WO2010121714A1

WO2010121714A1 - Complexes between phospholipids and protein vimentin, and in vitro methods for the detection of antibodies against these complexes

Info

Publication number: WO2010121714A1
Application number: PCT/EP2010/002173
Authority: WO
Inventors: Maurizio Sorice; Elena Ortona
Original assignee: Diametra S.R.L.
Priority date: 2009-04-24
Filing date: 2010-04-02
Publication date: 2010-10-28
Also published as: ITRM20090195A1; IT1396046B1

Abstract

The present invention relates to an analytic in vitro method for the detection, in serum or other biological samples, of antibodies against the complex between phospholipids and the protein vimentin, in relation to the diagnosis of the Antiphospholipid Syndrome (APS) and other autoimmune diseases.

Description

COMPLEXES BETWEEN PHOSPHOLIPIDS AND PROTEIN VIMENTIN,

AND IN VITRO METHODS FOR THE DETECTION OF ANTIBODIES

AGAINST THESE COMPLEXES

Field of the invention

The present invention relates to an analytic in vitro method for the detection, in serum or other biological samples, of antibodies against the complex formed by the electrostatic interaction between phospholipids, in particular cardiolipin, and the protein vimentin .

Prior Art

The Antiphospholipid Syndrome (APS), also known as Hughes

Syndrome, is an autoimmune disease characterised by arterial and venous thrombosis and recurrent miscarriages (Bertolaccini M. L. et al., Nature Clin Practice (2005) 1:40-46; Atanassova P.A., Yonsei Med. J. (2007) 48: 901-

926).

The interpretation of the clinical manifestations and laboratory analysis, upon which the diagnosis of this syndrome is based, is made difficult primarily by their overlapping with those of other autoimmune diseases, particularly the Systemic Lupus Erythematosus, or SLE.

The current consensus requires that at least one of the clinical criteria and one of the laboratory criteria, shown in Table 1, should be met, in order for a confident diagnosis of APS to be formulated.

Table l: Consensus criteria for the diagnosis of APS (da Miyakis S. et al., J. Tromb. Haemost. (2006) 4:295-306). Clinical criteria

Vascular thrombosis

>1 events of thrombosis within veins or minor arteries in any tissue or organ, confirmed by instrumental or histopathological analysis in the absence of relevant evidence of vascular wall inflammatioN

Pregnancy

>1 unexplainable death of morphologically regular foetuses from the 10th week of pregnancy onwards

>2 premature birth of morphologically regular newborns from the 34th week of pregnancy, caused by severe pre-eclampsia, eclampsia or placental insufficiency

>3 unexplainable consecutive miscarriages before the 10th week of pregnancy (excluding hormonal or anatomical abnormalities in the mother and genetic causes)

Laboratory criteria

1. Lupus Anticoagulant, found in the plasma in two or more occasions at least 12 weeks from each other, analysed according to the guidelines of the International Society on Thrombosis and HaemostasisJ or

2. Anti-cardiolipin antibodies, IgG or IgM isotype, found in medium or high level in serum or plasma (i.e., > 40 GPL or MPL or above the 99th percentile) in two or more occasions at least 12 weeks from each other, measured by a standardised enzyme-linked immunosorbent assay (ELISA); or

3. Anti-β2-glycoprotein antibodies, found in serum or plasma (level above the 99th percentile) in two or more occasions at least 12 weeks from each other, measured by a standardised ELISA according to the recommended procedures. Medium or high levels of anti-cardiolipin antibodies (aCL), IgG and / or IgM isotype, in the presence of the cofactor β 2- glycoprotein I, measured by a standardised (ELISA), or a positive assay of Lupus Anticoagulant (LA), in two or more occasions at least 6 weeks from each other, tested according to the guidelines of the International Society on Thrombosis and Haemostasis.

The basic criteria shown in Table 1 include the knowledge resulting from observations gathered in recent years (Miyakis S. et al., J. Thromb. Haemost. (2006) 4:295-306). In particular, with regard to laboratory criteria, for a diagnose of APS it is necessary^

^■ to have at least one positive Lupus Anticoagulant (LA) test in two or more occasions at least 12 weeks from each other; or

- to display medium or high levels of aCL antibodies in the presence of the cofactor β2-glycoprotein I ( > 40 GPL or MPL, or above the 99th percentile) in two or more occasions at least 12 weeks from each other, measured by a standardised ELISA; or

- to display a level of the anti-β2-glycoprotein I antibodies > 99th percentile, in two or more occasions at least 12 weeks from each other, measured according to the recommended procedures by a standardised ELISA. MPL and GPL are international units used to quantify the positivity of antiCL antibodies of IgM and IgG class. In particular, one MPL o GPL unit is equivalent, respectively, to lμg/ml of IgM or IgG in the sample under examination.

The LA assay is a combination of coagulation assays, during which the possible coagulation lag displayed by a sample, can be corrected by the addition of reagents containing a high concentration of phospholipids but not of normal plasma. The most used assays by the experts in the field are the activated partial thromboplastin time (aPTT), the dilute Russel's viper venom time (dRWT), and the Kaolin clotting time (KCT). In spite of this adjustment of the criteria to the most recent evidences, the diagnosis of APS in a patient remains rather complex. This complexity not only is related to the clinical and diagnostic resemblance between APS and other autoimmune diseases, but also to the very nature of the disease.

The anti-phospholipid antibodies existing in patients with APS belong to a family that includes not only specific anti-cardiolipin antibodies, but also antibodies that recognise other anionic phospholipids and their complexes with plasmatic proteins, such as prothrombin, β2- glycoprotein I, protein S and C, and antibodies known as lupus anticoagulant, affecting the coagulation cascade in various ways during APS (Oosting J. D. et al., Blood (1993) 81:2618-2625; Matsuura E. et al, J. Exp. Med. (1994) 179:457-462; Galli M. et al, Blood (1995) 86:617-623; Nojima et al., Clin. Chem. (2005) 51:545-552). The laboratory criteria upon which the APS diagnosis is now based are^: a) presence of medium or high levels of antibodies recognising cardiolipin complexed with the cofactor β2-glycoprotein I, or b) presence of antibodies recognising β2-glycoprotein I alone, or c) positivity for the Lupus Anticoagulant (LA) factor. However, the significance, in diagnostics as well, of the other antibodies that very likely play a role in the clinical manifestations of the disease is still unclear (Balasch J. et al., Hum. Reprod. (1999) 14:1956-1959; Bertolaccini M.L. et al, Br. J. Rheumatol. (1998) 37:1229-1232). Moreover there are cases where the patients are defined as false negative or "seronegative", for they present secondary clinical symptoms of APS (such as migraine, marked livedo reticularis, thrombocytopenia, consecutive miscarriages, heart valve problems) and are consistently negative for antibodies recognising cardiolipin complexed with β2- glycoprotein I and for the LA coagulation assays. Therefore these patients cannot be diagnosed with APS according to the consensus criteria above outlined (Hughes G.R.V. e Khamashta M. A., Ann. Rheum. Dis. (2003) 62:1127; McCarty G.A., APS Foundation of America Newsletter, Vol. 2, Summer 2006; Lockshin M.D. et al., Arthritis Rheum. (2000) 43:440-443). For these patients the diagnosis remains therefore pending, and without a precautionary anticoagulant treatment there is a serious danger of undergoing even severe thrombotic episodes.

In addition to these patients , there are all the other cases where the levels of aCL antibodies are low, and therefore not in accordance with the consensus criteria for the diagnosis of APS. In this regard, it is worth pointing out that it has not yet been attained a true standardization of the results obtained using diagnostic kits from different manufacturers, for the detection of antibodies recognising cardiolipin complexed with β2- glycoprotein I (Audrain M.A.P. et al., Rheumatology (2004) 43:181-185). Consequently, detection of the presence or the absence of low levels of aCL antibodies may depend upon the manufacturer of the kit, or even upon the production lot. An issue, then, concerning the diagnosis of APS is to improve the performance of the kits used for detection of antibodies against cardiolipin and β2-glycoprotein I. The presence of serum antibodies against a new complex between cardiolipin and a protein (different from β2-glycoprotein I) has been identified. This can contribute to the diagnosis of "seronegative" patients, not identifiable with the current diagnostic tools, and for which today there is no clear indication for them to start a precautionary anticoagulant therapy.

Vimentin is a protein whose molecular weight is 56 kDa and is a member of the family of the intermediate filament proteins, which are typical of eukaryotic cells and contribute to the formation of the cytoskeleton, together with microtubules and actin microfilaments.

Vimentin appears as a monomer, like other proteins of intermediate filaments class, it has a central alpha-helix domain, an amino-terminal head, and a carboxyterminal tail with a non-alpha-helical structure (Fuchs

E. and Weber K, Annu. Rev Biochem (1994) 63: 345-82.).

Two monomers can be wound together to form a "coiled-coil" dimer, as a result of the hydrophobic interactions between alpha helices.

The vimentin filaments interact with elements of the nucleus, endoplasmic reticulum and mitochondria, playing an important role in supporting and anchoring various organelles in the cytoplasm. In general, vimentin is considered as a component of the cytoskeleton responsible for maintaining cellular integrity (Katsumoto T., Mitsushima A., Kurimura T.,

Biol Cell (1990) 68:139-46; Goldman R. D., Khuon S., Chou Y., Opal P., Steinert P. J Cell Biol (1996) 134:971-83).

The presence of antrvimentin antibodies in patients with SLE was confirmed by previous studies (Alcover A. et al., Arthritis Rheum. (1984) 27:922-928; Alcover A. et al, Clin. Exp. Immunol. (1985) 61:24-30; Sanchez A. et al., J. Rheumatol. (1990) 17:205-209). However, in the complex diagnosis of this disease, the importance of the presence of these antibodies has never been confirmed, and it still does not fall within the consensus criteria for its diagnosis (Bertsias G. et ah, Ann. Rheum. Dis. (2008) 67:195- 205).

Blaschek et al. (Ann. Rheum. Dis. (1988) 47:708-716) studied the presence of anti-vimentin antibodies in patients with SLE, and the relationship between their levels and those of antibodies against cardiolipin and ssDNA, making a hypothesis for a cross-reactivity between epitopes existing on the molecules of those antigens in patients with SLE.

Recently a complex between negatively charged phospholipids and the protein vimentin has been found in the blood of patients presenting with a probable diagnosis of "seronegative" APS.

Summary of the Invention

Subject of the present invention is an analytical method that makes it possible to detect the presence of antibodies against the complex between a negatively charged phospholipid, in particular cardiolipin, and the protein vimentin. Such antibodies have been found in a larger portion of patients compared to the current assays that detect the presence of antibodies against the cardiolipin / β2-glycoprotein I complex.

The three-dimensional structure of vimentin is characterised by the presence of ionic interactions between positive and negative charges of amino acids, resulting in the formation of interchain and intrachain bonds

(Fuchs E. and Weber K, Annu. Rev Biochem (1994) 63: _pp. 345-82.). It is believed that the formation of the cardiolipin-vimentin complex can be ascribed to electrostatic interactions between the positively charged amino acids of vimentin and the negative charges of the anionic phospholipid cardiolipin. As a consequence of this interaction, new epitopes are exposed, which makes the complex much more antigenic than vimentin or cardiolipin alone, and detectable by the common immunochemistry assay methods

A still further subject of the present invention is a kit for detection of antibodies against a negatively charged phospholipid, particularly cardiolipin, complexed with the protein vimentin as a cofactor, in patients with coagulation disorders. The kit of the invention is particularly useful in a diagnosis scenario where patients with clinical symptoms of APS are negative for LA and antibodies against cardiolipin and β2-glycoprotein I.

An even further subject of the present invention is the use of the complex between a negatively charged phospholipid, particularly cardiolipin, and the protein vimentin in order to produce compounds in the pharmaceutical field, especially for diagnostic use.

More subjects will become clear from the detailed description and the attached claims.

Brief Description of the Drawings Figure l: Identification of endothelial protein vimentin as a cofactor in patients that are considered as false negative ("seronegative") for APS. (A) an extract of the proteins from endothelial cell membrane of the endothelial immortalized EAhy926 cell line (Edgell CJS et al. (1983) Proc Natl Acad Sci USA. 80: 3734-3737) was separated by two-dimensional gel electrophoresis and transferred onto a nitrocellulose membrane. The filter was incubated with sera from two "seronegative" patients, and revealed two "spots" of molecular weight of about 54 and 57 kDa. MALDI-TOF MS analysis of the two "spots" gave evidence of spectra corresponding to two isoforms of vimentin (B) and (C). Figure 2^'- Evidence for vimenin/cardiolipin complexes.

Vimentin/cardiolipin binding was revealed by coimmunoprecipitation. Vimentin/cardiolipin complexes were incubated with 10 μg of goat polyclonal anti-vimentin per mg of protein and rocked for 2 h at 4°C. At the end of the incubation, protein A-sepharose was added. A portion of the immunoprecipitate was subjected to phospholipid extraction and separated by high-performance thin layer chromatography (HPTLC) in a single dimension by using a solvent system of chloroform/methanol/acetic acid/water (1OO^;75^:7^:4, v/v/v/v). Phospholipids were stained by exposure to iodide vapors and also immunostained with the purified human aCL IgG. Cardiolipin was detected by HPTLC (Fig. 2A). The identity of vimentin in the immunoprecipitate was verified by Western blot (Fig. 2B). Next, we confirmed the antigenicity of vimentin/cardiolipin complex by ELISA (Fig. 2C). The reactivity of serum antibodies to vimentin/cardiolipin complex was completely prevented by previous absorption with the complex.

Figure 3: Cardiolipin dose-dependent binding curve of vimentin. Scalar doses of vimentin (R&D Systems), from 10 μg/ml to 1.25 μg/ml, in buffered saline solution (PBS) at pH 7.4, were incubated with cardiolipin at a concentration of 50 μg/ml in methanol (Sigma Chem. Co.), overnight at 4°C in polystyrene multiwell plates. After three washes with PBS containing 0.1% Tween 20, the wells were blocked for 2 hours at room temperature with 100 μl per well of blocking buffer, PBS containing 3% bovine serum albumin (Sigma Chem. Co.). After three washes with PBS containing 0.1% Tween 20 (PBS-Tween) the wells were incubated with goat polyclonal antibodies against human vimentin (R&D Systems) diluted 1^200 in blocking buffer. After three additional washes with PBS-Tween, the plates were incubated for 1 hour at room temperature, with peroxidase- conjugated secondary IgG antibodies, diluted 1^1000 in blocking buffer. The plates were washed three times as above. The binding of peroxidase was revealed by adding 100 μl per well of a solution of Ophenylenediamine dichloride, and incubating for 10 minutes at room temperature. The reaction was stopped with IN H2SO4, which was added at a concentration of 50 μl per well for 5 minutes. The absorbance was read at 492 nm using an ELISA plate reader (BioRad).

Figure 4- Detection by ELISA of antibodies against the cardiolipin / vimentin complex according to the method of the invention. The antibodies, of both IgG (Fig. 3A) and IgM (Fig. 3B) classes, were present in patients with APS, "seronegative" APS, and SLE, at a significantly higher rate, compared to healthy control subjects (p < O.OOOl). Particularly, the rate of IgG antibodies in patients with APS was significantly higher than that in patients with SLE and with rheumatoid arthritis (p < 0.0001).

Figure 5: Effect on endothelial cells of antibodies against the cardiolipin / vimentin complex. In order to verify the functional effectiveness of the antibodies, immortalized endothelial cells from cell line EAhy926 were incubated with purified IgG fractions from sera of "seronegative" patients. In (A) it is shown the phosphorylation of cellular protein IRAK- 1 (Interleukin-1 Receptor-Associated Kinase 1), in (B) it is shown the phosphorylation and translocation to the nucleus of the transcription factor NF-KB, in (C) it is shown the expression of VCAM on the cell membrane, in (D) it is shown an increased release of procoagulant factor known as Tissue Factor (TF).

Detailed Description of the Preferred Embodiment The present invention is based on the discovery that the protein vimentin can be complexed with a negatively charged phospholipid, thereby acting as a cofactor for its binding to specific antibodies that are possibly present in biological samples, typically blood derivatives, such as plasma and serum.

The identification of vimentin as a constituent of a complex with negatively charged phospholipid, particularly cardiolipin, and as a cofactor in the determination of the presence of antibodies against that phospholipid, was performed as described below in Example 1. Cardiolipin, phosphatidylserine, phosphatidylinositol, phosphatidic acid and lyso(bis) phosphatidic acid (LBPA), among others, are the preferred phospholipids. Particularly cardiolipin. In the present invention the terms "complex / complexes" mean the binding between two macromolecules, resulting, in the specific case, from an electrostatic interaction between a negatively charged phospholipid and a protein that in its primary structure exhibits amino acids with a positive charge. The term "negatively charged phospholipid" refers to an anionic phospholipid that bears a negative electric charge. In particular, cardiolipin is a diphosphatidylglycerol with two phosphate groups (PO₄ ⁼) in the structural formula, resulting in a double negative charge.

Moreover, in the present invention the term "cofactor" refers to a molecule (in this case a protein), which ,when bound to a phospholipid (in this case cardiolipin), increases its antigenicity, that is the ability to be recognised by the immune system cells. Example 5 and Figure 5 describe some of the effects produced in vitro by antibodies against the cardiolipin / vimentin complex, on endothelial cells. It was also discovered by the inventors that the cardiolipin / vimentin complex is recognised by antibodies, in the serum of patients with APS, in a larger population compared to the tests for determination of antibodies against the cardiolipin / β2-glycoprotein I complex. This way it is possible to contribute to the diagnosis of APS, also in a significant proportion of patients who are "seronegative" for antibodies against phospholipids and anti-β2 glycoprotein I. In the present case, and as previously observed by others in the case of the complex between β2- glycoprotein I and cardiolipin, only 26% of APS patients had serum IgG specific for vimentin not complexed with phospholipid, while 23% had specific IgM, with values of optical absorption significantly lower than those associated with the aCL / vimentin complex (p < 0.001). This indicates a strong correlation between the formation of the complex phospholipid / vimentin and the specific antibodies in vitro.

The method of the invention relates to the detection of antibodies against the phospholipid / vimentin complex through one of the methods commonly in use, such as immunoenzymatic assays (e.g. the ELISA technique), radioimmunoassays (RIA), immunofluorescence methods, as well as methods that use chemilumine scent substrates (Rosner MH et al. Environ Health Perspect, 1991, 94:131-134), or whatsoever immunochemistry method of testing.

In particular, the method of the invention allows to determine the presence and concentration of antibodies against the cardiolipin / vimentin complex.

More specifically, the method of the invention allows to support the diagnosis of APS in patients with clinical manifestations of the disease, but negative results of the laboratory tests are needed to confirm the diagnosis ("seronegative" patients), as previously discussed.

The method provided by the present invention comprises the following stages^ a) preparation of the phospholipid / vimentin complex; b) incubation of biological samples with that complex; c) determination of the concentration of the specific antibodies, possibly present in the sample, that recognised the phospholipid / vimentin complex. Preferably the sample is tested in triplicate.

The complex mentioned in stage a) is prepared on a solid substrate (preferably a 96-well ELISA plate) by incubating 50-150 μl of a solution containing 40-60 μg/ml of phospholipid in organic solvent, preferably methanol, with 50-150 μl of a 3-6 μg/ml solution of vimentin in PBS, for 8-12 hours at a temperature between 3 and 8°C. Ideally, during the execution of stage a) 100 μl of 5 μg/ml vimentin in PBS are added to 100 μl of 50 g/ml phospholipid (typically cardiolipin) in methanol, for 12 hours at 4°C.

Incubation of stage b) is performed with 10-150 μl of sample at room temperature for 60-120 minutes. Typically, 100 μl of sample (serum or plasma), diluted l^:100, are incubated at room temperature for 1 hour.

The determination of the antibody levels in the sample test is performed by adding to the wells 10-150 μl of secondary polyclonal antibodies against human IgG or IgM, followed by incubation for 60-120 minutes at room temperature. The secondary antibodies can be conjugated with a radioactive substance or an enzyme, or a chemiluminescent tracer, or a fluorescent tracer, or a dye , in order to be detected and allow a reading of the final result. The amount of antibodies present in the samples is determined using a standard curve generated with scalar quantities of antibodies to the phospholipid / vimentin complex. The sample derived from blood can be either serum or plasma. Other body fluids can also be analyzed (urine, synovial fluid, cerebrospinal fluid and saliva).

Any negatively charged phospholipid can be used in the method of the invention (e.g. cardiolipin, phosphatidylserine, phosphatidylinositol, phosphatidic acid, or lyso(bis) phosphatidic acid (LBPA), among others). In particular, the results showed reactivity of sera from patients with APS, also to the phosphatidylserine / vimentin complex. Specifically, the method allows the determination of antibodies against the complex formed by phospholipid cardiolipin and vimentin as a cofactor.

Typically, in the method of the invention the phospholipid / vimentin complex or the antibodies against this complex are immobilized on a suitable solid support, such as a chromatography column or a plastic surface like a dip stick or a microplate assay plate, or magnetic microp articles, or on other solid supports such as, polymeric latex ,or nitrocellulose, according to the methods known to the experts in the field (Rosner, MH et al. Environ Health Perspect, 1991, 94:131-134).

The material for the chromatography column can be, among others, a matrix of agarose beads, polyacrylamide, or dextran.

Among the plastic materials that can be used in the method of the invention, there are polymethyl methacrylate, polystyrene, polyethylene, polyethylene and polypropylene. Typically, any assay microplate is acceptable.

The solid surface can also be either an array or a gel (e.g. silica gel) in which the phospholipid and / or the cofactor, the complex or the antibody may be incorporated. Vimentin used in the method of the present invention can be of human or even murine origin, due to the high homology between the amino acid sequences in the two species (92%). Preferably it should be in pure form. More preferably the vimentin used in the method of the invention is of recombinant origin. Typically it is possible to use a standard ELISA technique where the antigen or antibody are labelled. The marking can be achieved for example with an enzyme or a radioisotope. Generally, the marker is an enzyme.

Another preferred embodiment of the present invention relates to a kit, for testing the presence of antibodies against the phospholipid / vimentin complex in the diagnosis of autoimmune diseases, including: a) protein vimentin; b) a negatively charged phospholipid, preferably chosen among cardiolipin, phosphatidylserine, phosphatidylinositol, phosphatidic acid, or lyso(bis) phosphatidic acid (LBPA) among others. Particularly, cardiolipin is preferred.

In a further and particularly preferred embodiment of the invention, the kit for testing the presence of antibodies against cardiolipin in the diagnosis of autoimmune diseases includes the phospholipid / vimentin complex immobilized on a solid support to provide a kit that is quick and easy to use ("rapid kit").

The kits, according to the invention, are typically packaged so to contain any additional components for the analysis, such as, for example, buffers, capture reagents, developing reagents, markers, reaction surfaces, control samples and instructions for use.

In the kits of the invention, the binding of antibodies to the complexes between phospholipids (typically cardiolipin) and vimentin can be determined using any means available in the field (and supplied along with the kit), such as, for example, a secondary antibody against human immunoglobulins in case the analysis is performed on human samples. The marker can be any of those used in the field, for example an enzyme or a radioisotope, or whatsoever marker used in the immunochemistry applications Typically, the marker is an enzyme.

The method of the invention allows to diagnose, with a single, practical and easy to perform assay, the presence of APS in a wider range of patients than other diagnostic methods currently in use. As a matter of fact, the response from this method, in conjunction with clinical data, is not only helpful to identify patients who would be diagnosed with APS if subjected to conventional tests for determining the presence of this autoimmune disease (especially tests for the presence of antibodies against cardiolipin complexed with β2-glycoprotein I, and antrβ2 glycoprotein I alone), but it also provides useful laboratory data in a substantial proportion of patients who, while presenting the clinical symptoms, cannot be diagnosed with APS because of the negativity of their laboratory tests, according to the current clinical consensus. The following examples are intended to further illustrate the present invention without limiting its scope.

Examples Example 1 Identification of vimentin as an endothelial protein cofactor in false - negative ("seronegative") patients for APS.

The proteins of the plasma membrane of EAHY926 endothelial cells, an immortalized cell line (Edgell CJS et al. (1983) Proc Natl Acad Sci USA. 80: 3734-3737), were purified using the commercial kit "Pierce Cell Surface Protein Isolation Kit" (Pierce, Rockford, IL). They were subsequently separated by two-dimensional poly aery lamide gel electrophoresis and transferred to a nitrocellulose membrane (Schleicher & Schuell). The membrane was then analyzed in the presence of sera from patients "seronegative" for APS, diluted 1^10O in PBS. Goat secondary antibodies against human IgG (BioRad), conjugated with peroxidase and diluted UlOOO in PBS, were used for detection, and the staining was developed with the substrate 3,3'-diaminobenzidine (Sigma Chem. Co.), which revealed the presence of two positive "spots" (Figure IA).

The gel stained with Coomassie blue, corresponding to the spots, was excised, bleached with 50% acetonitrile and digested with trypsin. Analysis with mass spectrometry was performed using MALDI micro MX (Micromass, Manchester, UK). The peaks obtained from the spectrum (Figure IB), corresponding to the digested peptides, were analyzed with Mascot using the database Swiss-Prot. The process led to the identification of the peptides, highlighted in the amino acid sequence of the protein vimentin (Figure 1C), as two isoforms of vimentin. Example 2 i) Binding of vimentin to cardiolipin.

Vimentin/cardiolipin complexes were re-suspended in a buffer containing 20 mM Tris-HCl, pH 7.5, 0.15 M NaCl, 1 mM EDTA, 0.02 % NaN3, 10 mM NaF. The mixtures were incubated with 10 μg of goat polyclonal anti-vimentin (R&D System, Minneapolis, MN, USA) per mg of protein and rocked for 2 h at 4°C. At the end of the incubation, protein A- sepharose (Sigma Chem Co) was added and the mixture was rocked at 4°C for an additional 1 h. As a negative control, immunoprecipitation was performed with an irrelevant goat IgG (Sigma Chem Co). A major portion of the immunoprecipitate was subjected to phospholipid extraction and separated by high-performance thin layer chromatography (HPTLC) in a single dimension by using a solvent system of chloroform/methanol/acetic acid/water (100^75:7:4, v/v/v/v). Phospholipids were stained by exposure to iodide vapors and also immunostained with the purified human aCL IgG. Vimentin/cardiolipin binding was revealed by coimmunoprecipitation. Cardiolipin was detected by HPTLC (Fig. 2A). The identity of vimentin in the immunoprecipitate was verified by Western blot (Fig. 2B). Next, we confirmed the antigenicity of vimentin/cardiolipin complex by ELISA (Fig. 2C). The reactivity of serum antibodies to vimentin/cardiolipin complex was completely prevented by previous absorption with the complex. ii) Dose-dependent binding of vimentin to cardiolipin.

Scalar doses of vimentin in buffered saline solution (PBS) at pH 7.4, from 10 μg/ml to 1.25 μg/ml, were incubated with 50 g/ml cardiolipin in methanol (Sigma Chem. Co.) overnight at 4°C, in 96-well polystyrene plates. After three washes with PBS containing 0.1% Tween 20 (PBS-Tween), the wells were blocked for 2 hours at room temperature with 100 μl per well of blocking buffer (PBS containing 3% bovine serum albumin, Sigma Chem. Co.). After three washes with PBS-Tween, the wells were incubated with goat polyclonal antibodies against vimentin (R&D Systems). After three additional washes with PBS-Tween, the plates were incubated for 1 hour at room temperature with polyclonal secondary IgG antibodies conjugated to the enzyme peroxidase (1:1000 in blocking buffer, Sigma Chem. Co.). The plates were then washed three times with PBS-Tween as above. The binding of peroxidase was revealed by adding 100 μl per well of a solution of O- phenylenediamine dichloride, and incubating for 10 minutes at room temperature. The reaction was stopped with IN H2SO4, which was added at a concentration of 50 μl per well for 5 minutes. The absorbance was read at 492 nm using an ELISA plate reader (BioRad).

The result (shown in Figure 3) confirms a dose-dependent binding between cardiolipin and vimentin. Example 3 Determination of the presence of antibodies against the cardiolipin / vimentin complex using ELISA.

Polystyrene 96-well plates were incubated overnight at 4°C with 100 μl per well of 50 g/ml cardiolipin (Sigma Chem. Co.) in methanol, and subsequently with 100 μl per well of recombinant human vimentin (5 μg/ml, R&D Systems) in PBS. The vimentin was let to adhere to the plate, overnight at 4°C.

After three washes with PBS containing 0.1% Tween 20 (PBS'Tween), the wells were blocked for 2 hours at room temperature with 100 μl per well of PBS containing 3% bovine serum albumin (blocking buffer, Sigma Chem. Co.).

After three washes with PBS-Tween the wells were incubated with

100 μl per well of serum from patients, diluted 1^100 in blocking buffer, for 1 hour at room temperature. Samples were tested in triplicate. A goat antibody against vimentin (R&D Systems), diluted 1^200 in blocking buffer was used as positive control.

After three additional washes in PBS-Tween, the plates were incubated for 1 hour at room temperature with secondary goat polyclonal antibodies, against human IgG or IgM (Sigma Chem. Co.), conjugated with horseradish peroxidase and diluted TlOOO in PBS + 1% BSA. The plates were again washed three times with PBS-Tween and the peroxidase reaction was revealed by adding 100 μl per well of a solution of

Ophenylenediamine dichloride, and incubating for 10 minutes at room temperature. The reaction was stopped with IN H2SO4, which was added at a concentration of 50 μl per well for 5 minutes. The absorbance was read at 492 nm using a BioRad plate reader.

The absorbance of the negative control wells (aspecific binding) was subtracted from each reading of the samples, and the threshold for positivity was determined as the mean + three times the standard deviation (SD) of the values obtained in samples from healthy patients.

The standard curve was obtained using, at scalar dilutions, the reference serum taken as positive control. The curve is standardized in arbitrary units: along the abscissa axis are the adopted scalar dilutions (see Figure 2). Example 4

Detection of antibodies against the cardiolipin / vimentin complex in patients with APS. In Table 2 are shown data obtained from a cohort of patients composed as follows^

29 sera from patients with clinical manifestations of APS (arterial and venous thrombosis and / or recurrent miscarriages) but negative for all the traditional tests for the determination of autoantibodies ("seronegative");

40 sera from patients with APS, (including 30 patients with primary APS and 10 with secondary APS). The patients were positive for aCL antibodies with ELISA and, possibly, for anti-β2-glycoprotein I and / or for the LA assay. Moreover, they presented with clinical symptoms stated according to the Sapporo criteria (Wilson WA et al., Arthritis Rheum. (1999) 42÷ 1309- 1311). The disease may manifest as isolated disorder (primary APS) or during a systemic autoimmune disease, such as Systemic Lupus Erythematosus (secondary APS);

30 sera from patients with SLE; ■ 30 sera from patients with rheumatoid arthritis;

32 sera from healthy subjects.

The threshold for positivity was calculated using the mean of the results obtained with sera from healthy patients + three times the standard deviation (SD) of the obtained values. Table 2. Prevalence of antibodies to vimentin/cardiolipin complex.

Data in Table 2 show that the sera of patients classified as

"seronegative" for antibodies against phospholipids (i.e., negative for

5 antibodies against the cardiolipin / β2- glycoprotein I complex and the LA, but with a case history of APS), analyzed by the method of the invention, showed a rate of approximately 50% of positivity for antibodies against the cardiolipin / vimentin complex. Moreover, as shown in Table 2 as well, almost all samples from patients with APS also displayed IgG and IgM

10 antibodies against the complex between cardiolipin and vimentin.

The sera were analyzed for their reactivity with vimentin in the absence of phospholipids. Positivity for vimentin in the absence of phospholipid was found in 26% of APS patients for the IgG isotype, and in

23% of APS patients for the IgM isotype. No other types of vimentin than

15 the human recombinant were tested (Table 3).

Table 3. Prevalence of antibodies to vimentin/cardiolipin complex, cardiolipin or vimentin

The results obtained and illustrated in Figure 4A and 4B show that

20 the antibodies are preferentially directed against the phospholipid-protein complex. Antibodies, both of IgG and IgM classes, were present in patients with APS and with "seronegative" APS, with a significantly greater rate than the healthy control subjects (p < 0.0001). The statistical analysis of data, using the nonparametric test of Mann- Whitney, showed the following results^ - IgG: each group of patients vs. healthy controls p < 0.0001

IgM: each group of patients vs. healthy controls p < 0.0001. Example 5

Determination of the presence of antibodies against the phosphatidylserine / vimentin complex using ELISA. With similar methods, it was verified that other negatively charged phospholipids can be complexed with vimentin. In particular, the results showed reactivity of sera from patients with APS to the phosphatidylserine / vimentin complex, demonstrating that this complex can represent a target for antibodies during APS. 96-well polystyrene plates were incubated overnight at 4°C with 100 μl per well of 50 g/ml phosphatidylserine (Sigma Chem. Co.) in methanol and then with 100 μl per well of recombinant human vimentin (5 mg/ml, R&D Systems) in PBS. After three washes with PBS containing 0.1% Tween 20 (PBS-Tween), the wells were blocked for 2 hours at room temperature with 100 μl per well of PBS containing 3% bovine serum albumin (blocking buffer, Sigma Chem. Co.).

After three washes with PBS-Tween the wells were incubated with 100 μl per well of serum from the patients, diluted 1^100 in blocking buffer, for 1 hour at room temperature. Samples were tested in triplicate. A goat antibody against vimentin, diluted 1^200 in blocking buffer, was used as positive control (R&D Systems). After three additional washes in PBS- Tween, the plates were incubated for 1 hour at room temperature with secondary goat polyclonal antibodies against human IgG or IgM (Sigma Chem. Co.), conjugated with horseradish peroxidase and diluted 1^1000 in PBS + 1% BSA.

The plates were again washed three times with PBS-Tween and the peroxidase reaction was revealed by adding 100 μl per well of a solution of £>phenylenediamine dichloride, and incubating for 10 minutes at room temperature. The reaction was stopped with IN H2SO4, which was added at a concentration of 50 μl per well for 5 minutes. The absorbance was read at

492 nm using a BioRad plate reader. The absorbance of the negative control wells (aspecific binding) was subtracted from each reading of the samples, and the threshold for positivity was determined as the mean + three times the standard deviation (SD) of the values obtained.

Analyzing the same sera from patients with APS tested in the example 4 we observed that 88% of patients with APS and 45% of

"seronegative" APS patients presented serum IgG to the phosphatidylserine

/ vimentin complex.

Example 6

Stimulation on endothelial cells by IgG purified antibodies, from sera of "seronegative" patients.

Endothelial cells of type Eahy926, an immortalized adherent cell line, were stimulated with purified IgG fractions from sera of "seronegative" patients (200 μg/ml in PBS) which were pre-incubated with cardiolipin micelles (1 mg/ml) and vimentin (200 μg/ml) in PBS. The positive control consisted of the same cells stimulated with LPS (100 ng/ml in PBS) or TNF- α (20 ng/ml in PBS, Sigma Chem. Co.). The negative control consisted of

Eahy926 cells incubated with purified IgG fractions from healthy donors.

The cells were incubated at 37°C in a humidified 5% CO2 atmosphere.

Incubation times were not the same in different experiments, as later indicated.

Example 6 A: Phosphorylation of cellular protein IRAK-I (Interleukin-1

Receptor- Associated Kinase 1).

To verify the phosphorylation of IRAK'l, the Eahy926 cells were stimulated as described above for 45 minutes. After treatment, cells were washed twice with PBS and detached from the flasks with trypsin, according to practice known to the experts of the field. The detached cells were washed by suspending them first in D-MEM culture medium (Sigma Chem. Co.), then in PBS, and centrifuged at 300 x g for 6 minutes at 4°C after each wash. The final pellet was suspended for 30 minutes at 4°C in a lysing solution consisting of 1% Nonidet P-40, Tris-HCl (100 mM pH 7.5), NaCl (150 mM), EDTA (5mM), PMSF (l mM), leupeptin (10 μg/ ml), Na₃VO₄ (ImM) and aprotinin (75 U, Sigma Chem. Co.). The cell lysate thus obtained was then centrifuged at 15000 x g for 30 minutes at 4°C, to remove cell debris and nuclei. The protein content of the recovered supernatant was determined by the Bradford assay (Bio-Rad). Subsequently, the samples were diluted, if necessary, in order for them to contain the same concentration of protein before being subjected to electrophoresis and then Western Blot. The electrophoretic separation was performed on 7.5% polyacrylamide gel with a mini Protean II Dual Slab Cell system (Bio Ra d). The determination of molecular weights was performed according to the electrophoretic mobility of standard proteins included in the gel (Bio-Rad). After the proteins were transferred on nitrocellulose, the filter was incubated for 1 hour at room temperature in TBS-Tween buffer (10 mM Tris-HCl pH 8.0, 150 mM NaCl, 0.05% Tween 20), with albumin 3% in order to block the aspecific binding sites of nitrocellulose. After this phase, the filter was incubated, for 12 hours at 4°C, with 2 μg/ml of rabbit polyclonal antibody against phospho-IRAK-1 (Cell Signaling Technology) in TBS- Tween. After three washes in TBS-Tween, the nitrocellulose was incubated for 1 hour at room temperature with goat polyclonal secondary antibodies, conjugated to the enzyme peroxidase (Sigma Chem. Co.) diluted 1^5000 in TBS-Tween. After three additional washes, performed as above, the reactivity of the antibodies was revealed by chemiluminescence reaction, using an ECL system for Western Blot (Amersham). The control of the protein content of the samples was performed after incubating the filter with monoclonal anti-actin (Sigma Chem. Co.), lμg/ml in TBS-Tween. This reaction was developed as above. Western blot analysis of cell lysates showed that IgG fractions from

SN-APS, as well as LPS, induced IRAK-I phosphorylation, as revealed by anti-phospho-IRAKl antibodies reactivity (Fig. 5A). On the contrary, cells stimulated with normal human IgG did not show anti-phospho-IRAK- 1 reactivity.

Example 6 B: Phosphorylation and translocation to the nucleus of the transcription factor NF-κB. In order to analyze the activation of nuclear transcription factor NF-

KB, phosphorylation and translocation of p65 subunit to the nucleus were evaluated.

Eahy926 cells were stimulated as above. After treatment, the cells were washed twice with PBS, detached from the flasks with trypsin, washed with culture medium, then with PBS, and centrifuged as above.

For the preparation of the nuclear extracts, the cell pellets obtained after the washes were first resuspended in a buffer consisting of HEPES (20 mM pH 7.9), KCl (20 mM), MgCl₂ (3.0 mM), Na₃VO₄ (ImM), leupeptin (10 μg/ml), PMSF (l mM), E64 (10 μM), pepstatin (0.5 μg/ml), DTT (5 mM), 1% Nonidet P-40 (Sigma Chem. Co.), then vortexed and left on ice for 30 minutes. Subsequently, the extracts thus obtained were centrifuged for 30 minutes at 10000 x g at 4°C. Pellets (which correspond to the nuclei) were resuspended in buffer B consisting of: HEPES (40 mM pH 7.9), NaCl (0.84 M), EDTA (0.4 mM), 50% glycerol, Na₃VO₄ (ImM), leupeptin (10 μg/ml), PMSF (1 mM), E64 (10 μM), pepstatin (0.5 μg/ml), DTT (5 mM), 1% Nonidet P-40 (Sigma Chem. Co.), then vortexed and left on ice for 1 hour. The nuclear extracts thus obtained were then centrifuged at 10000 x g for 1 hour at 4°C. The protein content of the recovered supernatant was determined by the Bradford assay (Bio-Rad). The samples were diluted, if necessary, in order for them to contain the same concentration of protein before being subjected to electrophoresis and then Western Blot, as previously described for the phosphorylation of IRAK-I. The nitrocellulose filter was incubated with rabbit polyclonal antibodies against phosphorylated p65 (2 μg/ml in TBS-Tween, Cell Signaling Technology). After three washes in TBS-Tween, the same filter was incubated with goat secondary antibodies, conjugated to peroxidase (Sigma Chem. Co.) diluted 1:5000 in TBS-Tween. The reaction was developed as above. The control of purity of nuclear extracts was performed by incubating the same filter of nitrocellulose with monoclonal anti-Histone 1 (Upstate Biotechnology) diluted to lμg/ml in TBS-Tween. The reaction was then developed as above.

Western blot analysis of nuclear extracts revealed that IgG fractions from SN-APS, as well as LPS, induced NF-kB phosphorylation, as revealed by anti-phospho-NF-kB p65 Ser antibodies reactivity (Fig. 5B). On the contrary, cells stimulated with normal human IgG did not shown anti- phospho-NF-kB p65 Ser reactivity. Example 6 C: Expression of VCAM on the EAhy926 cell membrane. Indirect immunofluorescence technique was used to analyze the expression of VCAM protein on the membrane of endothelial EAhy926 cells.

Briefly, for each sample the cells were stimulated for 20 hours as previously described. After treatment they were washed twice with PBS and detached from the flasks with trypsin, washed again and centrifuged as above. The obtained cell pellets were fixed for 20 minutes at 4°C with 4% formaldehyde in PBS. After this step the cells were centrifuged and washed three times in PBS, as described above, and then incubated for 1 hour at 4°C with a monoclonal antibody anti-VCAM (GeneTex, Inc) diluted to 1 μg/ml in PBS. After incubation, three more washes in PBS were performed, followed by centrifugation as above, and then the cells were incubated again for 30 min at 4°C with goat secondary antibodies conjugated to fluorescein isothiocyanate (FITC, a fluorochrome that emits green light, Sigma Chem. Co.) diluted 1^50 in PBS. After this period of incubation the cells were then washed with PBS as above; and resuspended in PBS. The fluorescence intensity was analyzed with a Becton Dickinson flow cytometer.

Cytofluorimetric analysis revealed that IgG fractions from SN-APS, as well as LPS, induced VCAM expression on the cell surface (Fig. 5C). On the contrary, cells stimulated with normal human IgG did not shown VCAM expression. Example 6 D^ Release of procoagulant factor known as Tissue Factor (TF) from EAhv926 cells.

To assess the increase in the release of procoagulant factor known as Tissue Factor (TF) from EAhy926 cells, the cells were first stimulated for 4 hours as previously described. After treatment the cells were washed twice with PBS, detached from the flasks with trypsin, washed again and centrifuged as above. After washing the supernatants were recovered and tested for the presence of TF using a commercial ELISA kit (American Diagnostica) as recommended by the manufacturer.

TF release by cells stimulated with IgG fractions from SN-APS or with LPS was significantly increased, as compared to untreated endothelial cells, as well as cells stimulated with control human IgG (Fig. 5D).

Claims

1. The complex between a negatively charged phospholipid and protein vimentin for use in medical field

2. An analytical method for determining the presence and the concentration of antibodies against the negatively charged phospholipid / vimentin complex in biological fluids, in which, at least one epitope of such complex is detected, said method of analysis, applicable in particular in the diagnosis of patients with Antiphospholipid Syndrome (APS), comprising the following stages: a) preparation of a negatively charged phospholipid / vimentin complex; b) coating of the negatively charged phospholipid / vimentin complex on an appropriate solid surface such as , but not only, elisa microplates, magnetic particles, latex particles, polymeric dyed particles, plastic tubes, cellulose derivatives; c) incubation of biological samples with the charged phospholipid / vimentin complex, in a matrix reaction, with binding of the specific antibodies present in the biological sample to the complex charged phospholipid/vimentin complex; d) detection of the of the antibodies, present in the sample, and bound to the charged phospholipid / vimentin complex.The detection can be made by reaction with a secondary antibody , marked with an appropriate label, and directed against the specific antibodies bound to the phospholipid/vimentin complex.

3. The method according to claim 2 characterised by the fact that the protein vimentin used in stage a) is in pure form, preferably of human or murine origin, more preferably recombinant.

4. Kit suitable for the execution of the method according to claims 2-3, comprising: a) protein vimentin; b) at least one negatively charged phospholipid; and, optionally! c) reaction surface and solid supports, buffers, capture reagents, developing reagents, markers, control samples and instructions for use.

5. Use according to claim 1 for identification of patients with Antiphospholipid Syndrome (APS).

6. Pharmaceutical compositions including the negatively charged phospholipid- vimentin complex with adjuvants and / or pharmacologically acceptable carrier agents.