, J Biol Chem 275, 20157-67 (2000)) c, Dectin-1 is expressed in bone marrow derived macrophages and in macrophage cell lines, as determined by RT-PCR analysis with dihydrofolate reductase (DHFR) as a control, d, Northern blot analysis reveals that Dectin-1 is expressed in a variety of tissues.

Figure 2. Zymosan binding to Dectin-1 transfected cells can be inhibited by β-1,3 and/or β-1,6 linked glucans. Dectin-1 transfected cells were pretreated with either 500μg/ml (open bars), lOOμg/ml (dotted bars) or lOμg/ml (striped bars) of the various carbohydrates before fluorescently labelled zymosan particles (50 per cell) were added. The amount of zymosan binding was quantitated by fluorimetry and expressed relative to an uninhibited control (100%; dashed line). Background zymosan binding to untransfected IH3T3 was normally around 10% (dotted line). Carbohydrates possessing side chain linkages^a and/or mixed linkages⁰ are indicated, and are composed of glucan polymers, except galactan (galactose monomer⁰) and mannan (mannose monomer^d).

Figure 3. Dectin-1 can mediate the binding and phagocytosis of both zymosan and yeast particles in transfected cells, a, internalisation of fluorescently labelled zymosan (green) via actin (red)-based phagocytic cups, b, GFP-Dectin-1 transfectants bind zymosan particles under phase microscopy, c, Under fluorescence microscopy, GFP-Dectin-1 (green) can be seen to accumulate around the fluorescently labelled zymosan (red) particles and to possess a vesicular distribution within the cell, d, Dectin-1 mediates the β-glucan dependent binding of S. cerevisiae and C. albicans, comparable to binding by RAW264.7 macrophages. Results are expressed relative to RAW264.7. e, Dectin-1 mediates the non-opsonic phagocytosis of fluorescently labelled C. albicans (green) via actin (red)- based phagocytic cups.

Figure 4, shows that, with increasing exposure of Dectin-1 in RAW macrophages (denoted as "bgr"), enhanced levels of zymosan binding are observed, compared to normal cells ("w '/wild type), thereby confirming that Dectin-1 binds zymosan (Bgr = cells which over-express Dectin-1).

Figure 5 shows that TNF production is increased in cells which over-express Dectin-1 ("bgr") compared to normal cells with wild type (wt) expression levels of Dectin-1. In particular, it shows that TNF production is increased as levels of available zymosan are increased.

The present invention will now be illustrated by the following, non-limiting Example.

EXAMPLE 1

Receptor screening, isolation and cloning

A RAW264.6 (ATCC# ΗB-71) cDNA library was generated using a cDNA synthesis kit (Stratagene) and directionally cloned into the pFBneo retro viral vector (Stratagene). The library was transfected into Phoenix ecotropic packaging cells; viral supernatants were harvested 48 hrs after transfection and used to transduce NIH3T3 fibroblasts (ATCC# CRL-1658). Fluorescein- labelled zymosan particles (Molecular Probes) were added to the transduced cells (100 particles/ cell) and allowed to bind for 1 hr. After washing, positive (zymosan-binding) cells were determined by fluorescence microscopy and isolated using cloning cylinders. A homogeneous population was subsequently obtained through repeated re-isolation of positive cells, the transduced gene isolated by PCR and then sequenced using pFBneo specific primers (Stratagene).

Once isolated, the gene encoding the postulated β-glucan receptor was recloned into pFBneo and fransduced back into NIH3T3 cells, in a manner similar to that described above, to generate the stable β-glucan receptor expressing cell line used in these experiments. The N-terminal GFP-Dectin-1 fusion was generated by the in-frame cloning of a full length PCR, from the second codon (K) to the stop codon into the GFP encoding plasmid, pEGFP-C3 (Clontech). The GFP-Dectin-1 coding sequence was subcloned into pFBneo and then transduced into NIH3T3 cells, as described above, to generate a stable GFP-Dectinl expressing cell line.

RT-PCR and Northern blotting

Complementary DNA was generated by reverse transcription of DNase-treated total RNA using the Advantage RT-PCR kit (Clontech). To detect Dectin-1 , PCR was performed with primers spanning the region encoding the extracellular portion from amino acids A66-L244. DHFR primers (Stratagene) were used as a control. Tissue expression was analysed using a 12 tissue polyA mRNA Northern blot (Origene Technologies), probed with the entire Dectin-1 coding sequence.

Receptor-ligand binding assays

To examine the effects of carbohydrates on zymosan binding, cells were pre-treated for 20 minutes with the various carbohydrates before the addition of fluorescein-labelled zymosan (50 particles /cell). After a further 1 hr incubation the cells were washed, lysed in 3% Triton XI 00 and the relative fluorescence determined using a fluorometer (Fluoroskan II; Titertek). Using these experimental conditions, the relative fluorescence gives a linear response (R²=0.98), reflecting the number of beads bound by the cells. Carbohydrates for these analyses were obtained from Sigma (laminarin, barley β-glucan, mannan), Megazyme (pullulan, galactan, lichenan,), CarboMer (curdlan), Pharmacia (dextran), Calbiochem (pustulan) and Seikagaku Corporation (laminariheptaose). Glucan phosphate was a gift from Dr. David Williams (East Tennessee State University, USA) (MuUer, A. et al, J Immunol 156, 3418-25 (1996)). A similar approach was used for the S. cerevisiae (Clontech AH109) and C. albicans (ATCC# 18804) binding assays, except that 20 heat killed, rhodamine green-X labelled (Molecular Probes) yeast particles were used per cell.

Immunofluorescence microscopy

For immunofluorescence, cells grown overnight on glass cover slips were incubated with either fluorescein or Texas red-labelled zymosan particles (Molecular probes) or with rhodamine green-X labelled C. albicans, as described above. The cells were subsequently fixed, permeabilised and stained with TRITC-labelled phalloidin (Sigma). After mounting, the cells were examined by confocal laser scanning microscopy using a Bio-Rad MRC-1024 mounted on a Diaphot 200 microscope (Nikon).

To isolate the macrophage β-glucan receptor, a RAW264.7 library was screened using the β- glucan-rich particle, zymosan (Di Carlo, F. J. et al, Science 127, 756-757 (1958)), and a single receptor which bound zymosan (Figure la) was isolated. The DNA sequence obtained identified the receptor as Dectin-1, a small type-II membrane receptor (~28kDa) possessing an extracellular C-type lectin-like domain fold and a cytoplasmic domain having an immunoreceptor tyrosine-based activation motif (ITAM) (Ariizumi, K. et al, J Biol Chem 275, 20157-67 (2000)) (Figure lb).

The isolation of a zymosan binding receptor from a macrophage cDNA library was consistent with the description of the β-glucan receptor as a receptor on monocytes and macrophages (Czop, J. K., Pathol Immunopathol Res 5, 286-96 (1986) and Muller, A. et al, J Immunol 156, 3418-25 (1996)), but contrasted with the reported dendritic cell specificity of Dectin-1 (Ariizumi, K. et al, J Biol Chem 275, 20157-67 (2000)). Therefore, the cellular and tissue expression profiles of this gene were re-examined and found to be expressed, not only on dendritic cells, but also in every macrophage population examined (Figure lc). Furthermore, Dectin-1 was expressed in many more tissues than previously reported, with the highest expression levels in the liver, lung and thymus (Figure Id).

To determine the substrate specificity of the macrophage β-glucan receptor, the ability of various carbohydrates to block binding of zymosan to H3T3 cells modified to express Dectin-1 (Figure 2) was assayed. This analysis showed that Dectin-1 functioned as a pattern recognition receptor (Medzhitov, R. et al, Curr Opin Immunol 9, 4-9 (1997)), recognising a variety of fungal and plant β-1,3 and β-l,6-linked glucans. The receptor was not inhibited by monosaccharides (not shown) or by carbohydrates possessing different linkages, such as β-l,4-linked cellulose or α-1,6 linked dextran. The partial inhibitory effect of yeast mannan observed at high concentrations was due to its known contamination with β-glucan (Goldman, R., Exp Cell Res 174, 481-90 (1988)). Laminarin and glucan phosphate, a structurally defined immunologically active β-glucan (Williams, D. L. et al, Carbohydr Res 219, 203-13 (1991)), were the most effective inhibitors and both have previously been shown to bind to the β-glucan receptor on monocytes and macrophages (Muller, A. et al, J Immunol 156, 3418-25 (1996) and Mueller, A. et al, Glycobiology 10, 339-46 (2000)). Although the unit ligand for the β-glucan receptors has been defined as a heptaglucoside (Janusz, M. J. et al, J Immunol 142, 959-65 (1989)), laminariheptaose did not inhibit zymosan binding to Dectin-1 fransfectants, nor was laminariheptaose inhibition of zymosan binding to either bone-marrow derived or RAW264.7 macrophages detectable (not shown).

Zymosan binding by Dectin-1 fransfectants was trypsin sensitive (not shown), a well documented characteristic of the β-glucan receptor (Czop, J. K., Pathol Immunopathol Res 5, 286- 96 (1986)).

The β-glucan receptor has also been shown to be involved in the uptake of soluble β-glucans (Muller, A. et al,. J Immunol 156, 3418-25 (1996)) and the phagocytosis of unopsonised zymosan (Czop, J. K. et al, J Immunol 134, 2588-93 (1985)). Dectin-1 was found to be able to mediate the actin-dependent phagocytosis of zymosan (Figure 3 a) and, by using a green fluorescent protein (GFP)-tagged receptor, the enrichment of the receptor to zymosan phagosomes was visualised (Figure 3b,c). Dectin-1 had a vesicular distribution in the cell, consistent with the observation that the β-glucan receptor is a recirculating receptor and present in a mobilisable intracellular pool (Tapper, H., et al, Biochem J 306, 829-35 (1995)).

The β-glucan receptor has also been implicated in the recognition and phagocytosis of intact Saccharomyces cerevisiae (Giaimis, J. et al, J Leukoc Biol 54, 564-71 (1993)). as well as the fungal pathogen C. albicans (Janusz, M. J. et al, Immunology 65, 181-5 (1988)), both of which possess β-1,3 and β-l,6-linked glucans within their cell walls (Bartnicki-Garcia, S., Annu Rev Microbiol 22, 87-108 (1968)). Both heat killed (Figure 3d) and live (not shown) S. cerevisiae yeasts, as well as heat killed C. albicans conidia (Figure 3d), were bound by Dectin-1 fransfectants in a β-glucan dependent fashion. Furthermore, C. albicans conidia were internalised (Figure 3e), demonstrating that Dectin-1 can mediate the non-opsonic phagocytosis of this opportunistic pathogen.

Soluble recombinant Dectin-1 has also been reported to bind to and stimulate T-lymphocyte proliferation (Ariizumi, K. et al, J Biol Chem 275, 20157-67 (2000)) in mice. Using a whole cell binding assay, we found that the binding of T-cells to NIH3T3 cells expressing Dectin-1 was not inhibited by β-glucans (not shown). Thus, Dectin-1 appears to have two ligand binding sites; one which recognises an endogenous ligand on T-cells (Ariizumi, K. et al, J Biol Chem 275, 20157-67 (2000)), which may be a protein, and a second binding site for exogenous carbohydrate ligands. The finding that the human receptor was capable of binding both zymosan and C. albicans in a β-glucan dependent fashion (not shown), confirmed that it was the functional equivalent of Dectin-1.

Example 2

Expression of the Dectin-1 β-Glucan Receptor On Macrophages.

Using specific carbohydrate inhibitors, we have shown that a β-glucan receptor (BGR), but not the MR, is the predominant receptor involved in this process in primary Macrophages. Furthermore, non-opsonic zymosan binding was unaffected by genetic CD1 lb deficiency or a blocking mAb against CR3, demonstrating that CR3 was not the β-glucan receptor mediating this activity.

To address the role of the β-glucan receptor, Dectin-1, a novel anti-Dectin-1 mAb, 2A11, was generated. Using this mAb, we showed that Dectin-1 was almost exclusively responsible for the β-glucan-dependent, non-opsonic recognition of zymosan by primary macrophages. These findings define Dectin-1 as the leukocyte β-glucan receptor.

Methods

Cells

Thioglycollate- (Tg) or Biogel-elicited peritoneal and bone-marrow derived (BMDM) macrophages were isolated from C57BL/6 mice by standard procedures and cultured overnight in 24 well plates. Animals were kept and handled according to institutional guidelines. C57BL/6 CDllb^";"mice, generated as previously described (Melo, M.D., et al, 2000, Cell. Immunol 205:13- 2319), were used. Cells were maintained in RPMI with 10% heat-inactivated FCS, 50 IU/ml penicillin G, 50μg/ml streptomycin and 2mM glutamine (RPMI medium); except for BMDM, which were cultured in RPMI medium supplemented with 15% (v/v) L-cell conditioned medium, as a source of M-CSF (Hume, D.A., et al, J. Cell. Physiol 117:189-194). BMDM were used 5 to 7 days after isolation and culture. Generation ofmAbs against Dectin-1.

The mAb, 2A11, specific for Dectin-1, was generated by immunisation of Fischer rats with NIH3T3 cells transduced with full length Dectin-1 (Brown, G.D., et al, Nature 413:36-37 and subsequent boosting with soluble recombinant, hemagglutinin (HA)-tagged, Dectin-1. Recombinant Dectin-1 was harvested from supernatants of the human 293T fibroblast cell line transfected with pcDNA3.1 (Invifrogen) encoding an N-terminal leader and HA-tag sequence fused to the extracellular portion of Dectin-1 (amino acids 66 to 244). Splenic B-cells from immunised rats were then fused with the Y3 rat myeloma cell line (Galfre, G., et al, 1979. Nature 277:131- 133), according to standard protocols. Hybridoma supernatants were initially screened by ELISA against the soluble recombinant form of Dectin-1. The mAb 2A11 (IgG2b) was subsequently selected based on its ability to recognise unfixed and unpermeabilised cells transduced with Dectin-1, hence detecting an extracellular epitope. Other antibodies used in this study were 5C6 (anti-CR3; (Rosen, H. et al, 1987. J. Exp. Med. 166:1685-1701)) and a rat IgG2b isotype control.

Flow cytometry and immunoprecipitation.

Flow cytometry was performed according to conventional protocols. Cells were examined by three-color FACS analysis using biotin-conjugated 2A11, phycoerythrin-conjugated F4/80 (Serotec) and FITC-conjugated 5C6. Biotin- or FITC- conjugated rat IgG2b were used as isotype controls. Allophycocyanin-conjugated streptavidin (BD-Pharmingen) was used to detect the biotin conjugates.

For immunoprecipitation, soluble recombinant HA-tagged Dectin-1 was incubated with 2A11 or rat IgG2b isotype control and then captured with sheep anti-rat IgG magnetic beads (Dynal). Western blotting, after SDS-PAGE, was performed according to standard protocols. The mouse mAb HA11 (anti-HA; Covance) and donkey anti-mouse IgG horseradish peroxidase conjugate (Jackson) were used to detect the HA-tagged protein using the ECL chemiluminescence substrate (Amersham).

Fluorescent zymosan binding assays.

The fluorescence-based binding assays using FITC-labelled zymosan were performed as described (Brown, G.D., et al, 2001. Nature 413:36-37 and Willment, J.A., et al, 2001. J. Biol. Chem. 276:43818-43823), except that in all experiments the cells were maintained at 4°C, to prevent the local release of opsonins, including complement (Ezekowitz, R.A., et al, 1985. J Clin. Invest. 76:2368-2376 and Ezekowitz, R.A., et al, 1984. J. Exp. Med. 159:244-26024). In brief, macrophages were plated at 2.5 x 10⁵ cells/well in 24 well plates in culture medium overnight. The cells were cooled to 4°C and washed three times with pre-chilled culture medium. Zymosan-FITC (Molecular Probes) was added to the cells at a ratio of 25 particles/cell for 1 hour on ice. After incubation with the carbohydrate or antibody inhibitors, described below, unbound zymosan was removed by extensive washing with medium and cells were then lysed with 3% Triton X-100. FITC in lysates was quantified using a Tifretek Fluoroskan II (Labsystems Group (UK) Ltd). Unless otherwise stated the amount of fluorescence was normalised to the uninhibited control and expressed at percent relative fluorescence. All experiments were repeated at least three times.

To obtain opsonised particles, FITC-labelled zymosan was incubated with neat normal mouse serum for 30 min at 37°C and then washed extensively in RPMI medium before use. Binding of opsonised FITC-labelled zymosan to cells was performed as for the unopsonised zymosan, described above.

Carbohydrate and antibody inhibition experiments.

All carbohydrates were obtained from Sigma, except for glucan phosphate, a structurally- defined and biologically active glucan, which was prepared as described (Muller, A., et al, 1996. J. Immunol. 156:3418-3425). To perform inhibition experiments, carbohydrates (lOOμg/ml) were added to washed cells for 20 min prior to the addition of FITC-labelled zymosan.

The antibody inhibition experiments were performed similarly, except that the cells were incubated with the mAbs in PBS for 1 hr, and then washed extensively prior to the determination of zymosan binding.

To determine the role of Dectin-1 in the direct recognition of soluble β-glucans, carbohydrates were added to the cells at lOOμg/ml for 1 h at 4°C, after which the cells were washed and stained in situ. The cells were then detached using a cell scraper and analysed by flow cytometry, as described above. To internalise the receptor, the washed cells were warmed at 37°C for 7 min and then immediately returned to 4°C. Results

The recognition of unopsonised zymosan is predominantly mediated by a β-glucan receptor.

The role of various receptors in recognition of zymosan was investigated. The ability of receptor-specific carbohydrates to inhibit the binding of FITC-labelled zymosan to a variety of primary macrophages was examined in the absence of complement. Laminarin, a soluble fungal derived β-glucan widely used to define leukocyte β-glucan receptor activity (Czop, J.K., et al, 1985. J. Immunol. 134:2588-2593; Giaimis, J., Y. et al, 1993. JLeukoc. Biol 54:564-571; and Thornton, B.P., et al, 1996. J. Immunol. 156:1235-1246), was found to exert a strong inhibitory effect on the recognition of unopsonised zymosan (data not shown). This is in accordance with early studies, which demonstrated a role for a β-glucan receptor in the non-opsonic recognition of zymosan (Czop, J.K., et al, 1985. J. Immunol. 134:2588-2593 and Goldman, R. 1988. Exp. Cell. Res. 174:481-490). In contrast, the addition of mannan, also previously shown to inhibit the recognition of unopsonised zymosan (Sung, S.S., et al, 1983. J. Cell. Biol. 96:160-166 and Giaimis, J., et al, 1993. JLeukoc. Biol. 54:564-5717), had no effect.

The results also show that inhibition of zymosan binding by mannan is a result of contamination of the yeast mannan preparations with β-glucans. This confirms the findings of earlier studies (Czop, J.K, et al, 1985. J. Immunol. 134:2588-2593 and Goldman, R. 1988. Exp. Cell. Res. 174:481-490). Therefore, the mannose receptor does not play a significant role in this process. Methyl glucoside, an inhibitor of the CDl lb lectin binding domain (Thornton, B.P., et al, 1996. J. Immunol. 156:1235-1246), had no effect, even at 40 fold molar excess compared to laminarin. This lack of inhibition was not due to the experimental conditions, as the ability of the lectin domain of CR3 to bind carbohydrate ligands at 4°C has been previously shown (Thornton, B.P., et al, 1996. J. Immunol 156:1235-1246). Therefore, CR3 may not be involved in the recognition of unopsonised zymosan.

CR3 is not involved in the non-opsonic recognition of zymosan by macrophages.

To confirm the lack of involvement of CR3, the ability of macrophages from CDl lb^"7" mice, which lack functional CR3 (Melo, M.D., et al, 2000. Cell. Immunol. 205:13-23), to recognise zymosan was examined. CR3-deficient and wild type BMDM bound unopsonised zymosan equally well, and in a β-glucan dependent fashion. As before, binding of unopsonised zymosan was not inhibited by mannan or methyl glucoside.

Opsonised zymosan showed markedly enhanced binding to the wild type cells but not to CDl lb^"A cells. This is consistent with the absence of CR3 from these cells. Furthermore, laminarin was still able to inhibit the binding of opsonised particles to wild type cells partially. Therefore, β-glucan recognition may also contribute to the binding of opsonised yeast particles. This finding helps to explain previous observations reporting the lack of specificity of C3- opsonised zymosan for CR3 in macrophages (Egwang, T.G., et al, 1983. J. Immunol. Methods 61:253-257). These results, therefore, demonstrate that a macrophage β-glucan receptor, distinct from CR3, is a major receptor involved in the non-opsonic recognition of zymosan and that this receptor also contributes to the recognition of opsonised zymosan particles.

Dectin-1 is expressed on the surface of macrophages.

To explore the role of this receptor in primary macrophages, a novel Dectin-1 -specific monoclonal antibody, 2A11 (IgG2b), was generated, as described above. The specificity of this mAb was confirmed by its ability to immunoprecipitate soluble recombinant HA-tagged Dectin-1. Furthermore, 2A11 specifically stained the surface of live Dectin- 1 transduced NJH3T3 fibroblasts, indicating that it recognised an extracellular epitope (data not shown).

As the epitope recognised by the mAb 2A11 was sensitive to fixation, the expression of Dectin-1 was examined on the surface of live primary macrophages, by flow cytometry. The Dectin-1 protein was detected on the surface of freshly isolated Tg-elicited peritoneal, Biogel- elicited peritoneal, and on BMDM (data not shown). Dectin-1 was found to be expressed on the surface of all macrophage populations tested as well as on monocytes, dendritic cells and neutrophils, demonstrating that Dectin-1 is not restricted to cells of the dendritic cell lineage. Similar levels of Dectin-1 were also observed on the surface of macrophages from CR3 deficient and wild type mice. This observation is compatible with a role for Dectin-1 in the non-opsonic, β-glucan-dependent, recognition of zymosan by CDl lb-deficient cells, see below.

Dectin-1 plays a major role in the β-glucan dependent recognition of zymosan by macrophages. Having shown that non-opsonic zymosan recognition by primary macrophages was mediated by a β-glucan inhibitable receptor, which was different to CR3, the role of Dectin-1 in this process was examined. The mAb, 2A11 inhibited the binding of unopsonized zymosan to a level comparable with the inhibition obtained with the exogenous β-glucans, glucan phosphate or laminarin, indicating that this mAb binds at or near the β-glucan binding site. 2A11 may also inhibit the binding of zymosan to NIH3T3 fransductants expressing Dectin-1. The mAb, 5C6 (Rosen, H., and S. Gordon. 1987. J. Exp. Med. 166:1685-1701), which blocks the lectin activity of CR3 (Xia, Y., et al, 1999. J. Immunol. 162:2281-2290), had no effect on zymosan binding, as expected.

Similar results were obtained when macrophages were plated on dishes coated with receptor specific antibodies, a process shown to sequester receptors from the upper, ligand-interacting, cell surface (Michl, j., et al, 1979. J Exp. Med. 150:607-621) (data not shown).

The effects of these antibodies on the recognition of opsonised zymosan by macrophages was also examined (data not shown). Although the individual levels of inhibition by 2A11 and 5C6 depend on the degree of opsonisation, the simultaneous addition of both antibodies has an additive effect in inhibiting the binding of opsonised zymosan. The levels of inhibition obtained with 5C6 were, however, similar to those previously obtained with other anti-CR3 antibodies (Xia, Y., et al, 1999. J Immunol. 162:2281-2290). Furthermore the inhibition obtained with 2A11 always correlated with that obtained with exogenously added β-glucans, indicating that Dectin-1 was mediating the β-glucan dependent recognition of opsonised zymosan, as discussed above.

Thus, Dectin-1 is the major receptor for unopsonised zymosan on macrophages, and it may also contribute to the recognition of opsonised zymosan particles.

The soluble β-glucans, glucan phosphate and laminarin, shown to be the most potent β- glucan inhibitors of Dectin-1 binding to zymosan (Brown, G.D., and S. Gordon. 2001. Nature 413:36-37 and Willment, J.A., et al, 2001. J. Biol Chem. 276:43818-43823), masked the 2A11 epitope on macrophages. Therefore, Dectin-1 is also involved in the recognition of these polysaccharides. Warming the cells briefly further reduced the detection of this receptor, consistent with the observation that β-glucans are internalised after receptor binding (Muller, A., et al, 1996. J Immunol. 156:3418-3425). No change in the level of surface expressed CR3 was observed, even after warming. Taken together, these results suggest that Dectin-1 is a major receptor for both unopsonised-particulate and soluble β-glucans on macrophages.

The above results establish a predominant role of Dectin-1 in the innate recognition of β-glucans. Dectin-1, is expressed on primary macrophages and is a major receptor for unopsonised zymosan. These conclusions are reinforced in these experiments by the use of structurally-defined, pure β-glucans, receptor-specific reagents, and the care taken in controlling for the contribution of opsonic factors released from the cells under study (Ezekowitz, R.A., et al, 1985. J Clin. Invest. 76:2368-2376 and Ezekowitz, R.A., et al, 1984. J. Exp. Med. 159:244-260).

EXAMPLE 3

Expression Pattern of the β-Glucan Receptor, Dectin-1.

In Example 2, the βGR was shown to be a major receptor on M0 (macrophages) for the non-opsonic recognition of β glucans. h this Example, we establish the expression pattern of βGR/Dectin-1 by performing a comprehensive analysis of the pattern of expression of this receptor in mice.

Methods

RNA analysis

For RT-PCR analysis, total RNA from various cell lines and primary cell types was prepared using the guanidine isothiocyanate-based RNA isolation kit (Stratagene). First strand cDNA synthesis was performed using an oligo(dT) primer from the Advantage RT-for-PCR kit (Clontech), as described by the manufacturer. The βGR transcript was amplified using primers coπesponding to the region of the cDNA encoding residues 66-244 of the primary protein sequence. Dihydrofolate reductase-specific primers (Stratagene) were used as a positive control. Commercially available membranes containing polyA mRNA isolated from various mouse tissues were purchased from Origene Technologies, and were probed as described by the manufacturer using a full-length βGR cDNA probe.

Tissue and cell preparation

All mice used in this study were C57BL/6J and were between 8 and 12 weeks of age. Animals were kept and handled in accordance with institutional guidelines. Splenocytes were harvested by standard methods using a combination of injection of spleens with "Liberase Blendzyme II" in RPMI (Roche Molecular Biochemicals) and mechanical dissociation. Femurs were collected and fresh bone marrow was flushed from within using Liberase Blendzyme II and incubated for 10 minutes at 37°C to disaggregate cells. Enzymatic activity was quenched with RPMI:20%FCS, erythrocytes lysed with Gey's solution and cell debris removed by centrifugation through 100% FCS at 300 x g.

Isolation of peripheral blood leukocytes

Mice were killed and peripheral blood collected by cardiac puncture into 0.1 volume of lOOmM EDTA. Cells were harvested by centrifugation and resuspended in 50 volumes of Gey's solution for lysis of erythrocytes. Peripheral blood leukocytes were then recovered by centrifugation through FCS as described above.

Isolation of alveolar M0

Bronchoalveolar lavage was performed by repeated washes with 1ml of PBS:5mM EDTA. Resident alveolar M0, the major leukocyte population in the lungs, were identified by size and autofluorescence using flow cytometery as previously described (Maus, U., et al, 2001, Am J PhysiolLung Cell Mol Physiol 280:L5820).

Induction of sterile peritonitis and recovery of peritoneal cells

To induce sterile peritonitis, mice were injected intraperitoneally with 4% thioglycollate (BD) up to 4 days prior to peritoneal lavage. After humane killing, inflammatory cells were collected by peritoneal lavage with ice cold 5 mM EDTA in PBS. Resident peritoneal cells were collected in the same way from untreated animals. Peritoneal M0 were identified by their expression of F4/80 and CR3 and distinguished from eosinophils by FSC/SSC profiles. To confirm the cellular composition of peritoneal exudates, differential counts were performed on cytospin preparations stained with Hema Gun (BDH).

FACS analysis

FACS was performed according to conventional protocols at 4°C in the presence of 2 mM NaN₃. Cells were blocked with 5% heat-inactivated rabbit serum; 0.5% BSA; 5 mM EDTA and 4 μg/ml 2.4G2 (anti-FcγRII and III) prior to the addition of primary antibodies. Biotinylated antibodies were detected using streptavidin-allophycocyanin (BD Pharmingen). Cells were fixed with 1% formaldehyde in PBS prior to analysis.

The following antibodies were used in this study: B220-CyChrome (RA3-6B2; BD Pharmingen), CD3-CyChrome (17A2; BD Pharmingen), F4/80-PE (Serotec), CDllc-PE (HL3; BD Pharmingen), Gr-l-PE (anti-Ly6C/G; BD Pharmingen), CD49b-PE (DX5 - "Pan NK-cell"; BD Pharmingen), 5C6-FITC (anti-CR3/CDllb) (21), 2All-biotin (rat IgG2b anti-βGR; Brown et al., manuscript in submission), 5D3-biotin (rat IgG2a anti-M0 mannose receptor; Martinez-Pomares, L. et al. unpublished data) and irrelevant rat IgG2b-biotin, IgG2a-biotin and IgG2b-FITC control antibodies.

In vitro non-opsonic zymosan binding assay

In vitro zymosan binding assays were performed as previously described (Brown, G. D., et al, 2001. Nature 413:3614, and Wilhnent, J. A., S. Gordon, and G. D. Brown. 2001, JBiol Chem 20:20). In brief, resident or 4 day thioglycollate-elicited peritoneal M0 were recovered, as described above, and plated at 5 x 10⁵ and 2.5 x 10⁵ cells/well respectively in 24 well plates in RPMI: 10% FCS overnight. The following day the cells were cooled to 4°C and washed three times with pre-chilled medium. All experiments were performed at 4°C to prevent receptor internalisation, to provide a direct measure of surface receptor involvement, and to prevent local release of opsonins including complement (Ezekowitz, R. A., et al, 1984, JExp Med 159:244 and Ezekowitz, R. A., et al, 1985, J Clin Invest 76:236822). Zymosan-FITC (Molecular Probes) was added to the M0 at a ratio of 25 particles/cell for 1 hour on ice. For in vitro blocking assays, carbohydrates (laminarin, β-methyl glucoside and mannan; all from Sigma and used at 100 μg/ml) or antibodies (2A11; (Brown et al, manuscript in submission); 5C6 (Rosen, H., et al, 1987, JExp Med 166:1685), which has been shown to block the CR3-mediated lectin activity (Xia, Y., et al, 1999, J Immunol 162:228124); or an irrelevant rat IgG2b confrol; all used at 100 μg/ml) were added to the chilled cells 30 minutes prior to the addition of zymosan. After incubation, unbound zymosan was removed by extensive washing with medium and cells were lysed with 3% Triton X-100. FITC in lysates was quantified using a Tifretek Fluoroskan II (Labsystems Group (UK) Ltd) as previously described (Brown, G. D., et al, 2001. Nature 413:3614, and Willment, J. A., S. Gordon, and G. D. Brown. 2001, J Biol Chem 20:20). For Ab modulation experiments poly-D- lysine conditioned tissue culture plates were coated with antibody at 100 μg/ml as previously described (Michl, J., et al, 1979, JExp Med 150:607.25).

Statistical Analysis

Statistics were calculated using GraphPad Prism™ (version 2.0; GraphPad Software). Oneway analysis of variance (ANOVA) with Bonferroni multiple comparison test was applied throughout. * = PO.05; ** = PO.01 and *** = PO.001.

Results

Expression of βGR transcript in macrophages and multiple mouse tissues.

All M0 cell lines lines (RAW264.7, J774 and P388D1), as well as bone marrow-derived M0 (BMDM0) and DC (BMDDC), showed evidence of βGR expression whereas the βGR transcript was not detectable in the mouse fibroblast cell line NIH3T3 (data not shown). The housekeeping gene dihydrofolate reductase (DHFR) was used as a PCR confrol.

The full length coding sequence was used to screen a multiple tissue Northern blot, showing that βGR expression was evident in most murine tissues (heart, kidney, liver, lung, small intestine, spleen, stomach, testis and thymus) with the exception of brain, muscle and skin (data not shown). Notably, there was only one discernible transcript detectable in these tissues. Control probing with β-actin confirmed equivalent loading between lanes (data not shown). Distribution of βGR surface expression in the spleen.

The surface expression of the βGR was examined using 2A11 on freshly isolated splenocytes. CDl lc^hlghDC were found to express the βGR, in a similar pattern to that reported previously (Ariizumi, K, et al, 2000, JBiol Chem 275:2015715). Notably, however, other CDl lc^Iow/"cells in the spleen, particularly those expressing complement receptor 3 (CR3), exhibited high surface expression of βGR. To further delineate which cell types were expressing βGR, the cells were subdivided into six populations based on their expression of CR3 and Gr-1 (a mAb recognising Ly-6G and Ly-6C), and their FSC/SSC profiles (Figure 2B). Gr-l^highCR3^highSSC^highneutrophils (population 3) exhibited high surface expression of βGR as did Gr-l^lowCR3⁺SSC^,owM0 (population 4B), which also expressed F4/80. CR3⁺Gr-1 ^"splenocytes, a mixed population containing DC (CDl lc^high), NK cells and other M0 (both CDl lc^int), showed heterogeneity in expression of βGR. NK cells, thought to express the βGR (8), and identified by high expression of the DX5 antigen (CD49b; (Arase, H., et al, 2001, J Immunol 167:114126)), did not show significant labelling with the 2A11 antibody. Gr-l^lowCR3^"splenocytes, previously reported to be a T cell subset (Lagasse, E., et al, 1996, J Immunol Methods 197:13921), and expressing CD3, were found to express low levels of surface βGR (data not shown). Analysis of all splenic T cells (CD3⁺) and B cells (B220⁺) for βGR surface expression, however, indicated that only a distinct subset of T cells exhibited significant surface expression of βGR. βGR⁺CD3⁺T cells were predominantly Gr-1⁺ and CD8⁺ but CD4⁺ cells were also observed. Splenic autofluorescent F4/80⁺M0 also expressed βGR, albeit at very low levels. Plasmacytoid DC, which were identified by their Gr-1⁺B220⁺CD1 lc^ιntCR3^" phenotype and analysed in 129Sv/Ev and Balb/c mice because of the relative sparsity of these cells in C57BL/6 (28), also exhibited low but detectable levels of βGR expression (data not shown).

Expression of βGR on peripheral blood leukocytes.

Expression of βGR/Dectin-1 has been observed on both human and mouse peripheral blood leukocytes (PBL) by Northern blot analysis (Yokota, K, A. et al, 2001, Gene 272:51 and Willment, J. A., et al, 2001, JBiol Chem 20:20). These observations were confirmed using FACS by identifying a significant population of PBL (approximately 10%) that expressed βGR, as both cell types showed significant levels of surface staining with 2A11, compared to a rat IgG2b control. Consistent with the data obtained from the spleen, peripheral blood neutrophils (Gr-l^hlghSSC^hlgh) and peripheral blood M₀ (CR3⁺F4/80⁺SSC^lo ) exhibited high surface expression of βGR (Figure 3A).

Surface expression of the βGR on myeloid cells in the bone marrow.

The βGR transcript has been previously observed in human bone marrow (Willment, J. A., et al, 2001, JBiol Chem 20:20). Murine bone marrow was examined for the expression of the βGR (data not shown). Gr-l^hlghCR3⁺ neutrophils appeared to be subdivided into two populations. Approximately one-third of the bone marrow Gr-l^hlghneutrophils had high βGR surface expression, the remaining two-thirds showed a lower level of expression. The βGR^hlghneutrophils had higher SSC and higher CR3 surface expression than the βGR^lowneufrophils, suggesting that the βGR^{hl h}neutrophils are in a more advanced state of maturation. This is consistent with the high βGR surface expression detected on circulating peripheral blood neutrophils. The Gr-l^low subgroup of bone marrow cells that has been reported to include cells of the M₀/M0-lineage, myeloid precursors and hematopoietic stem cells (Biermann, H., B. et al, 1999, JLeukoc Biol 65:217), contained cells with the highest βGR surface expression. Expression of CR3 and F4/80 indicated that these high βGR expressing cells belonged to the M₀/M0-lineage.

Expression of βGR by isolated primary M0.

Freshly isolated resident and thioglycollate-elicited peritoneal M0 were assayed for surface expression of βGR. βGR expression on alveolar M0 was also examined, following the discovery of high levels of the transcript in the lung. Both freshly isolated alveolar M0 and thioglycollate- elicited M0 expressed high surface levels of βGR, whereas resident peritoneal M0 exhibited lower expression (data not shown). Interestingly, we observed an upregulation of βGR on the surface of resident peritoneal M0 after 1 day of culture. Surface expression of βGR on thioglycollate-elicited M0 was relatively unaffected by 1 day of culture. Since the M0 MR is also a candidate receptor for the non-opsonic recognition of zymosan by resident peritoneal M0 (see below), the surface expression of this receptor was analysed on the same cells. As with the βGR, higher surface expression of the MR was found on thioglycollate-elicited and moderate expression was found on resident alveolar M0, but only very limited expression was found on the surface of resident peritoneal cells. Unlike the βGR, however, expression of the MR on resident peritoneal cells was relatively unaffected by 1 day of culture. Alveolar M0 expressed negligible CR3 and low levels of F4/80.

Surface expression of βGR during peritoneal inflammation.

The expression of the βGR, in an inflammatory context, was examined by looking at peritoneal exudate cells, 18 hours after the infraperitoneal administration of thioglycoUate, a model of sterile peritonitis. F4/80⁺CR3^{hi h}Gr-l^" M0, F4/80^"CR3⁺Gr-1⁺ neutrophils and F4/80⁺CR3⁺Gr-l^" SSC^high eosinophils (McGaπy, M. P., et al, 1991, JLeukoc Biol 50:471) were then tested for βGR surface expression (data not shown). Elicited peritoneal M0 exhibited the highest βGR expression, and significant amounts were also present on the inflammatory neutrophils. In contrast, recruited eosinophils showed no obvious surface expression of βGR.

Non-opsonic binding of zymosan to resident peritoneal M0.

Freshly isolated resident peritoneal M0 were observed to have a lower level of surface βGR expression compared to that on other M0 studied. Example 2 had shown that βGR was a major receptor for zymosan on thioglycollate-elicited and BMDM0. Therefore, resident peritoneal M0 were investigated to see if this applied to them. We compared the contribution of the βGR on both resident and elicited M0. The binding of unopsonised zymosan to elicited M0 was significantly inhibited by β-glucans (see Example 2). βGR was still a major receptor for zymosan on resident M0, although it contributed less to this process than in the thioglycollate-elicited cells. Furthermore, mannan had an inhibitory effect on the binding of zymosan to the resident M0, but not the thioglycoUate elicited cells. Combination of β-glucans and mannan did not have an additive effect. As with the elicited M0, methyl glucoside also failed to inhibit the initial binding of zymosan to resident peritoneal M0 suggesting no involvement of CR3 in this process (data not shown). These results implied that a secondary β-glucan independent, mannan-inhibited non- opsonic binding mechanism was operational on resident M0.

Antibody blocking experiments were performed on the resident peritoneal M0 to determine which specific receptors were involved. The anti-βGR mAb, 2A11, blocked the non-opsonic binding of zymosan to resident peritoneal M0 to the same degree as the soluble β-glucans laminarin and glucan phosphate, consistent with it being a major β-glucan receptor on M0. Anti- CR3 (5C6, which blocks the lectin activity of CR3) had no inhibitory effect, consistent with the results obtained with other primary M0 in Example 2. The surface expression of the M0 MR was found to be low on resident peritoneal M0 and higher on thioglycollate-elicited M0, which do not have a mannan inhibited component of zymosan binding, suggesting that the MR was not involved in the non-opsonic recognition of zymosan by peritoneal M0.

However, to exclude a role for the M0 MR in the binding of unopsonised zymosan in this experimental system, antibody blocking experiments were perfonned using a specific mAb to block surface receptor M0 MR. The mAb against the M0 MR (5D3) failed to block the non-opsonic binding of zymosan to resident peritoneal M0 (data not shown), however, this mAb has not yet been demonstrated to posses a blocking activity. We also performed antibody modulation experiments using specific mAb to deplete surface receptors from the upper ligand binding surface. Only 2A11, anti-βGR, coated tissue culture wells inhibited the non-opsonic binding of zymosan whereas anti-M0 MR and anti-CR3, did not.

EXAMPLE 4

In the data shown below in Figure 4, it is shown that, with increasing exposure of Dectin-1 in RAW macrophages (denoted as "bgr"), enhanced levels of zymosan binding are observed, compared to normal cells ("wt'Vwild type), thereby confirming that Dectin-1 binds zymosan (Bgr = cells which overexpress Dectin-1).

TNF production: over-expression of Dectin-1, also causes an increase in TNF production. Therefore, Dectin-1 mediates the biological effects of beta-glucans. Figure 5 shows that TNF production is increased in cells which over-express Dectin-1 compared to normal cells with wild type (wt) expression levels of Dectin-1. hi particular, it is shown that TNF production is increased as levels of available zymosan are increased.

Fig 4 shows that zymosan binding is increased in cells (denoted "bgr") which over-express Dectin-1. When these "bgr" cells are exposed to zymosan, they produce more TNF cytokine than wt cells with normal levels of Dectin-1, for the same level of zymosan. addition, as zymosan levels are increased, TNF production levels are also increased. TNF production is linked to levels of Dectin-1 expression and to the levels of glucan available for binding to the receptor. Thus Dectin-1 mediates the biological activities of beta- glucans.

Thus, it can be seen that Dectin-1 mediates the cellular effects of β glucan activity, such as modulation of TNF production.

Claims

Claims:

1. A substance capable of direct interaction with Dectin- 1 , and which inhibits the binding of zymosan to Dectin- 1.

2. A substance according to claim 1, said substance not inhibiting the binding of murine T cells to cells expressing murine Dectin-1.

3. A substance according to claim 1 or claim 2, which encourages multimerisation of membrane-bound Dectin-1.

4. A substance according to any preceding claim, which has a protein or saccharide structure, the protein not being derived from S. cerevisiae, and the saccharide being an octasaccharide or greater.

5. A substance according to any preceding claim, which is an antibody, or a fragment thereof.

6. A substance according to claim 5 which is humanised.

7. A substance according to claim 5 or 6, specific for the C-type lectin fold of Dectin 1.

8. A substance according to any preceding claim, which is labelled.

9. A substance according to any of claims 1 to 4, which comprises an oligosaccharide chain comprising in excess of 7 hexose sub-units, each sub-unit linked to the adjacent sub-unit by a β-1, 3 or β-1, 6 linkage.

10. A substance according to claim 9, which has a molecular weight greater than that of laminarin.

11. A substance according to claim 9, which has a molecular weight equal to or greater than that of glucan phosphate.

12. A substance according to any preceding claim, comprising a polypeptide substituted at one or more glycosylation sites thereon by one or more oligosaccharide chains comprising in excess of 7 hexose sub-units, each sub-unit linked to the adjacent sub-unit by a β-1, 3 or β-1, 6 linkage.

13. A substance according to any of claims 1 to 12 capable of activating macrophages via the macrophage β glucan receptor.

14. A substance according to any of claims 1 to 12 capable of partially activating macrophages via the macrophage β glucan receptor, such that full activation may be effected by a secondary insult.

15. Use of a substance according to any of claims 1 to 14 to detect the presence of Dectin-1.

16. Use of a substance according to any of claims 1 to 14 to detect cells expressing Dectin-1.

17. Use of a substance according to any of claims 1 to 14 to block the β glucan binding site.

18. Use of a substance according to any of claims 1 to 14 to isolate Dectin-1.

19. Use of a substance according to any of claims 1 to 14 to activate the β-glucan receptor.

20. Use of a substance according to any of claims 1 to 14 in an anti-cancer preparation.

21. Use of a substance according to any of claims 1 to 14 as an therapeutic antioxidant.

22. Use of a substance according to any of claims 1 to 14 as a haematopoietic stimulant.

23. Use of a substance according to any of claims 1 to 14 as an anti-ageing agent.

24. Use of a substance according to any of claims 1 to 14 as a prophylactic against diseases in post operative patients and/or those with depressed immune systems.

25. Use of a substance according to any of claims 1 to 14 for enhanced wound healing.

26. Use of a substance according to any of claims 1 to 14 for the treatment of a fungal infection.

27. Use of a substance according to any of claims 1 to 14 for the freatment of Candida related infections.

28. Use of a substance according to any of claims 1 to 14 for the treatment of transplantation rejection.

29. Use of a substance according to any of claims 1 to 14 for protection against radiation induced damage.

30. Use of a substance according to any of claims 1 to 14, which is capable of activating the β-glucan receptor, for the freatment or prophylaxis of an infection or condition selected from the group consisting of viral, bacterial, protozoal, parasitic and cancerous cells.

31. Use of a substance according to any of claims 1 to 14, which is capable of blocking the β-glucan receptor, for the freatment or prophylaxis of asthma, hay fever and allergic dermatitis.

32. Use of a substance according to any of claims 1 to 14 to provide long term protection against a microbial infection.

33. Use according to claim 32, wherein the infection is selected from E. coli and Staphylococcus aureus.

34. Use of a substance according to any of claims 1 to 14 in combination with an antibiotic, in therapy.

35. A pharmaceutical preparation comprising a substance according to any of claims 1 to 14.

36. A method of treatment of a condition described herein in a human or animal in need thereof, comprising administration of with a substance according to any of claims 1 to 14.