EP1697410A1

EP1697410A1 - Isolated cytotoxic factor associated with multiple sclerosis and method of detecting said cytotoxic factor

Info

Publication number: EP1697410A1
Application number: EP04816595A
Authority: EP
Inventors: Hervé Perron; Anne Eveno-Nobile; Jacques Portoukalian; Nicole Battail-Poirot
Original assignee: Biomerieux SA; Institut National de la Sante et de la Recherche Medicale INSERM
Current assignee: Biomerieux SA; Institut National de la Sante et de la Recherche Medicale INSERM
Priority date: 2003-12-23
Filing date: 2004-12-22
Publication date: 2006-09-06
Also published as: US7700294B2; JP2008505053A; FR2864086A1; US20070065881A1; WO2005063810A1

Abstract

The invention relates to an isolated cytotoxic factor which is associated with multiple sclerosis and which is selected from heterocomplex GM2AP/GM2/MRP14 and mutated GM2AP/GM2/MRP14, and to the method of detecting said factor in a biological sample to be tested. The inventive method comprises the following steps consisting in: (i) bringing the biological sample into contact with (a) at least one capture antibody selected from antibodies that bind specifically to protein GM2AP, to the mutated GM2AP protein, to protein MRP14, to complex GM2AP/GM2, to the mutated GM2AP/GM2 complex and to complex MRP14/GM2, and (b) at least one labelled detection antibody selected from antibodies that bind specifically to protein GM2AP, to the mutated GM2AP protein, to protein MRP14, to complex GM2AP/GM2, to the mutated GM2AP/GM2 complex and to complex MRP14/GM2; and (ii) detecting and/or quantifying the cytotoxic factor by detecting and/or quantifying the labelled detection antibody.

Description

ISOLATED CYTOTOXIC FACTOR ASSOCIATED WITH MULTIPLE SCLEROSIS AND METHOD FOR DETECTION OF T CYTOTOXIC FACTOR

Multiple sclerosis is a chronic disease of the central nervous system of man, evolving by succession of remission and pushing phases or according to a regular progression, whose anatomopathological characteristic consists in the formation of well delimited zones of de yelination in the white matter of the brain and spinal cord. Histologically, these areas present at the early stage of the lesion process, a breakdown of the peri-axonal myelin associated with an attack of the glial cells responsible for this demyelination. Macrophagic inflammatory activation involving microglial cells (tissue macrophages residing in the central nervous system), as well as probably macrophages from infiltrated blood monocytes, is associated with this demyelination process and contributes to the destruction of myelinated sheets. In the center of the demyelinated zone, a relative depletion in glial cells is found while a proliferation of astrocytes develops at the periphery and can invade the demyelinated plaque to generate a fibrous or gliotic plaque. These sclerotic structures are at the origin of the name given to the disease. Another characteristic of these plaques is their almost systematic association with a vascular element around which they develop. Histologically, there is a frequent alteration of the blood-brain barrier (BBB) constituted by the capillary endothelium. One of the determining elements in the maintenance of the BBB is constituted by the underlying presence of cytoplasmic extensions of astrocytes, called astrocytic feet. Presumably, the astrocytic feet induce the formation or allow the maintenance of tight junction structures which ensure the cohesion of the capillary endothelial barrier embodying the BBB. However, various pathological models report the alteration of the BBB and a depletion of the astrocytic feet. In addition, in the lesional process of MS, the alteration of the BBB contributes to amplifying the associated inflammatory response, by the influx of lymphoid cells from the bloodstream. The contribution of inflammation associated with immune cells is important in MS and participates in the lesional process. The etiology of MS is the source of a current debate because the disease could have various origins. Hypotheses have been put forward on a bacterial and / or viral origin. Furthermore, as described in patent application WO 95/21859, H. Perron et al. have been led to search for one or more effector agents of the pathogenic process leading to the typical formation of demyelination plaques and to an astrocytic gliosis. In the context of this study, they demonstrated the presence in cerebrospinal fluid (CSF) and serum of MS patients of at least one factor which exhibits toxic activity with respect to human or oligodendrocytic cells or animal. This toxic activity is characterized by cytomorphological disorganization of the network of intermediate filaments and / or degradation of the proteins of said filaments and / or cell death by apoptosis of glial cells. They established a significant correlation between the in vitro detection of this toxic activity in CSF and serum samples from MS patients and multiple sclerosis by a quantitative colorimetric assay with bromide. methyltetrazolium (MTT) of living cells, as described in patent application WO 95/21859. Furthermore, C. Malcus-Vocanson et al. ^{1, Λ} have shown that urine is a very favorable biological fluid for the detection of the activity of this toxic factor and developed a method using flow cytometry to detect and / or quantify adherent glial cells which have died by apoptosis. All the information concerning this process is described in patent application WO 98/11439. Tests have been carried out using a protein fraction of CSF and urine from MS patients in an attempt to identify this toxic factor. The protein content of each fraction was separated on a 12% SDS-PAGE gel and observed after coloration of the gel with silver. Among the proteins observed, a protein fraction centered on an apparent molecular weight of approximately 21 D was found in the majority associated with the toxic activity detected in vitro and a fraction centered on an apparent molecular weight of approximately 17 kD was found mainly associated with this toxic activity. An injection of the LCR fraction from MS patients into the Lewis rat brain and a post-mortem histological observation of rat brain sections made it possible to observe, three months after the injection, apoptosis of the astrocytic population and the formation of demyelination plaques. All the information is contained in patent application WO 97/33466. These observations are consistent with those that have been made on brain sections of MS patients after biopsy ⁵ . Proteins potentially associated with this toxic activity with respect to glial cells in biological samples from MS patients were studied as described in patent application WO 01/05422. The GM2AP proteins (precursor of the ganglioside activator GM2) and saposin B were thus measured in the urine of MS and non MS patients. The results presented in patent application WO 01/05422 showed that GM2AP and saposin B were present at high concentrations in the urine of MS patients compared to the concentrations found in non-MS individuals and that these two proteins which are co -detected in the urine of MS patients could represent a marker of the pathology. The inventors had also established a correlation between the detection of the proteins G 2AP and saposin B in the urine and the gliotoxicity measured in these urines by the MTT test and showed that there was a correlation between high urinary concentration and gliotoxicity for these two proteins. The inventors concluded that the GM2AP and / or saposin B proteins were involved in the mechanism of gliotoxicitis and that they could probably act in combination to induce gliotoxicitis. The present inventors have now wanted to know the activity of the proteins identified in patent application WO 01/05422 by using the MTT test and to see whether the gliotoxicity discovered in the urine of patients suffering from multiple sclerosis was linked to the proteins identified. Unexpectedly, the present inventors have shown that it was neither the proteins identified in WO 01/05422 taken individually, nor the association of the proteins GM2AP / saposin B which were involved in gliotoxicity and that in a completely surprising manner the agent responsible for gliotoxic activity and involved in cytotoxicity corresponded to a heterocomplex GM2AP / GM2 / MRP14 (calgranulin B) or mutated GM2AP / GM2 / MRP14, as described in the examples which follow. GM2 or ganglioside GM2 is a complex lipid found in brain tissue. Also, the subject of the present invention is the isolated, purified cytotoxic factor associated with multiple sclerosis, said cytotoxic factor being the heterocomplex GM2AP / GM2 / MRP1 or mutated GM2AP / GM2 / MRP14, it being understood that mutated GM2AP corresponds to the sequence SEQ ID NO: 2. These isolated, purified heteroco plexes are useful as markers of MS pathology and more precisely of a form of the disease, of a stage of the disease, of a period of activity of the disease, as well as in the follow-up of patients treated for this pathology. The present inventors then developed a method, a composition and a reaction mixture for detecting and / or quantifying the GM2AP / GM2 / MRP1 and mutated GM2AP / GM2 / MRP heterocomplexes in samples of individuals susceptible to sclerosis. in plaques or with clinical signs of this pathology. The method consists in detecting and / or quantifying the cytotoxic factor, associated with multiple sclerosis, in a biological sample, by isolating from said biological sample the heterocomplex GM2AP / GM2 / MRP1 or mutated GM2AP / G 2 / MRPl4. By isolation of the heterocomplex is meant all the conditions which allow the specific detection of the heterocomplex. The isolation of said heterocomplex can be carried out by any appropriate means. There may be mentioned ^'way of example, non-denaturing electrophoresis, the column chromatography, methods for the degradation of the biological medium compounds, with the exception of the heterocomplex (such as for example a treatment with proteinase) as well as any other method making it possible to detect a physicochemical characteristic of said heterocomplex, such as than molecular weight, isoelectric point or any other suitable means. In an embodiment of the invention, at least one antibody or at least two antibodies which bind (s) specifically to the heterocomplex are used, and said cytotoxic factor is detected and / or quantified by demonstrating the formation of d 'a complex constituted by the heterocomplex and the antibody or by the demonstration of a complex constituted by the heterocomplex and the two antibodies. Preferably at least one of said antibodies is a capture antibody and at least the other of the antibodies is a detection antibody. The capture antibody is chosen from the antibodies which specifically bind to the GM2AP / GM2 complex, to the mutated GM2AP / GM2 complex, to the MRP14 / GM2 complex, to the GM2AP / MRP14 complex and to the mutated GM2AP / MRPl4 complex, and the antibody detection is chosen from the antibodies which specifically bind to the GM2AP / GM2 complex, to the mutated GM2AP / GM2 complex, to the MRP14 / GM2 complex, to the GM2AP / MRP14 complex and to the mutated GM2AP / MRP14 complex. In another embodiment, the heterocomplex is isolated using at least two antibodies, at least one of which binds specifically to GM2AP or GM2AP mutated from the heterocomplex and at least the other binds specifically to MRP14 of the heterocomplex, and said cytotoxic factor is detected and / or quantified by demonstrating the formation of a complex constituted by the heterocomplex and the two antibodies. Preferably, at least one of said abovementioned antibodies is a capture antibody and at least the other of said antibodies is a detection antibody. In the abovementioned embodiments, the demonstration of the formation of the complex consisting of the heterocomplex and at least one antibody or by the heterocomplex and at least two antibodies can be carried out by any appropriate means, for example by screening according to size using a sorting apparatus, by screening according to molecular weight using a separation column or by direct or indirect labeling at least one antibody or by any other suitable means. In one embodiment, the method consists in (i) (i) having a biological sample to be tested, (ii) bringing said biological sample into contact with at least one capture antibody, said capture antibody being chosen from the antibodies which specifically bind to GM2AP protein, mutated GM2AP protein, MRP14 protein, GM2AP / GM2 complex, mutated GM2AP / GM2 complex and MRP14 / GM2 complex; and with at least one labeled detection antibody, said detection antibody being chosen from antibodies which specifically bind to the GM2AP protein, to the mutated GM2AP protein, to the MRP14 protein, to the GM2AP / GM2 complex, to the mutated GM2AP / GM2 and the MRP14 / GM2 complex, and (iii) the cytotoxic factor is detected and / or quantified by detection and / or quantification of the labeled detection antibody, it being understood that the mutated GM2AP corresponds to the sequence SEQ ID NO: 2. Preferably, the detection and / or quantification of the cytotoxic factor is carried out using different immunoassay principles which are well known to those skilled in the art, such as ELISA and ELFA, and advantageously a sandwich type immunoassay is used. The sandwich immunoassay can be carried out in one or more stages, ie without a washing stage or with one or more washing stages. The detection antibody or antibodies are labeled with any suitable label. The labeling can thus be radioactive labeling, labeling with an enzyme, labeling with a fluorescent molecule, labeling with a vitamin, a colorimetric marking. In the present invention, the marker is preferably a vitamin, biotin, the detection is carried out by the addition of streptavidin coupled to horseradish peroxidase and the development is carried out by the addition of orthophenylenediamine dihydrochloride. The antibody or capture antibodies are immobilized directly or indirectly on a solid phase. The term "antibody" used in the present invention includes monoclonal and polyclonal antibodies, their fragments and their derivatives. By antibody fragment is meant the fragments F (ab) 2, Fab, Fab ', sFv of a native antibody ⁶ ' ⁷ and by derivative is understood, inter alia, a chimeric derivative of a native antibody ⁸ ' ⁹ . These antibody fragments and antibody derivatives retain the ability to selectively bind to the target antigen. It may be advantageous to use humanized antibodies. The "humanized" forms of non-human, for example murine, antibodies are chimeric antibodies which comprise a minimal sequence derived from a non-human immunoglobulin. For the most part, humanized antibodies are human immunoglobulins (receptor antibodies) in which residues of a hypervariable region of the receptor are replaced by residues of a hypervariable region of a donor species (donor antibody), such as mouse, rat, rabbit or non-human primate, having the specificity, affinity and capacity desired. In some cases, the residues (FR) of the Fv region of human immunoglobulin are replaced by corresponding non-human residues. In addition, humanized antibodies may include residues which are not found in the recipient antibody or in the donor antibody. These modifications are made to improve the performance of the antibody. In general, the humanized antibody will comprise at least and preferably two variable domains, in which all or almost all of the hypervariable loops correspond to a non-human immunoglobulin and all or almost all of the FR regions will be those of a human immunoglobulin. The optionally humanized antibodies may also comprise at least part of a constant region (Fc) of an immunoglobulin, such as a human immunoglobulin ¹⁰ ' ¹¹ ' ¹² . Mention may in particular be made of the anti-GM2AP and anti-MRP14 antibodies described in application WO 01/05422. But the surprising discovery that the agent responsible for gliotoxic activity and involved in cytotoxicity corresponds to the heterocomplex GM2AP / GM2 / MRP14 or mutated GM2AP / GM2 / MRP14 allows the production of anti-heterocomplex antibodies which are capable of specifically link to the GM2AP / GM2 complex, to the mutated GM2AP / GM2 complex, to the MRP14 / GM2 complex, to the GM2AP / MRP14 complex or to the GM2AP / mutated GM2AP complex. The production of such antibodies is well known to those skilled in the art. The GM2AP / GM2 / MRP14 or GM2AP / GM2 / MRP14 hererocomplex is used as an immunogen to immunize BALB / c mice by intraperitoneal injection. The first injection is made with complete Freund's adjuvant. The other injections are given 4-8 weeks apart with incomplete Freund's adjuvant. A final reminder is made a few days before the fusion in physiological water. After this booster, the spleens of the immunized mice are removed and the splenocytes are collected. Then, the fusion of the spleen cells with cells of a myeloma line is carried out and the cells secreting antibodies are selected which recognize in ELISA the heterocomplex used for immunization. We finally select the clones producing specific antibodies of the immune heterocomplex, that is to say which do not recognize neither GM2AP, or mutated GM2AP, nor MRP14, alone. The aforementioned antibodies can be used in the method of detection and / or quantification of the cytotoxic factor either alone or in combination. In a preferred embodiment of the method of the invention, the sample to be tested is subjected to a preliminary treatment comprising: a step of digestion of the proteins of the sample with proteinase K; a step of inactivation of proteinase K, for example by precipitation with trichloroacetic acid, and a step of neutralization of the pH, for example by addition of a tris-random buffer. The biological sample to be tested is serum, plasma, urine or cerebrospinal fluid, preferably urine. Preferably, the antibodies used in the process of the invention are the following monoclonal and polyclonal antibodies: 10E11A11, 13D1E5, 13H9C9, 19C11C10, 2G12H5, 79, 2B9H2, 4A7B10, 5H7C10 and 196. But it is quite obvious that any antibody which has the characteristic of specifically binding to the GM2AP protein, to the mutated GM2AP protein, to the MRP14 protein, to the GM2AP / GM2 complex, to the mutated GM2AP / GM2 complex, to the MRP14 / GM2 complex, to the GM2AP / RP14 complex or to the complex Mutated GM2AP / MRP1 is part of the invention, the methods for obtaining such antibodies being well known to those skilled in the art, as described above. Preferably, the antibodies used in the ELISA sandwich detection and / or quantification test of the invention are the following monoclonal and polyclonal antibodies: capture antibodies 10E11A11, 13D1E5, 2G12H5, 4A7B10, 5H7C10, 2B9H2, and 79 detection antibody 10E11A11, 4A7B10, 5H7C10, 2B9H2 13H9C9, 19C11C10, 13D1E5 and 2G12H5.

The capture and detection antibodies are advantageously chosen from among the pairs: 2B9H2 / 10E11A11, 10E11A11 / 4A7B10 + 5H7C10, 13D1E5 + 2G12H5 / 4A7B10 + 5H7C10, 79 / 4A7B10 + 5H7C10, 79 / 2B9H2, 4A7B10 / 10H10 + 101071010 5H7C10 / 13H9C9 + 19C11C10, 2B9H2 / 10E11A11, 2B9H2 / 13H9C9 + 19C11C10, 13D1E5 + 2G12H5 / 4A7B10 + 5H7C10, 79 / 2B9H2, 4A7B10 + 5H7C10 / 5E7A10 / 5E7A10 / 5E7A10 13H9C9 + 19C11C10.

The aforementioned monoclonal and polyclonal antibodies are new and also form part of the objects of the present invention. Their mode of production will be described in more detail in the experimental part. The selected and preferred pairs of capture and detection antibodies are also new and also form part of the objects of the present invention. The present invention also relates to a composition for the detection and / or the quantification of the abovementioned cytotoxic (gliotoxic) factor in a biological sample to be tested, said composition comprising at least one antibody which specifically binds to the GM2AP / GM2 / MRP14 or GM2AP mutated / GM2 / MRP1 heterocomplex. Preferably, said composition comprises at least two antibodies which specifically bind to the heterocomplex. The subject of the present invention is also a composition for the detection and / or quantification of the abovementioned cytotoxic (gliotoxic) factor in a biological sample to be tested, said composition comprising in a reaction mixture and simultaneously at least one capture antibody and at least one labeled detection antibody, said antibodies being chosen from antibodies which bind specifically to the GM2AP protein, to the mutated GM2AP protein and to the heterocomplex protein MRP14. Preferably, the capture and detection antibodies are chosen from the following monoclonal and polyclonal antibodies: 10E11A11, 13D1E5, 13H9C9, 19C11C10, 2G12H5, 79, 2B9H2, 4A7B10, 5H7C10 and 196. Advantageously, said composition comprises at least one antibody of capture selected from antibodies 10E11A11, 13D1E5, 2G12H5, 4A7B10, 5H7C10, 2B9H2, and 79; and at least one detection antibody chosen from detection antibodies 10E11A11, 4A7B10, 5H7C10, 2B9H2 13H9C9, 19C11C10, 13D1E5 and 2G12H5. The preferred compositions include the following pairs of capture and detection antibodies: 2B9H2 / 10E11A11, 10E11A11 / 4A7B10 + 5H7C10, 13D1E5 + 2G12H5 / 4A7B10 + 5H7C10, 79 / 4A7B10 + 5H7C10, 79 / 2B9H2, 4A7B10 + 4A7B10 + 5H7C10 / 13H9C9 + 19C11C10, 2B9H2 / 10E11A11, 2B9H2 / 13H9C9 + 19C11C10, 13D1E5 + 2G12H5 / 4A7B10 + 5H7C10, 79 / 2B9H2, 4A7B10 + 5H7C10 / 10E11A11, 4A7B10 + 5H7C10 / 13D1E5 + 22G12H5, 2B9H2 / 13D1E5 + 22G12H5, 2B9H2.

The subject of the invention is also a reaction mixture for the detection and / or quantification of the abovementioned cytotoxic (gliotoxic) factor, said mixture comprising at least two antibodies, at least one of which binds specifically to GM2AP or GM2AP mutated from heterocomplex and at least the other specifically binds to MRP14 of the heterocomplex. By reaction mixture is meant a homogeneous or heterogeneous medium which simultaneously comprises at least the two abovementioned antibodies. Preferably, at least one of said antibodies is a capture antibody and at least the other of said antibodies is a detection antibody. Another object of the invention is a complex comprising the heterocomplex linked to at least two antibodies, of which at least one of the antibodies is specific for GM2AP or for mutated GM2AP and at least the other antibody is specific for RP14. The sequence SEQ ID NO: 1 corresponds to the sequence of the GM2AP protein. The sequence SEQ ID NO: 2 corresponds to the sequence of the GM2AP protein mutated in exon 2, at position 40 (replacement of an aspartic acid by a phenylalanine. The sequence SEQ ID NO: 3 corresponds to the sequence of the mutated GM2AP protein, having mutations in both exon 1, exon 2 and exon 4. In the following detailed description, when reference is made to the GM2AP protein, the sequence to take into account is the sequence identified in the sequence identifier in SEQ ID NO: 1. Furthermore, when reference is made to the mutated GM2AP protein, the sequence to be taken into account is the sequence identified in the sequence identifier in SEQ ID NO: 2; it being understood that in the sequences SEQ ID NO: 1 and SEQ ID NO: 2 can be found either at position 153 a valine or an alanine, as explained in the experimental part in Example 3. However, equivalent experiments can be carried out taking into consideration the mutated GM2AP protein exhibiting mutations in both exon 1, in exon 2 and in exon 4, as identified in the sequence identifier in SEQ ID NO: 3. Figure. The appended figure represents the dose-response curve of the ternary complex GM2AP + MRP14 + GM2 (GM2: 50 μg / ml final). The quantities of MRP14 are represented on the abscissa (in ng) and the percentage of cytotoxicity corresponding to the percentage of dead cells is represented on the ordinate. In this figure, the quantities of GM2AP in ng are respectively represented by the following symbols: φ: 5 ng, π: 10 ng, Ψ: 20 ng, υ: 50 ng and χ: 100 ng. A similar experiment was carried out with the mutated GM2AP protein instead of the GM2 protein. The results obtained are similar to those presented in the appended figure.

Examples.

Example 1: MTT test protocol. (i) “Coating” of the plates with poly-L-lysine. 250 μl of sterile poly-L-lysine solution (12.5 μg / l) are deposited in all the wells of 48-well plates (Falcon 3078). After a 2 hour incubation at 37 ° C, the solution is aspirated and replaced with 250 μl of sterile water to wash the wells. Once the wells are emptied by aspiration, they are dried under the air flow of a microbiological safety cabinet.

(ii) Cells used The CLTTl-1 cells are astrocytes derived from this transgenic mouse expressing the large T gene of polyoma virus ¹³ . These cells are cultured at 37 ° C. in a humid atmosphere at 5% C0 ₂ , in Dubelcco's Modified Eagle's Medium (DMEM) / Ham's F12 medium (50/50) 4.5 g / 1 of D-glucose supplemented with 10 % of Fetal Calf Serum (SVF) not decomplemented, gluta ax (580 mg / 1), penicillin (500 units / 1) and streptomycin (500 μg / 1). (iii) Cytotoxicity test The samples to be tested are prepared 24 hours before deposit for the toxicity test and incubated at 4 ° C. The 48-well plates "coated" with poly-'L-lysine are seeded with CLTT 1-1 cells at the rate of 250 μl of cell suspension (6000 cells / ml) per well, or 1500 cells / well. After 24 hours of incubation at 37 ° C, in a humid atmosphere and 5% CO 2, the samples are deposited on the surface of the cell medium. Each sample is deposited in triplicate. Certain wells make it possible to evaluate cellular controls (T) (no sample deposit) or so-called TUC controls (deposit of 10 μl of TUC solution). The TUC reagent (TRIS 20 mM, urea 250 mM, CaCl ₂ 1 mM) is a solution mimicking the chemistry of urine. The deposit is homogenized and, to avoid evaporation, a protective film is applied to the top of the plates. After 72 hours of incubation at 37 ° C, in a humid atmosphere and 5% of CO ₂ , the revelation by the MTT test is carried out. The cell supernatant is aspirated, taking care not to remove the cells from the bottom of the wells. 250 μl of MTT solution (0.5 mg / ml in culture medium) are gently deposited on the cells. After 3 hours of incubation at 37 ° C, the solution is aspirated and the formazan crystals formed in the cells are solubilized with isopropanol, HC1 IN (40 μl / ml). Once homogeneous, 70 μl of solution from each well of the 48-well plate are transferred to the wells of a 96-well plate, in order to read the optical density. The absorbances are read at 570 nm / 650 nm. The percentage of cytotoxicity can be calculated: MoyT = average of absorbances of controls σT = standard deviation of absorbances of controls CO ^" = CutOff = MoyT - 2 σT DO = average absorbance of samples% toxicity = (1 - (DO / CO)) x 100 To be valid, the absorbances of each sample (in triplicate) must not have a standard deviation greater than 10% of the average absorbance.

Example 2: Preparation of urine pools. 100 liters of MS urine (0.2-0.5 liters from patients' first morning urination) were collected. The urine of patients contaminated with bacteria or that of patients treated with drugs likely to interfere with the bioassay of gliotoxicite ⁴ were eliminated. The individual samples were tested for gliotoxicity and a final pool of 46 liters of urine with significant gliotoxicity, by the MTT test, was selected. In parallel, an equivalent volume of urine from healthy donors with negative gliotoxicity for each sample was obtained. The stages of concentration and purification of this material, the protein analysis and the identification strategy are presented below. • Purification of urinary proteins. The SEP positive and SEP negative urine pools were purified to obtain a high protein concentration. (i) Precipitation: Precipitation with ammonium sulphate (Prolabo - ref. 21 333 365) was carried out on the pools of SEP positive and SEP negative urine. The percentage of 60% of saturated ammonium sulphate for 40% of urine, ie 390 grams of ammonium sulphate per liter of urine was used. Each pool is divided into 1.8 liter fractions in 2 liter flasks to improve precipitation. The precipitation was carried out for 2 x 8 hours, at ambient temperature, with gentle stirring. After centrifuging the urine pools at 3000 rpm for 10 min., At a temperature of 10 ° C., the pellet obtained is taken up in a 20 mM Tris buffer containing 1 mM CaCl ₂ and 0.25 M urea. The mixture was then centrifuged at 3000 rpm for 10 min. The supernatant contains the concentrated proteins. It is either used immediately for the next step, or frozen if the next step cannot be performed continuously. (ii) Ion exchange chromatography: The protein-containing solution was then passed over a DEAE Fast Flow gel (trade name, marketed by PHARMACIA). This step is carried out at low pressure on a PHARMACIA column filled with gel. The pads are brought to the column by a peristaltic pump which allows a regular flow. The column equilibration buffer is the 20 mM Tris buffer, pH 7. The fraction corresponding to the precipitation supernatant and containing an excessively high amount of salts is dialyzed against this buffer before depositing on the column. Elution by a salt gradient makes it possible to recover the proteins. The elution gradient is carried out in steps of NaCl 100, 200, 300, 500 mM in the equilibration buffer of the column. The elution fractions are tested by the MTT test. Only the positive fractions, that is to say the fractions eluted at 200 Mm NaCl, are preserved. These fractions are treated immediately or stored in the lyophilized state, (iii) Purification: A steric exclusion chromatography based on the difference in size and shape of the proteins to be eluted was used. The fraction corresponding to the 200 mM NaCl elution is deposited on the column. During elution, proteins of low molecular mass are retained and therefore eluted later than large molecules. The purifications were carried out on HPLC with a TosoHaas TSK Prep G 3000 SW column, with a diameter of 21.5 mm and a length of 300 mm. The molecular weight exclusion limit is 500,000 daltons. The elution buffer used contains 100 mM phosphate, 100 mM sodium sulfate, pH 6.8. The separation of the protein mixture was carried out in 60 min. Only the fraction corresponding to a mass of 15-20,000 daltons has been kept. This fraction is dialyzed in a 20 mM Tris buffer containing 0.2 mM CaCl ₂ , pH 7.2, then lyophilized. At each step, only the fractions with significant toxic activity were selected for the next step. A check of the toxic activity of the proteins was carried out at each step, using the MTT test. only the fractions exhibiting significant toxic activity were retained for the additional purification step. (iv) Additional Purification of Urinary Proteins by Reverse Phase Chromatography: The urine pools from MS patients (positive MS pool) and non-MS patients (negative MS pool), obtained after purification, were taken up in distilled water, then diluted with a 0.2% TFA / 10% acetonitrile solution to obtain a final concentration of approximately 130 to 140 μg / ml. The separation by reverse phase C8 HPLC was carried out on a Brownlee Aquapore column (trade name) sold by the company Perkin Elmer (characteristics of the column: 300 angstroms / 7 μm / (100 × 4.6)). Two separate columns were used for the positive and negative pools, respectively. The injections were carried out by multi-injections of 250 μl. The proteins were eluted with a linear gradient of 5% to 15% of buffer A in 5 min., Then from 15% to 100% of buffer B in 95 min., At a flow rate of 0.5 ml / min. The separation buffers A and B used are respectively the 0.1% TFA buffer (Pierce n ° 28904) / MilliQ water and the 0.09% TFA / 80% acetonitrile buffer (Baker). Detection was carried out by measuring the UV absorbance at 205 and 280 nm. The fractions were collected in 1.5 ml and 0.5-1 ml fractions in the area of interest. The fractions were frozen after collection in dry ice. The collected fractions were then dried in Speed Vac and taken up in 100 μl of 0.1% TFA / 30% acetonitrile. 20 μl of the fractions were transferred into 500 μl eppendorfs, dried and washed twice with 100 μl of MilliQ water, then dried again. The toxic activity of the proteins contained in each fraction collected after elution was determined using the MTT test. Only fraction X76 / 43 of the positive SEP pool exhibits toxic activity in vitro. The number of this fraction corresponds to the order of the elution as a function of the elution conditions set out in the example. Its toxic activity was confirmed in vitro by FACS on murine astrocytic cells, as described in application WO 98/11439. Its profile on SDS-PAGE revealed protein bands at 55 kDa, 35kDa, 20 kDa, 18 kDa, 14kDa and 8 kDa. The corresponding fraction X76 / 43 of the negative SEP pool, obtained from the control urine, showed no toxic activity by the MTT test. His profile on SDS-PAGE showed bands at 55 kDa, 35 kDa and 20 kDa. • Analysis of the proteins obtained by separation on HPLC on SDS-TRICINE gel. The protein content of fraction X76 / 43 of the negative SEP control pool and of fraction X76 / 43 of the positive SEP pool was observed after separation on 16% SDS-TRICINE gel and staining of the gel with zinc / imidazole. The collection pool for fraction X76 / 43 obtained by HPLC was deposited on a 16% SDS-TRICINE gel pre-molded with 10 wells and 1 mm thick (sold by the company Novex). The conditions of use of the gel correspond to those recommended by the supplier. The sample is taken up in 75 μl of the sample buffer 1 time concentrated (SDS-TRICINE N ° LC 1676, 1 ml twice concentrated + 50 μl of β-mercaptoethanol (Pierce) diluted 1/2 in water) and 25 μl of the sample are deposited on the gel three times. The collection pool for fraction X76 / 43 from the negative SEP pool was deposited on the gel under the same conditions as those described for the positive SEP pool. The migration on the two gels was carried out in parallel in the same migration tank (XCELL II NOVEX (trade name)) at a constant voltage of 125 V for 2 hours. The tank is placed in a container containing ice. The gels were stained directly after migration by zinc / imidazole staining (staining kit 161-0440 sold by the company BIORAD) to obtain a reversible negative staining. • Trypsin digestion of the gel strips. All the protein bands visualized in the deposits of fraction X76 / 43 were cut and subjected to proteolysis in a trypsin solution overnight. The gel strips are cut with a scalpel into 1 mm slices and transferred into eppendorf tubes. The eppendorfs are subjected to a centrifugation peak to bring down the pieces of gel and after centrifugation 100 μl of washing buffer (100 Mm NH ₄ CO ₃ /50% CH ₃ CN) are added to the pieces of gel. After 30 min. stirring at room temperature, the supernatant is removed in fractions of 20 μl and the washing step is repeated twice. The eppendorfs are dried for 5 min. in speed vac. 20 μg of trypsin (Modified sequenal grade PROMEGA V5111) (trade name) are taken up in 200 μl of digestion buffer (5 mM TRIS, pH 8) and are dissolved for 30 min. at room temperature, with intermittent stirring and 20 to 30 μl of resuspended trypsin are added to the gel pieces. The eppendorfs are centrifuged and stored in a hot room at 28 ° C overnight. After digestion the gel strips can be used immediately for mass measurements or frozen for later use. High apparent molecular weight proteins were found in both fractions. On the other hand, the bands of apparent molecular weights of 8, 14, 18 kDa are only visible in the fraction X76 / 43 of the positive SEP pool. Example 3: Mass spectrometry and protein sequencing. • MALDI-TOF mass spectrometry analysis of proteolytic fragments. 30 μl of extraction buffer (2% TFA / 50% acetronitrile) are added to the samples. The eppendorfs to be analyzed are subjected to a centrifugation of 5 min., Then to a sonication of 5 min. and finally to a centrifugation of 1 min. On a stainless steel disc, 14 deposits of 0.5 μl of matrix (α-cyano-4-hydroxy-trans-cinnamic acid saturated in acetone) are made. A thin uniform microcrystalline layer is obtained. 0.5 μl of a 2% TFA / water solution are deposited on this sublayer over the 14 deposits, then 0.5 μl of sample to be analyzed are added. In this drop thus formed, 0.5 μl of a saturated solution of Oc-cyano-4-hydroxy-trans-cinnamic acid in 50% acetonitrile / water are added. After drying at room temperature for 30 min., The crystalline deposits are washed with 2 μl of water which are immediately removed by a breath of air. All the spectra are obtained on a BRUKER BIFLEX (trade mark) mass spectrometer equipped with a reflectron. The measurements (90 to 120 laser shots over the entire deposit) are accumulated to obtain a mass spectrum which is the most representative of all the peptides present in the matrix-sample sandwich. For each deposit, a calibration with the peptides of the trypsin autolysis was made in order to be able to use a measurement accuracy of less than 100 ppm. The results of mass spectrometry were sought in the databases using the MS-FIT algorithm of the Protein Prospector software (http A / prospector .ucsf.edu). • N-terminal sequencing of digestion peptides. - (i) Extraction and separation by HPLC of the digestion peptides. The peptides obtained after digestion with trypsin were extracted in 3 times 30 min. in a sonication bath with 0.1% TFA / 60% acetonitrile. The extraction solutions are combined and dried to 20 μl in speed vac. After dilution in 80 μl of buffer A (0.1% TFA / water), the extractions of the gel bands, digested with trypsin, are injected on a column C18 / MZ- Vydac / (125x1, 6) mm / 5 μm (trade name). The elution of the peptides is done at a flow rate of 150 μl / min. and in a gradient ranging from 5% of buffer B (0.09% TFA / 80% acetronitrile) to 40% of buffer B in 40 min., then from 40% of buffer B to 100% of buffer B in 10 min. The detection is made by measuring the UV absorbance at 205 nm. The peaks are collected in 500 μl eppendorf tubes. The individual peptide peaks collected are then subjected to an analysis of the N-terminal amino acid sequence. - (ii) N-terminal sequencing: The fractions corresponding to a single mass peak were analyzed by Edman degradation on a sequencer (Model 477A PERKIN ELMER / Applied Biosystems). The sequencing conditions are those described by the manufacturer. A micro cartridge was used for depositing the samples and the PTH-AminoAcid are identified with an online HPLC system (Model 120A PERKIN ELMER / Applied Biosystems). •Results

The results of the mass spectrometry analysis and sequencing are presented in Table 1 below.

Table 1

MW: mean molecular weight ISM: identification by mass spectrometry IS: identification by sequencing NI: remaining peaks not identified ND: not determined *: identical to the G-terminal 20 kDa fragment of perlacan probably resulting from the previous proteolysis of the complete protein of 467 kDa in the urine or during the purification process.

A mixture of co-purified proteins was still present both in the final SEP purification fraction and in the corresponding control fraction. The proteins identified in these two samples were considered to be irrelevant due to the absence of gliotoxic activity in these two fractions. Consequently, GM2AP (18kDa), MRP14 or calgranulin B (14kDa), and saposin B (10 kDa) were considered as potential candidates for gliotoxic activity. Furthermore, the N-terminal sequencing of the trypsin-digested fragments of the 18 kDa band in the SEP fraction showed the presence of polymorphism in different positions of GM2AP. A mutation in exon 1, at position 19 of the amino acid sequence of GM2AP, where an alanine is replaced by a threonine. A mutation in exon 2, where an aspartic acid is replaced by a phenylalanine at position 40 of the amino acid sequence of GM2AP. This mutation has never been found in the genomic DNA of healthy or sick donors. Two other mutations in exon 2, respectively at positions 59 and 69 of the amino acid sequence of GM2AP which correspond to the replacement of an isoleucine by a valine and of a methionine by a valine. A mutation in exon 4, which consists of replacing a valine with an alanine at position 153 of the amino acid sequence of GM2AP, appeared to be a new polymorphism not described after different sequencing of the genomic DNA of lymphocytes from healthy individuals (blood donors) and patients with multiple sclerosis. This mutation in exon 4 was found in 3 out of 27 MS patients tested, as well as in 8 out of 27 control individuals suggesting a normal polymorphism. Another mutation is found in exon 4, at position 171 of the amino acid sequence of GMPA2, where a lysine is replaced by a glutamine. The amino acid sequences of GM2AP and mutated GM2AP are respectively represented in the sequence identifier in SEQ ID NO: 1 and SEQ ID NO: 2, it being understood that in these two sequences SEQ ID NO: 1 and SEQ ID NO: 2 we can find either at position 153 a valine or an alanine, since the mutation in exon 4 for this position suggests a normal polymorphism. Example 4: Recombinant proteins. Recombinant proteins (purchased or produced by transfection) were used to assess the gliotoxic potential of the candidate proteins. The so-called “non-human” proteins, that is to say recombinant proteins produced in a prokaryotic expression system (E. coli) by transformation with a plasmid containing the insert to be expressed, or in a eukaryotic expression system in yeasts or insect cells infected with baculovirus having integrated the insert to be expressed, the following were used: The protein MRP14 (or Calgranulin B or S100A9) fused in N-terminal with a histidine tail and produced in E. coli; the protein MRP8 (or Calgranulin A or S100A8) produced in E. coli; and the native human heterocomplex MRP14 / MRP8 (or Calprotectin), purchased from Dr. C. Kerkhoff (University of Munster, Germany). The protein GM2AP (precursor of the activator of the ganglioside GM2) fused in N-terminal with a histidine tail produced in Baculovirus and the protein Sap B (Saposin B) produced in yeast, purchased from Pr K. Sandhoff (Kekule Institute, University of Bonn, Germany). These proteins have their own physiological activity described in the literature. The so-called “human” proteins, that is to say recombinant proteins produced in a eukaryotic expression system in human cells transfected with an appropriate plasmid having integrated the insert to be expressed were produced according to the protocol described below. The 293T cells (primary human embryonic kidney cells transformed with a type 5 adenovirus, expressing the T antigen) were cultured at 37 ° C. in a humid atmosphere at 5% CO ₂ , in DMEM 4.5 g / l of D-glucose supplemented with 10% of Fetal Calf Serum (SVF) decomplemented, glutamax (580 mg / 1), penicillin (500 units / 1) and streptomycin (500 μg / 1). To carry out the transient transfection, appropriate plasmids containing the cDNA of the proteins of interest, MRP14, GM2AP and GM2AP mutated (Aspartic acid / Phenylalanine / position 40) preceded by a secretory peptide (IgK) at the N terminal, . The 293T cells are transfected with a “Transfectant” reagent composed of lipids which complex and transport the DNA in the cells. The 293T cells are trypsinized, seeded to 2 million cells per 75 cm ² flask, and incubated overnight at 37 ° C, in a humid atmosphere and 5% C0 ₂ in 10 ml of culture medium (DMEM 4.5 g / 1 of D-glucose supplemented with 10% of decomplemented fetal calf serum (SVF), glutamax (580 mg / 1), penicillin (100 units / ml) and streptomycin (100 μg / l)). The transfection solution is prepared extemporaneously using the ratio 3/2 [volume of Transfectant (μl) / quantity of plasmid DNA (μg)] qs 1ml of ^" medium without SVF. After 45 minutes of contact at room temperature, the solution transfection is added dropwise onto a non-confluent cell mat After 72 hours of incubation at 37 ° C., in a humid atmosphere and 5% CO ₂ , the supernatants are recovered and centrifuged for 10 minutes at 2500 rpm. produced is then carried out either by the MRP Enzyme Immunoassay assay kit (trade name) marketed by BMA Biomedicals AG, Augst, Switzerland, following the instructions for the recombinant human protein MRP14, or by the technique of semi-quantitative Western Blot with antibodies polyclonal of anti-GM2AP rabbit. These techniques give indicative values for a relative comparison. The crude supernatants from this production will be used in particular for toxic activity and detection tests.

Example 5: Toxicity of "non-human" proteins. The toxicity of the “non-human” recombinant proteins MRP14, MRP8, GM2AP, SapB was evaluated by the MTT test. Proteins were tested in a defined range from the assessment of the concentration of each protein in different urine. The ranges are produced in different buffers, either in the TUC solution, or in two types of urine: urine from patients with multiple sclerosis who were toxic by the MTT test (urine MS), and urine from '' recruitment of non-MS donors who were not toxic by the MTT test (normal urine). The urine had previously been treated 30 min. at 56 ° C and filtered. The results show that, taken individually, the proteins tested in the TUC solution and in normal urine are not toxic by the MTT test. No significant effect of the GM2AP, MRP14 and Saposin B proteins is demonstrated in MS urine by the MTT test. There is an inhibition of toxicity with an MRP8 dose equal to or greater than 3 ng. These results are presented in Table 2.

Table 2A: Range in the TUC solution

For MRP14 and MRP8 it is a percentage of average cytotoxicity over 2 ND tests: not determined * in another MS urine, the same inhibition of toxicity is observed

Combinations of GM2AP / MRP14, Saposin B / MRP14 and Saposin B / GM2AP / MRP14 proteins were then prepared in TUC eb solution in both types of urine as described above. In “control” combinations, the MRP14 / 8 heterocomplex or the MRP8 protein replace the MRP14 protein in the different GM2AP / MRP14 / 8, Saposin B / GM2AP / MRP8 combinations. The “control” combinations were prepared in the same way. All the combinations were incubated overnight at 4 ° C before being tested for their toxicity by the MTT test. The results are presented in Table 3.

Table 3A Range in the TUC solution

MRP14 / 8: native human heterocomplex Sap. B: Saposin B *: average of two tests The results in Table 3A show that the combinations GM2AP / MRP14, GM2AP / MRP14 / 8, Saposin B / MRP14 and GM2AP / MRP14 / Saposin B have no toxic effect in TUC, whatever the quantity tested. Only the GM2AP (10ng) / MRP14 (0.5ng) combination seemed to present a toxicity, but this toxic activity was not subsequently found in two additional comparable trials. In addition, additional tests have been carried out with the GM2AP / MRP14 combination using different amounts of GM2AP and of MRP14. The results obtained confirmed that the GM2AP / MRP14 combination has no toxic effect in TUC, whatever the quantity tested.

Table 3B: Range in normal urine.

Sap. B: Saposin B

As can be seen from Table 3B, the GM2AP / MRP14 combination is toxic in normal urine since the toxicity increases as the amount of GM2AP protein increases. However, this toxicity appears not very stable and not very reproducible and seems to be dependent on the urine sample (see comparison of the percentage of cytotoxicity between normal urine 1 and normal urine 2, in Table 3B). The combination of Saposin B / MRP14 is at the limit of significance in normal urine. The results obtained with the GM2AP / MRP14 / Saposin B combination are difficult to interpret. The toxicity of the combinations of GM2AP / MRP14 and Saposin B / MRP14 proteins was also tested against normal urine and toxic urine from patients with multiple sclerosis (MS urine). The results are presented in Table 3C.

Table 3C Range in non MS and MS urine

Sap. B: saposin B

The combination Saposin B / MRP14 has no toxic effect in normal urine and in MS urine, regardless of the amount tested. The GM2AP / MRP14 combination does not have a toxic effect with respect to normal urine 1, but has a toxic effect with respect to normal urine 2 (when GM2AP increases, the toxicity of urine increases). There is also an opposite effect with respect to MS urine. When the amount of MRP14 increases, the toxicity of urine decreases.

Example 6 Toxicity of “Human” Proteins The GM2AP, GM2AP proteins mutated in exon 2 and MRP14 produced as described in Example 3 were tested for their toxicity by the MTT test, using culture supernatants from 293T cells. containing them. The following combinations were also carried out using culture supernatants from 293T cells: GM2AP / MRP14, mutated GM2AP / MRP1, GM2AP / MRP14 / MRP8. The combinations prepared were then incubated overnight at 4 ° C, then they were tested for their toxicity by the MTT test.

The results are presented in Table 4.

Table 4A

C%: cytotoxicity in percentage ND: not determined Approximate protein concentrations in the supernatants MRP14 lots 1 and 2350 ng / ml; GM2AP batch 1: 300 ng / ml batch 2: 200 ng / ml

For certain values, indicated in bold type, certain GM2AP / MRP14 combinations are weakly cytotoxic (from 20 to 30% of cytotoxicity) with an optimum for the GM2AP (20 ng) / MRP14 combination (0.5 ng). MRP14 alone is not cytotoxic. GM2AP alone is not considered to be cytotoxic, even if very low toxicity was found in tests 1 and 2 carried out on batch 2. In fact, the reproducibility cannot be perfect because it depends on the production batch of the supernatants . Table 4B

C%: cytotoxicity in percentage ND: not determined Approximate concentration of GM2AP and MRP14 in the supernatant: 2 μg / ml Rejection:% of cytotoxicity rejected because the standard deviation of the OD of the samples is greater than 50 *: standard deviation of the OD of the samples between 16 and 11 No comment: standard deviation of the OD of the samples less than 10.

The results show that the proteins alone, in the supernatants, are not toxic, except in a non-specific manner in very large quantities (100 ng of MRP14). Only the GM2AP (100 ng) / MRP14 (100 ng) combination can be considered to have relative cytotoxicity.

If the GM2AP protein is replaced by the GM2AP protein mutated in this combination, the same type of toxicity is obtained for certain mixtures, as shown below.

Table 4C

C%: percentage cytotoxicity *: standard deviation of the ODs of the samples between 14 and 11 Without comment: standard deviation of the ODs of the samples less than 10 Approximate protein concentration in the supernatants: mutated G 2AP: 200 ng / ml; MRP14: 350 ng / ml.

For many values, indicated in bold type, the mutated GM2AP / MRPl4 combination is toxic. The mutated GM2AP protein alone has no cytotoxic effect. MRP14 alone is considered to have no cytotoxic activity.

The cytotoxicity of the combinations of supernatants containing the recombinant human proteins, GM2AP / MRP14 and mutated GM2AP / MRPl4 is found in the same order of magnitude, with greater stability depending on the production batch of the proteins, than with non-recombinant proteins. human. However, this does not correspond to the stability, reproducibility and intensity of the gliotoxic activity found in the biological fluids of MS patients. 4D table

C%: percentage cytotoxicity Approximate protein concentration in the supernatants: GM2AP (lot 1): 300 ng / ml, GM2AP (lot 2): 200 ng / ml. Concentration of native MRP14 / 8: 1.3 mg / ml. It appears from the results of Table 4D that GM2AP alone has no cytotoxic activity and that for certain values, indicated in bold, the combination GM2AP / MRP14 / MRP8 has a cytotoxic effect. This cytotoxicity is dependent on the batch of supernatant used.

Table 4E

C%: percentage cytotoxicity Approximate protein concentration in mutated GM2AP supernatants: 200 ng / ml. Concentration of native MRP14 / 8: 1.3 mg / ml.

It appears from Table 4E that the mutated GM2AP / MRP14 / MRP8 combination does not exhibit cytotoxic activity.

These studies show that none of the proteins identified in the gliotoxic fraction purified from MS urine reproduced, alone, the desired gliotoxic activity and that the combinations of proteins produced in the form of "non-human" or "human" recombinants only reproduce weakly and not very reproducible (even if an improvement is noted with “human” recombinants) the gliotoxic activity. The obtained results do not meet all the criteria characterizing gliotoxic activity (high activity, stability, reproducibility, dose-response effect).

The results show that an essential component is missing which was not identified in the protein analysis.

The inventors then surprisingly found that lipids, in particular complex lipids, are interesting candidates in this context. To this end, ganglioside GM1, ganglioside GM2 and sulfatide were tested. Among these lipids, ganglioside GM2 proved to be the only convincing, as shown by the examples which follow.

Example 7: Toxicity of “human” recombinant proteins in association with ganglioside 6M2. The ganglioside GM2 (supplied by Professor J. Portoukalian (Lyon France)) is added at a concentration of 50-— μg / ml final to the combinations of —recombinant “human” proteins already produced, involving the proteins MRP14, GM2AP and GM2AP mutated . The combinations GM2AP / MRP14 and mutated GM2AP / MRPl4 were tested in a range of proteins: 0, 5, 10, 20, 50, 100 ng for the recombinant proteins GM2AP and mutated GM2AP and up to 200 ng for the protein MRP14. These ranges were made in association or not with ganglioside GM2. After mixing, the combinations are incubated overnight at 4 ° C, their toxicity is then evaluated by the MTT test.

The results obtained are described in Table 5 and in the appended figure. Table 5A Measurement of the gliotoxic activity of “human” proteins combined and associated with the ganglioside GM2 (50 μg / ml final)

C%: cytotoxicity in percentage gGM2: GM2 ganglioside *: OD standard deviation of the samples between 13 and 11 Without comment: OD standard deviation of the samples less than 10 Approximate concentrations in the supernatants: GM2AP and MRP14: 2 μg / ml

The combination GM2AP / MRP14 associated with a constant concentration of ganglioside presents a gliotoxic effect which increases in parallel with the quantity of protein MRP14. In addition, for increasing amounts of the GM2AP protein (20 and 10 ng), a typical dose-response effect increasing in stages is obtained. However at the extreme points, when there is not enough GM2AP protein (5 ng) there is no toxicity. On the contrary, if there are any too much GM2AP protein (50 ng and 100 ng) there is saturation of toxicity with a plateau at around 60%. In fact, only CLTTl-1 cells proliferating in the culture during exposure to the gliotoxic factor are sensitive. This explains that the gliotoxicity plateaus do not reach 100%.

Table 5B Measurement of the gliotoxic activity of "human" proteins combined, associated or not, with ganglioside GM2 (50 μg / ml final)

C%: cytotoxicity in percentage gGM2: ganglioside GM2 *: OD standard deviation of the samples between 13 and 11 Without comment: OD standard deviation of the samples less than 10

Without GM2 ganglioside, the mutated GM2AP / MRPl4 combinations are not gliotoxic. An overall increase in the cytotoxicity of the mixture with the ganglioside GM2 is observed compared to the combinations without ganglioside. The variability of the measurements is apparently greater with the use of the mutated GM2AP protein. Overall, the activity appears significant and reaches a maximum plateau (cf.: maximum reached on the pool of proliferating cells during the test, as previously discussed) for the highest concentrations, according to a two-variable dose effect, GM2AP mutated and MRP14.

In order to know whether the action of GM2 ganglioside is specific for the toxicity of combinations of recombinant human proteins GM2AP / MRP14 (5 ng of MRP14 and 50 ng or 100 ng of GM2AP), other lipids were tested in parallel: ganglioside GM1 and sulfatide. The concentration ranges used are 0, 10, 20, 30, 50 μg / ml final. Once the lipids have been added, the combinations are incubated overnight at 4 ° C., their toxicity is then evaluated in the MTT test. The results, presented in Tables 5C and 5D, show that only the associations with the ganglioside GM2 for the GM2AP / MRP14 combinations at the doses 30 μg / ml and 50 μg / ml are toxic for glial cells (27% and 30% respectively). The other lipids do not show any toxicity with the protein combinations.

Table 5C Influence of the ganglioside GM2 in the gliotoxic activity of the recombinant “human” proteins combined GM2AP / MRP14

For the GM2AP 100 ng test, this is an average of two tests. gGM2: GM2 ganglioside

Table 5D Influence of the ganglioside GM2 in the gliotoxic activity of the “human” recombinant proteins combined GMA2AP / MRP14

The results of the study show that: the activity is associated with a protein heterocomplex involving the proteins GM2AP or GM2AP mutated and MRP14; it is the addition of a lipid, such as ganglioside GM2, which has made it possible to obtain activity levels, reproducibility and dose-response effects, compatible with the reproduction of the desired gliotoxic activity; the mutation found on the GM2AP protein is not essential to the determinism of gliotoxin in vitro. However, in vivo, it can be decisive if it is necessary for the bioavailability process of the GM2AP protein (for example in the extracellular medium of the central nervous system). These elements therefore demonstrate that a mutated MRP14 / GM2AP or MRP14 / GM2AP heterocomplex associated with the ganglioside GM2 is the main, even unique, vector of gliotoxic activity.

Example 8: Development of an immunoassay of the gliotoxic complex - Preparation of the samples before the ELISA test. (i) Samples tested. The samples tested are: on the one hand human recombinant proteins in combination (GM2AP + MRP14) with or without GM2 ganglioside, diluted or not in normal urine, in order to detect the active recombinant complex, on the other hand normal urine and MS for direct detection in urine. The samples, once prepared, are incubated for 24 hours at 4 ° C before the detection test. The “human” recombinant proteins are used in the form of crude production supernatants, recovered after the transient transfection of the 293T cells, with the appropriate negative controls in parallel. The MRP14 and GM2AP protein assay systems used are semi-quantitative and the quantities specified are indicative. The results are presented in the following examples. (ii) Processing of samples. As shown, in the following examples, the detection method using anti-MRP14 and anti-GM2AP antibodies in a “sandwich” ELISA format, make it possible to obtain positive results. However, the inventors have optimized this detection method by carrying out a preliminary treatment of the sample comprising a step of digestion by proteinase K of the proteins present, followed by a step of inactivation of this protease by an original method of precipitation. with trichloroacetic acid, then neutralization of the pH with a tris-maleate buffer, selected for its subsequent compatibility with an ELISA sandwich test. This treatment of the sample, original in its various stages, was subsequently applied to the analyzes which are presented in the following examples and is described in detail below. The samples (mixture of recombinant proteins or urine) are treated with proteinase K before detection of the complex according to the following protocol: 0.3 g of proteinase K is added for 100 μl of sample. After digestion for one hour at 37 ° C., precipitation with trichloroacetic acid is carried out in order to inhibit the action of proteinase K. Trichloroacetic acid 90% (90 g of trichloroacetic acid for 48 ml of distilled water ), is added to the sample (15% of the initial volume of the sample). The mixture is incubated for 30 minutes at 4 ° C. After centrifugation for 30 minutes at 13,000 rpm, the pellet is taken up with a volume equal to the initial volume of the sample with the 0.2 M TRIS Maleate buffer pH 6.2 (in tests without concentration factor) or in a minimum volume (to achieve a volume concentration of undigested proteins). After control by the Western Blot technique of the samples treated with proteinase K, an observation can be made and the treatment can be optimized by increasing the quantity of proteinase K and its time of action.

Example 9: Heterocomplex detection protocol in a sandwich ELISA assay. (i) Obtaining antibodies: the following antibodies were produced according to the protocols described below: polyclonal antibodies (bioMérieux): rabbit polyclonal antibody 196 (anti-pept ide MRP 14) rabbit polyclonal antibody 79 (anti-recombinant protein GM2AP). monclonal antibodies (bioMérieux): - 4A7B10 - 5H7C10 - 2B9H2 - 10E11A11 - 13H9C9 - 19C11C10 - 13D1E5 - 2G12H5

Anti-GM2AP monoclonal antibodies: 10E11A11, 13D1E5, 13H9C9, 19C11C10, 2G12H5. The mice were immunized according to the following protocol: on day D0 intraperitoneal injection of 75 μg of the GM2AP-MRP14 complex in the presence of complete Freund's adjuvant. On days D23, D37 new - intraperitoneal injection of the same amount of GM2AP-MRP14 complex in the presence of an incomplete Freund's adjuvant. Four days before fusion make an intravenous injection of 50 μg of GM2AP antigen diluted in physiological water. 1900 supernatants were screened by indirect ELISA technique. The plates were “coated” with 100 μl of antigen (the GM2AP-MRP14 complex) at 1 μg / ml in 0.05M bicarbonate buffer, pH 9.6. The “coated” plates were incubated overnight at a temperature of 18-22 ° C. The plates were saturated with 200 μl of PBS-1% milk and incubated for 1 hour at 37 ° +/- 2 ° C. 100 μl of supernatants or ascites liquid diluted in PBS-tween 20 buffer, 0.05% were added and the plates were incubated for 1 hour at 37 ° +/- 2 ° C. 100 μl of polyclonal goat anti-Ig (H + L) antibody mice conjugated to alkaline phosphatase (Jackson Immunoresearch ref: 115-055-062), diluted in PBS-BSA buffer 1% to 1/2000, were added and the plates were then incubated for 1 hour at 37 ° +/- 2 ° C. 100 μl of PNPP (Biomérieux ref 60002990) at a concentration of 2 mg / ml in DEA-HCL (Biomérieux ref 60002989), pH = 9.8, were added. The plates were incubated for 30 minutes at 37 ° +/- 2 ° C. The reaction was blocked by adding 100 μl of NaOH, IN. Three washes are carried out between each step with 300 μl of PBS-tween 20, 0.05%. An additional wash in distilled water is carried out before adding the PNPP. 150 supernatants were positive in indirect ELISA with an OD> 0.9. After the specificity tests the five aforementioned antibodies are produced. Anti-MRPl4 monoclonal antibodies: 2B9H2, 4A7B10, .5H7C10. The mice were immunized according to the following protocol: on day D0 an intraperitoneal injection of 75 μg of the GM2AP-MRP14 complex in the presence of complete Freund's adjuvant. On days D23 and D37 intraperitoneal injection of Aa the same amount of complex in the presence of incomplete Freund's adjuvant. Four days before fusion an intravenous injection of 50 μg of MRP14 antigen diluted in physiological water. 1100 supernatants were tested and screened by the indirect ELISA technique, as described above. 300 supernatants tested positive with an OD> 1. After the specificity tests the three aforementioned antibodies are produced. Rabbit polyclonal antibody 79 (recombinant anti-protein GM2AP). The rabbits were immunized according to the following protocol: on OJ day, the ^1st blood test of 10 ml, 75 μg of GM2AP were injected intraperitoneally in the presence complete Freund's adjuvant (AFC) (75 μg of immunogen + qs 0.5 ml of physiological water 9 ° / oo + 0.5 ml AFC). On days D28 and D56 the same amount of immunogen was injected intraperitoneally under the same conditions in the presence of 0.5 ml of incomplete Freund's adjuvant (AFI). In a 2 ^nd day J63 jack da blood of 30 ml was performed by ear without anticoagulant. A 3 * ^my blood test was performed under the same conditions on day J70. Rabbit polyclonal antibody 196 (anti-peptide MRP14). Rabbits were immunized according to the following protocol: Rabbits were immunized according to the following protocol: at day, the ^ere taken 10 ml of blood, 80 ug of immunogen was injected intraperitoneally in the presence of adjuvant Complete Freund (AFC) (80 μg of immunogen + qs 0.5 ml of physiological water 9 ° / ₀₀ 4- 0.5 ml AFC). On days D28 and D56 the same amount of immunogen was injected intraperitoneally under the same conditions in the presence of 0.5 ml of incomplete Freund's adjuvant (AFI). On February 1 day J63 ^ven taking 30 ml blood was performed by ear without anticoagulant. A ^3rd blood test was performed under the same conditions on day J70. These antibodies are used in capture or detection. When used in detection in the ELISA sandwich test, the antibodies are biotinylated. (ii) ELISA sandwich test: The treatment of the samples (proteinase K and TCA precipitation), if any, is carried out after the overnight incubation at 4 ° C. and before the sandwich ELISA detection test. The capture antibody is “coated” with 1 μg in carbonate-bicarbonate buffer (50 mM) pH 9.5, 100 μl are deposited in the wells of a 96-well microplate. The plate is covered with a protective film and incubated overnight at room temperature. After 3 washes in PBS (Phosphate Buffered Saline) Tween 0.05%, the non-specific sites are blocked by PBS Tween 0.05%, goat serum (1/10 °) for monoclonal antibodies or 100 μl of hydrolyzate of casein for polyclonal antibodies. After 3 washes in 0.05% Tween PBS, the samples, treated or not, are deposited at the rate of 100 μl per well and incubated for 1 hour 30 minutes at 37 ° C. with shaking. After 3 washes in 0.05% Tween PBS, 100 μl of detection antibodies biotinylated at 1 μg / ml are deposited in each well and incubated for 1 hour 30 minutes at 37 ° C. After 3 washes in 0.05% Tween PBS, 100 μl of streptavidin coupled to HRP (horseradish peroxidase) at 0.2 μg / ml are placed in each well and incubated for 1 hour 30 minutes at 37 ° C. in order to amplify the signal. After 3 washes in PBS Tween 0.05%, 100 μl of OPD solution (orthophenylene diamine dihydrochloride) at 2 g / 1 are placed in each well and incubated for 10 minutes at room temperature. The reaction is stopped with 100 μl of H ₂ S0 ₄ IN. The optical density is read at 492 nm. Likewise for the production of anti-heterocomplex antibodies, the GM2AP / GM2 / MRP14 or mutated GM2AP / GM2 / MRPl4 hererocomplex is used as an immunogen to immunize BALB / c mice by intraperitoneal injection. The first injection is made with complete Freund's adjuvant. The other injections are given 4-8 weeks apart with incomplete Freund's adjuvant. A final reminder is made a few days before the fusion in physiological water. After this booster, the spleens of the immunized mice are removed and the splenocytes are collected. Then we carry out the cell fusion splenic with cells of a myeloma line and the cells secreting antibodies are selected which recognize in ELISA the heterocomplex used for immunization. Finally, the clones producing antibodies specific for the immune heterocomplex are selected, that is to say which do not recognize either GM2AP, or mutated GM2AP, nor MRP14, alone.

Example 10: Detection of the human recombinant heterocomplex. The immunoenzymatic assays of the gliotoxic activity characterized molecularly in the previous examples pass through an antigen / antibody system involving only the proteins involved (the proteins GM2AP, GM2AP mutated and MRP14) and by antibodies (alone or in combination) able to detect this molecular complex. The recombinant complex corresponds to the association of the supernatants of recombinant proteins GM2AP (1000 ng) and MRP14 (50 ng) associated with 50 μg / ml final of ganglioside GM2. (i) Detection of the recombinant gliotoxic heterocomplex without treatment with proteinase K. In order to know whether the toxic combinations were directly detectable, the “human” recombinant proteins MRP14 and GM2AP are combined with the ganglioside GM2, incubated overnight at 4 ° C and tested by the sandwich ELISA test using anti-MRP14 and anti-GM2AP antibodies. The combinations [MRP14, GM2AP and ganglioside GM2] are diluted in normal urine (not gliotoxic in the MTT toxicity test). The results are presented in Table 6. These results show that couples of anti-GM2AP capture antibodies / anti-MRP14 detection recognize the recombinant complex in an extremely reproducible manner. The results are presented in Table 6.

Table 6

Positive tests: number of positive tests Total tests: total number of tests,

(ii) Detection of the recombinant gliotoxic heterocomplex after treatment with proteinase K. As described above, the gliotoxic activity resists proteinase K. Also, by treating the samples (combination GM2AP + MRP14 + GM2) with proteinase K, uncomplexed proteins are destroyed, and background noise is reduced. As in the previous section, ^" the combinations are incubated at 4 ° C. overnight. But before testing them, the samples are treated with proteinase K and precipitated with TCA (trichloroacetic acid), according to the protocol described in the example. 8 (ii). The results are presented in Table 7. These results show in particular that the anti-MRPl4 / anti-GM2AP detection couples [4A7B10 + 5H7C10] / [13H9C9 + 19C11C10], [4A7B10 + 5H7C10] / 10E11A11, 2B9H2 / 10E11A11 and 2B9H2 / [13H9C9 + 19C11C10] detect the recombinant complex in the supernatants diluted in the urine, after treatment with proteinase K, in an extremely reproducible manner. The background noise is significantly attenuated.

Table 7

Positive tests: number of positive tests Total tests: total number of tests

Example 11: Detection of the heterocomplex in the urine of patients. Direct detection of the complex in the urine of patients was tested on two representative urines: MS urine and normal urine. The results are described in Table 8. These results show that the anti-MRP1 4 / anti-GM2AP detection antibody couples [4A7B10 + 5H7C10] / [13D1E5 + 2G12H5], [4A7B10 + 5H7C10] / 10E11A11, 2B9H2 / [ 13D1E5 + 2G12H5] and 2B9H2 / [13H9C9 + 19C11C10] detect the complex.

Table 8

For urine treated with proteinase K, there is no concentration with TCA

The methods described in the examples are useful as diagnostic tools, allowing the assay of a biological marker of multiple sclerosis, since the correlations between the gliotoxic activity and the clinic have proved to be very good ¹ ' ³ ' ⁴ . Bibliographic references:

1. Malcus-Vocanson, C. et al. (1998) A urinary arker for multiple sclerosis [letter]. Lancet 351, 1330.

2. Menard, A. et al. (1997) Gliotoxicity, reverse transcriptase activity and retroviral RNA in monocyte / macrophage culture supernatants from patients with multiple sclerosis. FEBS ett 413, 477-85.

3. Menard, A. et al. (1998) Detection of a gliotoxic activity in the cerebrospinal fluid from multiple sclerosis patients. Neurosci Lett 245, 49-52.

4. Malcus-Vocanson, C. et al. (2001) Glial Toxicity in urine and Multiple Sclerosis. Multiple Sclerosis 7, 383-388.

5. N. Benjelloun et al. Cell. Mol. Biol. , 1998, 44 (4), 579-583.

6. Blazar et al., (1997) Journal of Immunology 159: 5821-5833. 7. Bird et al., (1988) Science 242: 423-426.

8. Arakawa et al., (1996) J. Biochem 120: 657-662.

9. Chaudray et al., (1989) Nature 339: 394-397.

10. Jones et al., Nature 321: 522-525 (1986).

11. Reichmann et al., Nature 332: 323-329 (1988).

12. Presta et al., Curr. Op. Struct. Biol. 2: 593-596 (1992).

13. Galiana et al., J. Neurosci. Res. (1990) 26: 269-277.

Claims

1. Isolated cytotoxic factor, associated with multiple sclerosis, said cytotoxic factor being chosen from the heterocomplex GM2AP / GM2 / MRP14 and mutated GM2AP / GM2 / MRP14 in which GM2AP mutated corresponds to the sequence SEQ ID NO: 2.

2. Method for detecting and / or quantifying a cytotoxic factor, associated with multiple sclerosis, in a biological sample, according to which a heterocomplex chosen from the heterocomplex GM2AP / GM2 / MRP14 and mutated GM2AP is isolated from said biological sample / GM2 / MRP14 in which GM2AP mutated corresponds to the sequence SEQ ID NO: 2.

3. Method according to claim 2, according to which the heterocomplex is isolated using at least one antibody which specifically binds to the heterocomplex, and said cytotoxic factor is detected and / or quantified by demonstrating the formation of a complex constituted ^{"~ '} by the heterocomplex and the antibody.

4. Method according to claim 3, according to which the heterocomplex is isolated using at least two antibodies which specifically bind to the heterocomplex, and said cytotoxic factor is detected and / or quantified by demonstrating the formation of a complex consisting of the heterocomplex and the two antibodies.

5. Method according to claim 4, according to which at least one of said antibodies is a capture antibody and at least the other of said antibodies is a detection antibody.

6. Method according to claim 2, according to which the heterocomplex is isolated using at least two antibodies, at least one of which binds specifically to GM2AP or GM2AP mutated from the heterocomplex, and at least the other specifically binds to heterocomplex MRP14, and said cytotoxic factor is detected and / or quantified by demonstrating the formation of a complex consisting of the heterocomplex and the two antibodies.

7. The method of claim 6, wherein at least one of said antibodies is a capture antibody and at least the other of said antibodies is a detection antibody.

8. Method according to any one of claims 2 to 7, according to which the sample to be tested is subjected to a preliminary treatment comprising: a step of digestion of the proteins of the sample by proteinase K, a step of inactivation of proteinase K, and a pH neutralization step.

9. The method of claim 8, wherein the step of inactivation of proteinase K is performed by precipitation with trichloroacetic acid, and the step of neutralizing the pH is performed by adding a tris-maleate buffer.

10. Method according to any one of claims 2 to 9, in which the biological sample is chosen from serum, plasma, urine and cerebrospinal fluid.

11. Composition for the detection and / or quantification of a cytotoxic factor associated with multiple sclerosis, said cytotoxic factor being chosen from the heterocomplex GM2AP / GM2 / MRP14 and mutated GM2AP / GM2 / MRP14 in which GM2AP mutated corresponds to the sequence SEQ ID NO: 2, characterized in that it comprises at least one antibody which specifically binds to the heterocomplex.

12. Composition according to claim 11, characterized in that it comprises at least two antibodies which specifically bind to the heterocomplex.

13. Reaction mixture for the detection and / or quantification of a cytotoxic factor associated with multiple sclerosis, said cytotoxic factor being chosen from the heterocomplex GM2AP / GM2 / MRP14 and mutated GM2AP / GM2 / MR 14 in which GM2AP mutated corresponds to the sequence SEQ ID NO: 2, characterized in that it comprises at least two antibodies of which at least one binds specifically to GM2AP or GM2AP mutated from the heterocomplex and at least the other binds specifically to MRP14 of the heterocomplex.

14. Reaction mixture according to claim 13 characterized in that at least one of said antibodies is a capture antibody and at least the other of said antibodies is a detection antibody.

15. Complex comprising the heterocomplex GM2AP / GM2 / MRP14 or mutated GM2AP / GM2 / MRPl4, said heterocomplex being linked to at least two antibodies, at least one of the antibodies of which is specific for GM2AP or for mutated GM2AP and at least l the other antibody is specific for MRP14.