EP2673298A1 - Anti-gb3 antibodies useful in treating disorders associated with angiogenesis - Google Patents

Abstract

The present invention is situated within the field of novel anti-cancer therapies, specifically within the field of anti-angiogenic molecules. It relates in particular to anti-GB3 antibodies having specific CDR sequences, as well as to the use of anti-GB3 antibodies not coupled to a therapeutic molecule in treating disorders associated with angiogenesis, such as solid tumors.

Description

 Anti-Gb3 Antibodies useful in the treatment of angiogenesis-associated diseases

FIELD OF THE INVENTION

The present invention is in the field of new anticancer therapies, more specifically in the field of anti-angiogenic molecules. In particular, it relates to anti-Gb3 antibodies having specific CDR sequences, as well as to the use of anti-Gb3 antibodies that are not coupled to a therapeutic molecule in the treatment of diseases associated with angiogenesis.

PRIOR ART

Cancer cells, because of their genetic instability, can acquire resistance to cancer treatments, which is at the origin of therapeutic failures or cancer recurrences.

 There is therefore a need for new anti-cancer agents that are less sensitive to the genetic variability of tumor cells.

 Unlike the tumor cells themselves, the endothelial cells of the tumor's blood vessels are genetically stable. In addition, they are essential for neovasculansation without which the tumor can not continue to grow, for lack of nutrients. Therefore, a strategy to inhibit angiogenesis, i.e., proliferation of tumor endothelial cells, would not be sensitive to the genetic variability of the tumor, with endothelial cells being genetically stable.

 Targeting endothelial cells also has other benefits

In the context of a solid tumor, the tumor cells are difficult to reach. Indeed, with traditional anti-cancer therapies, the penetration of the treating agents into the tumor tissues is diminished because of the high interstitial pressure in most tumors. This is not the case in vessels that are a more easily accessible target. Because of this accessibility, endothelial cell targeting allows efficient distribution and rapid accumulation in the tumor vessels. Therefore, targeting the vasculature is a strategy that can be applied to the majority of tumor types.

 In an adult organism, the vast majority of endothelial cells are in a quiescent state. On the other hand, the endothelium of the tumor networks is composed of endothelial cells presenting a proliferating "angiogenic" phenotype that it is theoretically possible to target. In fact, the vessels that irrigate the tumors have structural and functional characteristics different from those of the normal vessels. It is thus expected that this strategy is selective and causes few side effects on so-called "physiological" angiogenesis.

 Finally, targeting vascularization and inhibiting the formation of metastases are crucial aspects, since invasion is one of the most serious aspects of the disease.

Thus, targeting proliferating endothelial cells has many advantages, both in terms of accessibility of targeted cells, the number of tumors that can be treated, reduction of side effects if only proliferating endothelial cells are targeted, and prevention of metastases.

Globotriasosylceramide or globotriaosylceramide (Gb3), also called CD77, P ^k , ceramide trihexoside (CTH), and Burkitt Lymphoma Antigen (BLA) is a glycosphingolipid in the form of Galactoseal → 4Galactose i → 4Glucose i → Cerami de, synthesized by the enzyme Lactosylceramide. 4-alpha-galactosyltransferase (A4GALT). It is expressed by the HUVEC endothelial cell line, more when this line is proliferating than when the cells are confluent and are no longer in the exponential phase of growth. It therefore constitutes a marker of the endothelial cells involved in angiogenesis (Heath-Engel et al., Obrig et al).

 The ceramide, the hydrophobic part of the molecule anchored in the outer leaflet of the plasma membrane, is formed of a fatty acid chain composed of 16 to 28 carbon atoms, linked by an amide linkage to a sphingoid base, generally sphingosine. . It can have variations in the length but also in the number of unsaturation of its chain of fatty acids. In particular, the main possible variations for the sphingoid base and the fatty acid chain are shown in Table 1 below.

Table 1. Main molecular species of fatty acids and sphingoid bases used in ceramide

 Fatty acids Sphingoid bases

 Palmitic acid (16: 0) Sphingosine (18: 1)

 Stearic acid (18: 0) Dihydrosphingosine (18: 0)

 Oleic acid (18: 1) C20-dihydrosphingosine (20: 0) Arachidonic acid (20: 0) C20-sphingosine (20: 1)

 Behenic acid (22: 0)

 Lignoceric acid (24: 0)

 Nervonic acid (24: 1)

Gb3 is a natural receptor of certain bacterial toxins such as Shiga toxin, produced by Shigella dysenteriae, and verotoxins produced by Escherichia coli. These toxins are composed of two subunits A and B, the subunit B being involved in the binding to Gb3, while the subunit A corresponds to the toxic part inhibiting the protein synthesis. The binding of these bacterial toxins to Gb3 leads to the transport of toxin from the plasma membrane to the endoplasmic reticulum via early endosomes and the Golgi apparatus. Subunit A produces its toxic effects in the endoplasmic reticulum.

 Shiga toxin and verotoxins have therefore been proposed to exert cytotoxic activity on endothelial cells, in particular on proliferating endothelial cells (Heath-Engel et al., Obrig et al., WO98 / 51326).

Verotoxins and Shiga toxin have also been proposed as a therapeutic agent in the treatment of tumors expressing Gb3 on their surface, the subunit B for targeting, while the A subunit plays a cytotoxic role. In particular, verotoxin 1 has been shown to be capable of inducing apoptosis of Burkitt lymphoma cells (Tétaud et al).

 However, these approaches are not really applicable in cancer therapy because of the very important side effects of toxic subunit A on the rest of the human body. Indeed, besides the vascular tissue, Gb3 is expressed in many other tissues, such as the tissue of the stomach, esophagus, prostate and kidney. Since Shiga toxin and verotoxins are small molecules (about 68 kilodaltons (kDa)), they can leave the bloodstream and enter healthy tissues, where they have the same cytotoxic properties for Gb3-expressing cells as for tumor cells. Because of their lack of specificity of action, these toxins lead to unacceptable side effects. Thus, for example, Bast et al show that two hours after administration to rabbits, verotoxin 1 is no longer present in the bloodstream and can be found in target organs expressing Gb3.

 Moreover, in addition to the associated side effects, the small molecular weight of these toxins and their exit from the bloodstream also reduces their duration of action on the cells of the blood vessels, thus also decreasing their effectiveness.

 It has also been proposed to use only the B subunit of these toxins, and especially Shiga toxin, to target tumor cells expressing Gb3, a cytotoxic therapeutic molecule being coupled to the B subunit (Janssen et al. Viel et al). However, provided that the therapeutic molecule coupled to the B subunit (only about 8 kDa per B subunit) is of low molecular weight, then this strategy poses the same problems of toxicity and duration of action on the cells. blood vessels than the previous one.

 There is therefore a need for new molecules capable of binding to Gb3 and inhibiting the angiogenic activity of endothelial cells of tumor blood vessels, which are less toxic and have a longer duration of action.

Monoclonal antibodies can be generated for many types of ligands, including glycosphingolipids, although some are immunogenic. Monoclonal antibodies directed against Gb3 were already used for labeling Gb3-expressing cells, in particular monoclonal antibody 38.13 (Bordron et al., Chark et al).

 However, it has been described that this monoclonal antibody 38.13 does not bind Gb3 in the same way as verotoxins (Chark et al). In addition, it has also been reported that an anti-Gb3 monoclonal antibody obtained from ascites 1A4, induced apoptosis of Burkitt's lymphoma cells by a mechanism different from that of verotoxin 1 (Tétaud et al). It was therefore unlikely that an anti-Gb3 monoclonal antibody could have the same anti-angiogenic effect as verotoxins.

 Moreover, W098 / 51326, which describes the use of Gb3-binding molecules as anti-angiogenic agents, provides as such agents verotoxins and anti-Gb3 antibodies coupled to a toxic molecule, such as toxins. The authors of this document therefore considered that an anti-Gb3 antibody alone would not have an anti-angiogenic effect.

 In addition, Bordron et al describes the apoptotic effect on endothelial cells of different anti-endothelial cell antibodies (undetermined specific antigenic specificities, probably mixtures of antibodies of different specificities) from patients suffering from autoimmune diseases, as well as than several monoclonal antibodies of known specificity, among which an anti-Gb3 monoclonal antibody (clone 38/13). The results presented in this document show that only 8% of the endothelial cells are in apoptosis state following their contact with the 38/13 anti-Gb3 monoclonal antibody, which percentage is lower than that obtained in the presence of a single medium. , without antibodies (14%). These results also suggest that anti-Gb3 monoclonal antibodies do not have the same effect on endothelial cells as Shiga toxin or verotoxins.

However, the inventors have surprisingly found that several anti-Gb3 monoclonal antibodies have anti-angiogenic activity on proliferating endothelial cells. These monoclonal antibodies can therefore be used in the treatment of diseases associated with angiogenesis, and in particular solid tumors, with all the previously mentioned advantages that this targeting comprises (less risk of resistance, applicable to any type of tumor, targets easily). reached, prevention of metastases). In addition, they have several advantages over Shiga toxin and verotoxins:

 Depending on their isotype, the antibodies have a molecular weight that varies between 150 (IgG) and 1000 (IgM) kDa, and therefore have a higher molecular weight than the Shiga toxin and verotoxins. This should allow them to stay longer in the bloodstream and therefore have a longer duration of action than these toxins on the endothelial cells of tumor neovessels. The isotypes IgG (in particular IgG1) and IgM are the most advantageous. Indeed, the IgM have the largest molecular weight, about 80% remaining in the bloodstream, and they have an in vivo half-life of about 10 days. Regarding IgG, 45% remain in the bloodstream, and their half-life is about 20 days (Nikolayenko et al). This is to be compared with the fact that the administration of verotoxin 1 to rabbits shows that only 2% of verotoxin 1 is still present in the bloodstream after two hours, the rest being found in tissues expressing Gb3 .

 These antibodies are not toxic in themselves, which should limit their side effects on other tissues expressing Gb3, the fact that they remain mainly in the bloodstream also limiting these side effects,

 Mouse antibodies obtained by the inventors can be humanized, so as to minimize any immune response of the recipient, which is not possible for bacterial toxins, which generate an immune response may reduce their effectiveness.

These antibodies therefore have in common to recognize Gb3 and to be useful in the treatment of diseases associated with angiogenesis. In addition, two of these antibodies (3E2 and 22F6) further have in common to recognize the upper band but not the lower band of the Gb3 doublet expressed by HMEC-1 cells, which appears to be particularly overexpressed when endothelial cells proliferate. , compared to quiescent cells. This is likely to give them increased specificity for endothelial cells during angiogenesis.

DESCRIPTION OF THE INVENTION

Antibody

 The present invention therefore relates to an antibody directed against the glycosphingolipid membrane globotriaosylceramide (Gb3), or a functional fragment or a derivative thereof, characterized in that it has at least one complementarity determining region (CDR) having a amino acid sequence selected from SEQ ID NO: 1 to 42, or having an amino acid sequence of at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96% at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 1 to 42.

 In one embodiment, the anti-Gb3 antibody, functional fragment or derivative according to the invention has a heavy chain comprising at least one complementarity determining region (CDR) having an amino acid sequence selected from SEQ ID NO 1 to 3, 7 to 9, 13 to 15, 19 to 21, 25 to 27, 31 to 33, and 37 to 39, or having an amino acid sequence of at least 80%, preferably at least 85% at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 1 to 3, 7 to 9, 13 to 15, 19 to 21, 25 to 27, 31 to 33, and 37 to 39.

 In another embodiment, the anti-Gb3 antibody, functional fragment or derivative according to the invention has a light chain comprising at least one complementarity determining region (CDR) having an amino acid sequence chosen from SEQ ID NO: 4 to 6, 10 to 12, 16 to 18, 22 to 24, 28 to 30, 34 to 36, and 40 to 42, or having an amino acid sequence of at least 80%, preferably at least 85% %, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 4 to 6, 10 to 12 , 16 to 18, 22 to 24, 28 to 30, 34 to 36, and 40 to 42.

Advantageously, the anti-Gb3 antibody, functional fragment or derivative according to the invention has: a heavy chain comprising at least one complementarity determining region (CDR) having an amino acid sequence selected from SEQ ID NO: 1 to 3, 7 to 9, 13 to 15, 19 to 21, 25 to 27, 31 at 33, and 37 to 39, or having an amino acid sequence of at least 80%, preferably at least 85%, at least 90%, at least 95%, at least

96%), at least 97%, at least 98%>, at least 99% identity with one of SEQ ID NO: 1 to 3, 7 to 9, 13 to 15, 19 to 21, 25 to 27 , 31 to 33, and 37 to 39, and

 a light chain comprising at least one complementarity determining region (CDR) having an amino acid sequence selected from SEQ ID NO: 4 to 6, 10 to 12, 16 to 18, 22 to 24, 28 to 30, 34 at 36, and 40 to 42, or having an amino acid sequence of at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%), at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 4 to 6, 10 to 12, 16 to 18, 22 to 24, 28 to 30, 34 to 36, and 40 to 42.

 The anti-Gb3 antibody, functional fragment or derivative according to the invention may also advantageously have a heavy chain comprising three CDR-H (heavy chain CDR) having the following amino acid sequences, or sequences having at least 80% preferably at least 85%, at least 90%, at least 95%, at least 96%), at least 97%, at least 98%, at least 99% identity with the following sequences:

 a) CDR1-H-3E2: SEQ ID NO: 1, CDR2-H-3E2: SEQ ID NO: 2, CDR3-H-3E2: SEQ ID NO: 3,

 b) CDR1-H-14C11: SEQ ID NO: 7, CDR2-H-14C11: SEQ ID NO: 8,

CDR3-H-14C11: SEQ ID NO: 9,

 c) CDR1-H-15C11: SEQ ID NO: 13, CDR2-H-15C11: SEQ ID NO: 14, CDR3-H-15C11: SEQ ID NO: 15,

 d) CDR1-H-22F6: SEQ ID NO: 19, CDR2-H-22F6: SEQ ID NO: 20, CDR3-H-22F6: SEQ ID NO: 21,

e) CDR1-H-25C10: SEQ ID NO: 25, CDR2-H-25C10: SEQ ID NO: 26, CDR3-H-25C10: SEQ ID NO: 27, f) CDR1-H-11E10: SEQ ID NO: 31, CDR2-H-11E10: SEQ ID NO: 32, CDR3-H-11E10: SEQ ID NO: 33, or

 g) CDR1-H-16G8: SEQ ID NO: 37, CDR2-H-16G8: SEQ ID NO: 38, CDR3-H-16G8: SEQ ID NO: 39.

 The anti-Gb3 antibody, functional fragment or derivative according to the invention may also advantageously have a light chain comprising three CDR-L (light chain CDR) having the following amino acid sequences, or sequences having at least 80% preferably at least 85%>, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the following sequences: a) CDR1- L-3E2: SEQ ID NO: 4, CDR2-L-3E2: SEQ ID NO: 5, CDR3-

L-3E2: SEQ ID NO: 6,

 b) CDR1-L-14C11: SEQ ID NO: 10, CDR2-L-14C11: SEQ ID NO: 11, CDR3-L-14C11: SEQ ID NO: 12,

 c) CDR1-L-15C11: SEQ ID NO: 16, CDR2-L-15C11: SEQ ID NO: 17, CDR3-L-15C11: SEQ ID NO: 18,

 d) CDR1-L-22F6: SEQ ID NO: 22, CDR2-L-22F6: SEQ ID NO: 23, CDR3-L-22F6: SEQ ID NO: 24,

 e) CDR1-L-25C10: SEQ ID NO: 28, CDR2-L-25C10: SEQ ID NO: 29, CDR3-L-25C10: SEQ ID NO: 30,

 f) CDR1-L-11E10: SEQ ID NO: 34, CDR2-L-11E10: SEQ ID NO: 35,

CDR3-L-11E10: SEQ ID NO: 36, or

 g) CDR1-L-16G8: SEQ ID NO: 40, CDR2-L-16G8: SEQ ID NO: 41, CDR3-L-16G8: SEQ ID NO: 42.

 Advantageously, the anti-Gb3 antibody, functional fragment or derivative according to the invention has heavy and light chains respectively comprising CDR-H and CDR-L having the following amino acid sequences, or sequences having at least 80 %, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the following sequences:

 a) CDR1-H-3E2: SEQ ID NO: 1, CDR2-H-3E2: SEQ ID NO: 2, CDR3-

H-3E2: SEQ ID NO: 3, CDR1-L-3E2: SEQ ID NO: 4, CDR2-L-3E2: SEQ ID NO: 5, CDR3-L-3E2: SEQ ID NO: 6, b) CDR1-H-14C11: SEQ ID NO: 7, CDR2-H-14C11: SEQ ID NO: 8, CDR3-H-14C11: SEQ ID NO: 9, CDR1-L-14C11: SEQ ID NO: 10, CDR2-L-14C11: SEQ ID NO: 11, CDR3-L-14C11: SEQ ID NO: 12,

 c) CDR1-H-15C11: SEQ ID NO: 13, CDR2-H-15C11: SEQ ID NO: 14, CDR3-H-15C11: SEQ ID NO: 15, CDR1-L-15C11: SEQ ID NO: 16, CDR2

L-15C11: SEQ ID NO: 17, CDR3-L-15C11: SEQ ID NO: 18,

 d) CDR1-H-22F6: SEQ ID NO: 19, CDR2-H-22F6: SEQ ID NO: 20, CDR3-H-22F6: SEQ ID NO: 21, CDR1-L-22F6: SEQ ID NO: 22, CDR2-L-22F6: SEQ ID NO: 23, CDR3-L-22F6: SEQ ID NO: 24,

 e) CDR1-H-25C10: SEQ ID NO: 25, CDR2-H-25C10: SEQ ID NO: 26,

CDR3-H-25C10: SEQ ID NO: 27, CDR1-L-25C10: SEQ ID NO: 28, CDR2-L-25C10: SEQ ID NO: 29, CDR3-L-25C10: SEQ ID NO: 30,

 f) CDR1-H-11E10: SEQ ID NO: 31, CDR2-H-11E10: SEQ ID NO: 32, CDR3-H-11E10: SEQ ID NO: 33, CDR1-11E10L: SEQ ID NO: 34, CDR2- L-11E10: SEQ ID NO: 35, CDR3-L-11E10: SEQ ID NO: 36, or

 g) CDR1-H-16G8: SEQ ID NO: 37, CDR2-H-16G8: SEQ ID NO: 38, CDR3-H-16G8: SEQ ID NO: 39, CDR1-L-16G8: SEQ ID NO: 40, CDR2-L-16G8: SEQ ID NO: 41, CDR3-L-16G8: SEQ ID NO: 42.

 In an advantageous embodiment, the anti-Gb3 antibody, functional fragment or derivative according to the invention has a heavy chain comprising a variable region having a sequence chosen from SEQ ID NO: 43 to 49 or a sequence having at least 80% preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 43 at 49.

 In another advantageous embodiment, the anti-Gb3 antibody, functional fragment or derivative according to the invention has a light chain comprising a variable region having a sequence chosen from SEQ ID NO: 50 to 56 or a sequence having at least 80 %, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 50 to 56.

Advantageously, the anti-Gb3 antibody, functional fragment or derivative according to the invention has: a heavy chain comprising a variable region having a sequence selected from SEQ ID NO: 43 to 49 or a sequence having at least 80%, preferably at least 85%>, at least 90%, at least 95%, at least 96% at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 43 to 49, and

 a light chain comprising a variable region having a sequence selected from SEQ ID NO: 50 to 56 or a sequence having at least 80%), preferably at least 85%, at least 90%, at least 95%, at least 96% at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 50-56.

 In one advantageous embodiment, the anti-Gb3 antibody, functional fragment or derivative according to the invention may be chosen from among the anti-Gb3 monoclonal antibodies generated by the inventors or variants thereof, which have heavy chains. and light ones whose variable regions have the following amino acid sequences or sequences having at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%), at least 99% identity with the following sequences:

 a) Antibody 3E2: heavy chain: SEQ ID NO: 43, light chain: SEQ ID NO: 50,

 b) 14C11 antibody: heavy chain: SEQ ID NO: 44, light chain: SEQ ID NO: 51,

 c) Antibody 15C11: heavy chain: SEQ ID NO: 45 light chain: SEQ ID NO: 52,

 d) 22F6 antibody: heavy chain: SEQ ID NO: 46, light chain: SEQ ID NO: 53,

 e) 25C10 antibody: heavy chain: SEQ ID NO: 47, light chain: SEQ ID NO: 54,

 f) 11E10 antibody: heavy chain: SEQ ID NO: 48 light chain: SEQ ID NO: 55, and

g) 16G8 antibody: heavy chain: SEQ ID NO: 49, light chain: SEQ ID NO: 56. By "antibody" or "immunoglobulin" is meant a glycoprotein composed of two types of glycopolypeptide chains called "heavy chain" and "light chain", an antibody consisting of two heavy chains and two light chains, linked by disulfide bridges. Each chain consists of a variable region and a constant region. The constant region of a particular isotype of heavy or light chain is normally identical from one antibody to another of the same isotype, except somatic mutations. In contrast, the variable region varies from one antibody to another. In fact, the genes coding for the heavy and light chains of the antibodies are generated by recombination of respectively three and two distinct gene segments called VH, DH and JH-CH for the heavy chain and VL and JL-CL for the light chain. The CH and CL segments do not participate in the recombination and form the constant regions of the heavy and light chains respectively. The recombinations of the VH-DH-JH and VL-JL segments form the variable regions of the heavy and light chains, respectively. The VH and VL regions have 3 hyper-variable areas or complementarity determining regions (CDRs), called CDR1, CDR2 and CDR3, the CDR3 region being the most variable, since it is located at the recombination zone. These three CDR regions, and particularly the CDR3 region, are in the portion of the antibody that will be in contact with the antigen and are therefore very important for antigen recognition. Thus, the antibodies that retain the three CDR regions and each of the heavy and light chains of an antibody retain for the most part the antigenic specificity of the original antibody. In a number of cases, an antibody retaining only one of the CDRs, and in particular CDR3, also retains the specificity of the original antibody. The CDR1, CDR2 and CDR3 regions are each preceded by the FR1, FR2 and FR3 regions respectively, corresponding to the framework regions (FR region framework) which vary at least from one VH or VL segment to another. The CDR3 region is also followed by an FR4 framework region.

The CDRs of an antibody are defined from the amino acid sequence of its heavy and light chains with respect to criteria known to those skilled in the art. Various methods for determining CDRs have been proposed, and the portion of the amino acid sequence of a heavy or light chain variable region of an antibody defined as a CDR varies depending on the method chosen. The first one The method of determination is that proposed by Kabat et al (Kabat et al., Sequences of Proteins of Immunological Interest, 5 ^th Ed., US Department of Health and Human Services, NIH, 1991, and later editions). In this method, CDRs are defined by looking for the amino acids responsible for antigen binding of the antibody. A ^2nd method was proposed by IMGT, this time based on the determination of the hypervariable regions. In this method, a unique numbering has been defined to compare the variable regions regardless of the antigen receptor, the type of chain or the species (Lefranc et al., 2003). This numbering provides a standardized delimitation of the framework regions ((FR1-IMGT: positions 1 to 26, FR2-IMGT: 39 to 55, FR3-IMGT: 66 to 104 and FR4-IMGT: 118 to 128) and complementarity determining regions (CDR1-IMGT: positions 27 to 38, CDR2-IMGT: positions 56 to 65 and CDR3-IMGT: positions 105 to 117) Finally, there is also a so-called "common" numbering, in which the sequence of a particular CDR corresponds to the common sequence between Kabat numbering and IMGT numbering Throughout the present description, CDR sequences are indicated in the IMGT numbering In particular, the CDRs were determined using the IMGT / V-QUEST program available on http: // imgt.cines.fr and described in Brochet et al (2008).

Table 2 below summarizes the amino acid sequences of the CDRs and variable regions of the heavy and light chains of the anti-Gb3 antibodies generated by the inventors:

3E2 Antibody 25C10 Antibody

 Heavy chain Heavy chain

CDR1 SEQ ID NO: 1 CDR1 SEQ ID NO: 25

CDR2 SEQ ID NO 2 CDR2 SEQ ID NO 26

CDR3 SEQ ID NO 3 CDR3 SEQ ID NO 27

Variable Region SEQ ID NO 43 Variable Region SEQ ID NO 47

 Light chain Light chain

CDR1 SEQ ID NO 4 CDR1 SEQ ID NO 28

CDR2 SEQ ID NO: 5 CDR2 SEQ ID NO: 29

CDR3 SEQ ID NO 6 CDR3 SEQ ID NO 30

Variable Region SEQ ID NO 50 Variable Region SEQ ID NO 54

 14C11 Antibody 11E10 Antibody

 Heavy chain Heavy chain

CDR1 SEQ ID NO 7 CDR1 SEQ ID NO 31

CDR2 SEQ ID NO 8 CDR2 SEQ ID NO 32

CDR3 SEQ ID NO 9 CDR3 SEQ ID NO 33

Variable Region SEQ ID NO 44 Variable Region SEQ ID NO 48

 Light chain Light chain

CDR1 SEQ ID NO: 10 CDR1 SEQ ID NO: 34

CDR2 SEQ ID NO 1 1 CDR2 SEQ ID NO 35

CDR3 SEQ ID NO 12 CDR3 SEQ ID NO 36

Variable Region SEQ ID NO 51 Variable Region SEQ ID NO 55

 15C11 Antibody 16G8 Antibody

 Heavy chain Heavy chain

CDR1 SEQ ID NO 13 CDR1 SEQ ID NO 37

CDR2 SEQ ID NO 14 CDR2 SEQ ID NO 38

CDR3 SEQ ID NO 15 CDR3 SEQ ID NO 39

Variable Region SEQ ID NO 45 Variable Region SEQ ID NO 49

 Light chain Light chain

CDR1 SEQ ID NO 16 CDR1 SEQ ID NO 40

CDR2 SEQ ID NO 17 CDR2 SEQ ID NO 41

CDR3 SEQ ID NO 18 CDR3 SEQ ID NO 42

Variable Region SEQ ID NO 52 Variable Region SEQ ID NO 56

 22F6 Antibody

 Heavy chain

 CDR1 SEQ ID NO 19

 CDR2 SEQ ID NO 20

 CDR3 SEQ ID NO 21

 Variable Region SEQ ID NO 46

 Light chain

 CDR1 SEQ ID NO 22

 CDR2 SEQ ID NO 23

 CDR3 SEQ ID NO 24

SEQ ID NO 53 variable region By "functional fragment" is meant an antibody fragment retaining the antigen-binding domain and therefore having the same antigenic specificity as the original antibody, such as Fv, ScFv, Fab, F (ab fragments). ') 2, Fab', scFv-Fc or di-antibodies ("diabodies").

 By "derivative" of an antibody is meant a binding protein formed of a carrier peptide and at least one of the CDRs of the original antibody to preserve its ability to recognize Gb3.

 The antibodies, functional fragments or derivatives according to the invention can be obtained by genetic recombination or by chemical synthesis, according to technologies well known to those skilled in the art.

 According to a preferred embodiment, the antibody according to the invention is a monoclonal antibody. By "monoclonal" is meant an antibody obtained from a substantially homogeneous antibody population, i.e. the antibodies forming this population are essentially identical except for possible natural mutations that may be present in minor quantities. These antibodies are directed against a single epitope and are therefore very specific. By "epitope" is meant the site of the antigen to which the antibody binds. Regardless of the antigen, an epitope is a three-dimensional region formed by portions of the antigen that may or may not be adjacent in a linear, non-dimensional structure of the antigen.

 The antibodies, functional fragments or derivatives are of course not in a natural form, but isolated, obtained by purification from a natural source, by genetic recombination or by chemical synthesis.

For the purpose of the present description, the "percent identity" between two nucleic acid or amino acid sequences means the percentage of identical nucleotides or amino acids between the two compared sequences, obtained after optimal alignment of the two sequences. . This percentage is purely statistical and the differences between the two sequences are distributed randomly over their entire length. The comparison of two nucleic acid or amino acid sequences is generally performed after optimally aligning them, the comparison being possible by segment or using an "alignment window". The optimal sequence alignment can be achieved using different software well known to those skilled in the art, including BLAST NR (nucleic acid) or BLAST P (protein) software.

 The percentage identity between two nucleic acid or amino acid sequences is determined by comparing the two optimally aligned sequences, in which the compared nucleic acid or amino acid sequence may have deletions or insertions by compared to the reference sequence. The percent identity is calculated by determining the number of positions at which the nucleotide or amino acid is identical between the two sequences, preferably between the two complete sequences, and dividing it by the total number of positions in the window. alignment (preferably the complete sequences) and multiplying the result by 100. This percentage of identity can be calculated easily using for example the BLAST software with default parameters.

When the CDR or the variable region of an antibody according to the invention has an amino acid sequence which is not 100% identical to one of those described above and in the sequence listing (reference sequences ) but which has at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with at least such a reference sequence, it may have insertions, deletions or substitutions with respect to the reference sequence. When it comes to substitutions, the substitution is preferably carried out by an "equivalent" amino acid, that is to say any amino acid whose structure is close to that of the original amino acid and is therefore unlikely to alter the biological activities of the antibody. Examples of such substitutions are shown in the following Table 3:

Table 3. Substitutions with equivalent amino acids

All the above mentioned antibodies recognize Gb3. In human HMEC-1 cells used by the inventors, Gb3 is represented by immunolabeling with a doublet corresponding to the presence of 2 different fatty acid chains at the ceramide level. Porcine cells also express two forms of Gb3, represented by a doublet. The 3E2 antibody recognizes both forms of Gb3 of the porcine doublet, as well as the upper band, but not the lower band, of the doublet expressed by HMEC-1 cells. Preliminary results suggest that the upper band of the Gb3 doublet expressed by HMEC-1 cells is over-expressed in HMEC-1 cells undergoing angiogenesis compared to quiescent HMEC-1 cells (see Figure 14, where the percentage of cells recognized by the specific antibody 3E2 of this upper band increases when the cells are in complete medium, while the percentage of cells recognized by the antibody 1A4, which recognizes the two bands indifferently, does not vary). Therefore, the specificity of the 3E2 antibody, and any antibody having the same CDR1, CDR2 and CDR3 or the same variable regions, for this upper band presumably further increases its specificity for proliferating endothelial cells, for example. compared with quiescent endothelial cells, which is likely to further reduce its side effects. The 22F6 antibody also recognizes this upper band of the Gb3 doublet expressed by the HMEC-1 cells and should therefore be particularly specific for proliferating endothelial cells, as well as any antibody having the same CDR1, CDR2 and CDR3 or the same variable regions. . This is not the case of the antibodies of the prior art 38.3 and A4, which recognize the two bands of the doublet (see FIGS. 7 and 8), the antibody 38.3 even seeming to recognize mainly the lower band of the doublet (see FIG. 7-1). Thus, in one advantageous embodiment, the antibody, functional fragment or derivative according to the invention is the antibody 3E2 or 22F6 as defined above by the sequences of the variable regions of their heavy and light chains, advantageously the 3E2 antibody, or an antibody of the same antigenic specificity. In particular, the antibody, functional fragment or derivative according to the invention may have at least one complementarity determining region (CDR) having an amino acid sequence chosen from SEQ ID NO: 1 to 6 and 19 to 24, or having an amino acid sequence of at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 1 to 6 and 19 to 24. More specifically, the antibody, functional fragment or derivative according to the invention may have a heavy chain comprising at least one complementarity determining region ( CDR) having an amino acid sequence selected from SEQ ID NO: 1 to 3, and 19 to 21, or having an amino acid sequence of at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 1 to 3, and 19 to 21 and / or has a light chain comprising at least one complementarity determining region (CDR) having an amino acid sequence selected from SEQ ID NO: 4 to 6, and 22 to 24, or having an amino acid sequence having at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO 4 to 6, and 22 to 24. It may also advantageously possess a heavy chain comprising three CDR-H (heavy chain CDRs) having the following amino acid sequences, or sequences having at least 80%), preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the following sequences: a) CDR1-H-3E2: SEQ ID NO: 1, CDR2-H-3E2: SEQ ID NO: 2, b) CDR1-H-22F6: SEQ ID NO: 19, CDR2-H-22F6: SEQ ID NO: 20, CDR3-H-22F6: SEQ ID NO: 21,

 and / or possess a light chain comprising three CDR-L (light chain CDR) having the following amino acid sequences, or sequences having at least 80%, preferably at least 85%,> 90%, at least minus 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the following sequences:

 a) CDR1-L-3E2: SEQ ID NO: 4, CDR2-L-3E2: SEQ ID NO: 5, CDR3-L-3E2: SEQ ID NO: 6,

 b) CDR1-L-22F6: SEQ ID NO: 22, CDR2-L-22F6: SEQ ID NO: 23,

CDR3-L-22F6: SEQ ID NO: 24.

 Advantageously, the anti-Gb3 antibody, functional fragment or derivative according to the invention can have:

 a heavy chain comprising a variable region having a sequence selected from SEQ ID NO: 43 and 46 or a sequence having at least

80%), preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 43 and 46, and / or

 a light chain comprising a variable region having a sequence selected from SEQ ID NO: 50 and 53 or a sequence having at least

80%), preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 50 and 53.

In a particularly advantageous embodiment, the anti-Gb3 antibody, functional fragment or derivative according to the invention can be chosen in particular from the anti-Gb3 monoclonal antibodies generated by the inventors or variants thereof, which possess chains. heavy and light whose variable regions have the following amino acid sequences or sequences having at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97% , at least 98%), at least 99% identity with the following sequences: a) Antibody 3E2: heavy chain: SEQ ID NO: 43, light chain: SEQ ID NO: 50, and

 b) 22F6 antibody: heavy chain: SEQ ID NO: 46, light chain:

 SEQ ID NO: 53.

As indicated in the introduction, the most advantageous isotypes are the IgG isotypes, in particular IgG1, and IgM, more particularly the IgM isotype, which has the highest molecular weight. Therefore, any antibody according to the invention is preferably of IgG or IgM isotype, preferably IgM.

 In addition, any antibody, functional fragment or derivative according to the invention may advantageously be chimeric or humanized. Indeed, this makes it possible to avoid the immune reactions of the patient against the administered antibody. In particular, the antibody according to the invention may advantageously be any of the chimeric or humanized versions of the antibodies 3E2, 14C1 1, 15C11, 22F6, 25C10, 11E10, and 16G8 described above, advantageously 3E2 and 22F6 antibodies. , and in particular of the antibody 3E2.

By "chimeric" antibody is meant an antibody which contains a naturally occurring variable (light chain and heavy chain) derived from an antibody of a given species in association with the constant light chain and heavy chain regions of an antibody of a heterologous species to said given species. The chimeric type antibodies according to the invention can be prepared using genetic recombination techniques. For example, the chimeric antibody may be made by cloning a recombinant DNA comprising a promoter and a sequence coding for the variable region of a non-human monoclonal antibody, in particular murine monoclonal antibody, according to the invention and a sequence coding for the constant region. human antibody. A chimeric antibody of the invention encoded by such a recombinant gene will for example be a mouse-human chimera, the specificity of this antibody being determined by the variable region derived from murine DNA and its isotype determined by the constant region derived from the murine DNA. Human DNA. This will notably be the case for the chimeric antibodies obtained from the murine monoclonal antibodies 3E2, 14C11, 15C11, 22F6, 25C10, 11E10, and 16G8 described in the present description. For methods of preparing antibodies For example, it is possible to refer to Verhoeyn et al. (BioEssays, 8: 74, 1988).

 By "humanized" antibody is meant an antibody which contains CDRs regions derived from an antibody of non-human origin, the other parts of the antibody molecule being derived from one (or more) human antibodies. In addition, some of the skeletal segment residues (referred to as FR) may be modified to maintain binding affinity (Jones et al., Nature, 321: 522-525, 1986, Verhoeyen et al., Science, 239: 1534-1536 1988, Riechmann et al., Nature 332: 323-327, 1988). The humanized antibodies according to the invention may be prepared by techniques known to those skilled in the art such as "CDR grafting", "resurfacing", Superhumanization, "Human string content", "FR libraries" technologies. "," Guided Selection "," FR shuffling "and" Humaneering "as summarized in the review by Almagro et al.

The anti-Gb3 antibody, or functional fragment or derivative thereof according to the invention may also have been optimized for certain effector functions. Thus, it may include mutations increasing its affinity for the Fc receptor to which its isotype binds. It can also be produced under specific conditions (cells, medium, etc.) to obtain a particular glycosylation of the antibody. In particular, for IgG, certain substitutions of the Fcγ portion and low or no fucosylation make it possible to increase the affinity for the FcγRIII receptor (see in particular Shields et al., Journal of Biological Chemistry, Vol 276, No. 9 , Issue of March 2, pp. 6591-6604, 2001, EP1176195A1, EP1331266A1, WO 01/77181). Alternatively or in addition, the antibody can also, especially when it is an IgG isotype and in particular an IgG1 isotype, have been modified to limit its ability to fix the complement and therefore its activity of complement-dependent cytotoxicity (CDC), while preserving its antibody-dependent cellular cytotoxicity (ADCC) functions. Indeed, in the case of an anti-GD2 antibody, it has been shown that a point mutation in the constant part of the human IgG1 region (replacement of the lysine at position 322 of the human Fc constant part of the light chain κ by alanine) allowed to reduce its complement fixation while maintaining its ADCC properties, and to reduce some side effects associated with the administration of this antibody (US7432357B2). So, in a mode of Advantageously, the anti-Gb3 antibody according to the invention is an antibody with a low CDC activity, that is to say that in a CDC activity test in vitro, the CDC activity of the antibody at a concentration of 0, 1 μg / ml is not significantly different from that of the control without antibodies. In particular, an advantageous embodiment relates to a humanized anti-Gb3 IgG1 isotype antibody, comprising an κ isotype light chain, in which the lysine at position 322 of the human Fc constant part of the κ light chain is replaced by an alanine or an equivalent amino acid, preferably an alanine. The invention also relates to a nucleic acid (also called nucleic or nucleotide sequence) coding for any of the anti-Gb3 antibodies, or functional fragments or derivatives according to the invention as described above. All the different nucleic sequences, because of the degeneracy of the genetic code, encoding a particular amino acid sequence are within the scope of the invention. Such nucleic sequences may have been optimized to promote its expression in a host cell of interest. Particular sequences of interest are those of the variable regions of the heavy and light chains of the antibodies generated by the inventors, represented as follows:

 Antibody 3E2: heavy chain: SEQ ID NO: 57, light chain: SEQ ID NO: 64,

 14C11 antibody: heavy chain: SEQ ID NO: 58, light chain: SEQ ID NO: X65

 15C11 antibody: heavy chain: SEQ ID NO: 59, light chain: SEQ ID NO: 66,

 - 22F6 Antibody: heavy chain: SEQ ID NO: 60, light chain: SEQ

ID NO: 67

 25C10 Antibody: heavy chain: SEQ ID NO: 61 light chain: SEQ ID NO: 68,

 11E10 antibody: heavy chain: SEQ ID NO: 62, light chain: SEQ ID NO: 69,

16G8 antibody: heavy chain: SEQ ID NO: X63 light chain: SEQ ID NO: 70. The invention also relates to a vector comprising at least one of the nucleic sequences described above. Such a vector comprises the elements necessary for the expression of said nucleic sequence, and in particular a promoter, a transcription initiation codon, termination sequences, and appropriate transcriptional regulatory sequences. These elements vary according to the host serving for the expression and are readily selected by those skilled in the art in view of his general knowledge. The vector may especially be plasmidic or viral. It is used to clone or express the nucleic sequences according to the invention.

 The invention also relates to a host cell comprising one or more nucleic sequences or one or more vectors according to the invention. The host cell may be of procary or eukaryotic origin, and may especially be selected from bacterial cells, insect cells, plants, yeast or mammals. The antibody, functional fragment or derivative according to the invention can then be produced by culturing the host cell under appropriate conditions. It can also be obtained by chemical synthesis, in particular in the solid or semi-solid phase.

The antibodies, functional fragments or derivatives according to the invention have anti-angiogenic properties as such. In particular, they exhibit cytostatic and cytotoxic activity on proliferating endothelial cells. They can therefore be used alone in the treatment of diseases associated with angiogenesis, as defined below. Thus, in one embodiment, the antibodies according to the invention described above are not coupled to a therapeutic molecule, in particular a cytotoxic, cytostatic or anti-angiogenic molecule such as a toxin, in particular an angiogenesis-inhibiting toxin. For the purposes of the present invention, it is considered that an antibody is "coupled" to a molecule if there is a covalent bond between the antibody and the molecule Thus, the fact that the antibody according to the invention is not coupled to a therapeutic molecule means that it is not bound by a covalent bond to such a molecule. However, this does not prevent the antibody, which is not coupled to a therapeutic molecule, from being administered in combination (simultaneously or sequentially) with another therapeutic molecule, including a cytotoxic, cytostatic or anti-angiogenic molecule in the case of tumors. the two molecules can act independently of one another. In another embodiment, the anti-Gb3 antibodies, or functional fragments or derivatives thereof according to the invention as described above can nevertheless also be coupled to another therapeutic molecule. Treatment of diseases associated with angiogenesis

 The present invention also relates to an antibody, directed against the Gb3 membrane glycosphingolipid and not coupled to a therapeutic molecule, for its use as a medicament in the treatment of angiogenesis-associated diseases. It also relates to the use of an antibody directed against the Gb3 membrane glycosphingolipid and not coupled to a therapeutic molecule, for the preparation of a medicament for the treatment of diseases associated with angiogenesis. It also relates to a method of treating angiogenesis-associated diseases in a subject in need, comprising administering an effective amount of an antibody directed against Gb3 membrane glycosphingolipid and not coupled to a therapeutic molecule. It also relates to a method for inhibiting angiogenesis in a subject in need, comprising administering an effective amount of an antibody directed against Gb3 membrane glycosphingolipid and not coupled to a therapeutic molecule.

By "diseases associated with angiogenesis" is meant any disease whose evolution requires angiogenesis to develop, that is to say the formation of new vessels from a pre-existing vascular network. Among these diseases associated with angiogenesis include solid tumors; psoriasis; angiomas; prominent eye diseases, including age-related macular degeneration (AMD), diabetic retinopathy, or neovascular glaucoma; autoimmune diseases, including lupus or rheumatoid arthritis, as well as atherosclerosis-related diseases; obesity or Alzheimer's disease. Indeed, the development of all these diseases involves angiogenesis. Of course, blood cell diseases, such as leukemias or lymphomas, are "liquid" tumors that do not require angiogenesis, and are therefore not considered to be angiogenesis-associated diseases within the meaning of the present invention. In an advantageous embodiment, the anti-Gb3 antibody, not coupled to a molecule cytotoxic, cytostatic or anti-angiogenic, is used in the treatment of solid tumors.

 Solid tumors for which anti-Gb3 antibodies are useful include adenomas, sarcomas, and carcinomas; and in particular adenocarcinomas, ovarian, breast, pancreatic, skin, lung, brain, kidney, liver, nasopharyngeal cavity, thyroid, central nervous system (neuroblastoma by example), prostate, colon, rectum, cervix, testes, or bladder.

 In an advantageous embodiment of the application to the treatment of diseases associated with angiogenesis, in particular for the treatment of solid tumors, the anti-Gb3 antibody used is any of those described above in the section concerning the antibodies as such.

 In another advantageous embodiment, the anti-Gb3 antibody recognizes the upper band but not the lower band of the Gb3 doublet expressed by HMEC-1 cells. In particular, it may be antibodies 22F6 and 3E2, or an antibody possessing at least one CDR or all the CDRs, or even one and / or the other of the variable regions of these antibodies.

In the treatment of solid tumors, the anti-Gb3 antibody may be administered in combination with another treatment, including surgical resection of the tumor, radiotherapy treatment or chemotherapy with another therapeutic molecule, provided that another therapeutic molecule is not coupled to the antibody. The treatment (radiotherapy / radiotherapy / other therapeutic molecule) can then be administered simultaneously (i.e. at the same time, in the same composition or in two separate compositions) or sequentially ( one then the other, possibly alternately). They can also be administered first simultaneously then sequentially, or vice versa. The combination may also be administered first (simultaneously or sequentially) and then the treatment continued with only one of the two therapies. Provided that a therapeutic molecule is not coupled to the antibody, it can be selected from all other anticancer molecules, whether they also target angiogenesis or the tumor itself. DESCRIPTION OF THE FIGURES

Figure 1. (A) HPTLC profiles of growth phase (lane 1) or confluent (track 2) HMEC-1 cells, rat brain ganglioside markers (lane 3) and standard neutral glycolipids (lane 4). (B) HPTLC profiles of HMEC-1 cells incubated in depleted (lane 1) or complete (lane 2) culture media, ganglioside markers of rat brain (lane 3), and a mixture of standard neutral glycosphingolipids (lane 4). (C) HPTLC profiles of HMEC-1 cells treated with PET (lane 1) or S1P (lane 2), rat brain ganglioside markers (lane 3), and standard neutral glycolipids (lane 4). The migration solvent used is C / M / H ₂ O, 0.2% CaCl ₂ , 55: 45: 10 v / v / v. All bands are revealed by orcinol.

 Figure 2. Protocol for obtaining antibodies against HMVEC-L endothelial cells cultured in the presence of T84 human adenocarcinoma cells.

 Figure 3. Principle of 96-well microplate limiting dilution cloning. (Panel A) In position A1 are seeded 1000 hybridoma cells which are then diluted cascade half up, down wells A to H, then from left to right, wells 1 to 12. Panel B represents the theoretical number of cells remaining after dilution.

 Figure 4. Electrophoretic analysis of IgM 3E2 mAb on a 6% polyacrylamide gel under non-reducing conditions (A) on a 12% polyacrylamide gel under reducing conditions (B).

 Figure 5. Analysis of flow cytometric labeling of HMEC-1, Raji and NXS2 cells with purified anti-Gb3 monoclonal antibodies 3E2. C (A), 25C 10.a (B), 16G8.h (C), 14Cl1. ld (D), 15C7.2e (E). All cells were labeled with 10 μg / ml of antibody. Percentages of positive cells are noted in the panel figures A, B, C, D, E and F.

Figure 6. Study of the Specificity of the 3E2 Antibody by ELIS A (Panel A) Fixation of the 3E2 Antibody Measured by ELISA on Raji Cells, HMEC-1 Cells and IMR32 Cells (Panel B) Attachment of the 3E2 antibody and the 38.13 anti-Gb3 rat monoclonal antibody to the HMEC-1 cells measured by ELISA. The initial concentration of the antibodies is equal to 10 μg / ml (3 independent analyzes). Figure 7. Immunostaining on HPTLC of HMEC-1 cells with mAb 3E2, mAb 3, 8, and supernatant of uncleaved hybridoma 3E2. Ganglioside markers of rat brain (lane 1) and glycolipid extracts of HMEC-1 cells (lanes 2, 3, 4 and 5). (A) The bands are chemically revealed by orcinol. (B) Immunostaining is carried out with 5 μg / ml anti-Gb3 38-13 rat mAb (lane 3), 5 μg / ml mAb 3E2 (lane 4) and the hybridoma supernatant 3E2 before cloning (track 5) ·

 Figure 8. Immunostaining on HPTLC of HMEC-1 cells by mouse mAbs 3E2 and A4. Rat brain ganglioside markers (lane M), neutral glycolipid mixture (lane 1), purified porcine Gb3 (lane 2) and glycolipid extracts of HMEC-1 cells (lane 3). (Panel A) The lines are chemically revealed by orcinol. (Panel B) Immunostaining is performed with 5 μg / ml of MAb 3E2 or 1 A4.

 Figure 9. Analysis of the specificity of the antibody 3E2 by flow cytometry on HMEC-1, Raji and NXS2 cells. Expression of Gb3 on HMEC-1, Raji and NXS2 cells by flow cytometry with 10 μg / ml mAb 3E2 (A), 1A4 (B) and IgM isotype control, κ (C). Percentages of positive cells are noted in Table 10.

 Figure 10. HPTLC glycolipid profile revealed in orcinol of HMEC-1 (4), Raji (5), NXS2 (6) and T84 (7) cells. Rat brain ganglioside markers (lane 1), a mixture of standard neutral glycolipids (lane 2) and purified porcine Gb3 (lane 3).

 Figure 11. Expression of Gb3 on HMEC-1, HMVEC-L and HUVEC cells by flow cytometry with 10 μg / ml mAb 3E2 (A), 1 A4 (B) and isotypic IgM control, κ (C) . Percentages of positive cells are noted in FIG. 11 and the fluorescence intensity average values are listed in Table 11.

 Figure 12. glycolipid profile by HPTLC revealed to orcinol cells HMEC-1 (4), HMVEC-L (5), HUVEC (6). Rat brain ganglioside markers (lane 1), mixture of standard neutral glycolipids (lane 2) and purified porcine Gb3 (lane 3) ·

Figure 13. Nucleotide sequences of the variable regions of the V _H (panel A) and V _L (panel B) gene segments of the 3E2 antibody. Figure 14. Flow cytometric analysis of Gb3 expression of HMEC-1 cells incubated for 24 h in depleted media (MA) and in complete medium (MC). Fixation was measured using mAb 3E2 (Panel A) and mAb 1 A4 (Panel B). The percentage of positive cells and the fluorescence intensity average values are listed in Table 14 (3 independent experiments).

 Figure 15. Flow cytometric analysis of Gb3 expression of HMEC-1 cells incubated in depleted media supplemented with SμΜ SIP or PET. Fixation was measured using mAb 3E2 at 24 h (Panel A) and 72 h (Panel B). The percentage of positive cells and the average fluorescence intensity values are listed in Table 15 (3 independent experiments).

 Figure 16. Cell viability of HMEC-1, Raji and NXS2 cells measured by MTT after 24 h incubation of mAb 3E2 and 1A4. Cell viability of HMEC-1 cells in the presence of mAb 3E2 and A4 (Panel A) and Raji and NXS2 cells in the presence of Γ mAb 3E2 (panel B), 3 independent experiments.

 Figure 17. Kinetics of cell viability of HMEC-1 cells measured by

MTT after 24 h, 48 h and 72 h incubation of mAb 3E2 at 20 μg / ml (3 independent experiments).

 Figure 18. Representation of the number of HMEC-1 cells as a function of the incubation time of the mAb 3E2 at 20 πιΐ.

 Figure 19. Percentage of HMEC-1 cells in division and estimation of their cell division time in the presence of the antibody 3E2. (Panel A) Proportion of cells that enter cell division relative to the number of cells initially, as a function of incubation time (n = 3). (Panel B) Cell division time: proportion of cells entered in division as a function of their cell division time, relative to the total number of cell divisions over the 24 hours of incubation (n = 3).

Figure 20. Aortic ring test performed with mAb 3E2 and 1A4 (n = 3). (Panel A) Microvessels in development from transverse sections of murine aortic rings. Photo 1, growth of microvessels at implantation of aortic sections (J0). Photo 2, microvessel growth after five days after incubation in complete culture medium. Photos 3 and 4, aortic rings treated with respectively 20 μg / ml and 40 μg / ml of mAb 3E2. (Panel B) Scores assigned length, density and number of vascular buds from analysis of aortic rings images (scale from 0 to 5, n = 5).

 Figure 21. Evaluation of the CDC activity of mAb 3E2 and 1A4. Specific lysis of the cells was determined by flow cytometry by measuring the percentage of cells having incorporated propidium iodide. The CDC activity is measured in the presence of decomplemented human serum (') or not decomplemented (c'). (Panel A) CDC activity performed on HMEC-1 cells, (Panel B) Raji cells and (Panel C) on NXS2 cells.

 Figure 22. Nucleotide sequences of the variable regions of the heavy chain of 7 anti-Gb3 antibodies.

 Figure 23. Nucleotide sequences of the variable regions of the anti-Gb3 antibody light chain.

Figure 24. Reduction of the growth of a subcutaneous tumor by the 3E2 antibody. Female AI mice were inoculated subcutaneously in the right flank with 10 ⁶ NXS2 cells (neuroblastoma line) alive (viability> 95%) resuspended in 150 μl of PBS. When the tumors reached a volume V of 100-150 mm ³ , the mice received a single injection of 3E2 antibody directed against the upper band of the doublet Gb3 expressed by the HMEC-1 cells, anti-GD2 14G2a antibody, IgM isotype control antibody or PBS alone. The graph represents the tumor volume, expressed as a percentage of the tumor volume during the injection of the treatment, as a function of the duration after treatment.

Figure 25. The 3E2 antibody inhibits the growth of metastases in vivo. (A) Representative photograph of mouse livers at day 28 after intravenous injection of NXS2 cells. The scale bars represent 1 cm. (B) Number of metastases evaluated for each animal liver (PBS n = 4, 3E2 n = 6, 14G2a n = 4, IgM control n = 13, mean ± SEM, * p <0.05). EXAMPLES

EXAMPLE 1 Materials and Methods of Examples 2 to 6

1.1. Cell culture.

Endothelial cells: HMVEC-L, HMEC-1, HUVEC.

HMVEC-L cells are primary human microvascular endothelial cells from lungs (Cambrex, Clonetics ^®, USA). Each ampoule corresponds to a single Caucasian donor. They were seeded at a density of 5 10 ⁴ cells / cm ² in 6-well plates, then cultured in a humid atmosphere enriched with C0 ₂ at 37 ° C in the middle of specific EGM -2 ^® MV Culture (Cambrex, USA) having 25 ml of FCS (fetal calf serum), 0.2 ml of hydrocortisone, 2 ml of hFGF-B, 0.5 ml of VEGF, 0.5 ml of R3-IGF-1, 0.5 ml of ascorbic acid, 0.5 ml of hEGF and 0.5 ml of GA-1000, per 500 ml of medium. They are delivered in the third pass and can be grown until the seventh pass. The line FDVIEC-1 (human microvascular endothelial cells) was obtained by the team of Pr. FJ Candal (Center for Disease Control, Atlanta, Ades et al., 1992) after immortalization of microvascular endothelial cells of human skin by transfection with a plasmid containing the coding region of the SV40 virus T antigen. These cells retain the main characteristics specific to microvascular endothelial cells (Ades et al., 1992) and are cultured between passages 13 and 20 in order to limit phenotypic drift related to cell culture. They are seeded at a cell density of 2 × 10 ⁴ cells / cm ² and cultured for 5 days until they are confluent in a C0 ₂ enriched humid atmosphere at 37 ° C. in MCDB 13 1 culture medium (Invitrogen, Cergy-Pontoise, France) containing 15% inactivated FCS at 56 ° C for 45 min, 10 ng / ml EGF (Becton-Dickinson, BD Biosciences, Le Pont de Claix, France), 2 μg / ml hydrocortisone (Sigma-Al). drich, Saint Quentin-Favier, France), 2 mM L-glutamine (Gibco-BRL, Cergy Pontoise, France), 100 IU / ml penicillin and 100 μg / ml streptomycin (Gibco-BRL). HUVEC cells are human macrovascular primary cells extracted from umbilical cord veins (Cambrex). Each ampoule corresponds to several donors. They are seeded at a density of 3.5 × 10 ⁴ cells / cm ² in 6-well plates and then cultured in a humid atmosphere. enriched in C0 ₂ at 37 ° C in the specific culture medium EGM -2 (Cambrex, USA) according to: 10 ml of FCS, 0.2 ml of hydrocortisone, 2 ml of hFGF-B, 0.5 ml of VEGF, 0.5 ml of R3-IGF-1, 0.5 ml of ascorbic acid, 0.5 ml of hEGF, 0.5 ml of GA-1000 and 0.5 ml of heparin, per 500 ml of middle. They can be grown from the first to the sixth pass.

Tumor lines: Raji, NXS2, IMR32 and T84.

The Raji human lymphoma cells and the IMR32 human neuroblastoma cell line are from ATCC (Rockville, USA). The NXS2 cells of murine neuroblastoma were provided by Prof. H. Lode (Charity Children's Hospital, Berlin, Germany). The Raji cells are inoculated at 0.1 × 10 ⁶ cells / ml in RPMI 1640 medium (Invitrogen) supplemented with 10% inactivated FCS, 2 mM L-glutamine, 100 IU / ml penicillin and 100 μg / ml. streptomycin. The IMR32 cells are cultured in the same medium but are inoculated at 2 × 10 ⁴ cells / cm ² . The NXS2 cells, inoculated at a rate of 2 × 10 ⁴ cells / cm ^2, are maintained in DMEM medium containing 4.5 g / l of glucose (Sigma), 10% of inactivated SVF, 2 mM of L-glutamine, 100 IU / ml of penicillin and 100 μg / ml of streptomycin. The T84 epithelial cells of a human colon adenocarcinoma were provided by Dr. M. Neunlist (INSERM UMR913, Nantes). They are seeded at a density of 2 × 10 ⁵ cells / cm ² and cultured in DMEM / F12 medium (1: 1, Gibco-BRL) with 10% inactivated FCS, 2 mM L-glutamine, 100 IU / ml of penicillin and 100 μg / ml streptomycin.

Reagents used.

 Mouse monoclonal antibody IgA 1A4 directed against neutral glycolipid

Gb3 was graciously given to us as ascites by Dr. J. Wiels (Institut Gustave Roussy, Villejuif, France). It was purified by affinity chromatography using a mannan column (Pierce, Rockford, USA) and conditioned at 4 ° C in PBS. The 38.13 anti-Gb3 rat IgM monoclonal antibody and isotype control of mice (clone 11E10) were obtained in purified form from Beckman Coulter (Fullerton, CA, USA). The "isotypic" control antibody, packaged in a 100 mM borate solution, was dialysed in PBS and then 0.22 membrane filtered. μηι. Biotinylated secondary goat antibodies such as mouse anti-IgG + IgM (H + L) antibodies, mouse anti-IgG (H + L) antibodies, mouse anti-mouse antibodies were provided by Jackson Immunoresearch (Laboratories, Westgrove, PA, USA). Peroxidase-coupled goat antibodies specific for the rat μ chain and biotinylated goat secondary antibodies specific for mouse subclasses (Fc _Y i, Fc _Y 2a, Fc _Y 2b, Fc _Y 3) come from the same supplier. . Biotinylated goat antibodies specific for mouse light (κ and λ) chains were provided by Southernbiotech (Birmingham, AL, USA). These antibodies have been adsorbed against human, bovine or rabbit serum proteins. The streptavidin-R-phycoerythrin complex (R-PE) was obtained from Biosource (Camarillo, CA, USA), the streptavidin-horseradish-peroxidase complex and the streptavidin-FITC complex were obtained from Beckman Coulter.

The mixture of neutral glycosphingolipids comprising lactosylceramide (LacCer), galactosylceramide (GalCer), ceramide trihexoside (Gb3) and globoside (Gb4) as well as purified Gb3 were obtained from Matreya (Pleasant Gap, PA, USA). These glycolipids are either of porcine origin (Gb3 and Gb4) or of bovine origin (LacCer and GalCer). The purified Gb3 was redissolved with 1 ml of a chloroform: methanol (2: 1 v / v) mixture so as to obtain a final concentration of 1 μg / μl. Sphingosine-1-phosphate (S1P) was obtained from Biomol International (Plymouth Meeting, PA, USA) and rehydrated (Morita et al., 2000) in order to obtain a concentration of 10 μΜ in the PET diluent consisting of 5% of polyethylene glycol, 2.5% ethanol and 0.8% Tween-80. The mixture of gangliosides markers standards (GMi, GDi _has GDi, GTi _b) was prepared in the laboratory by extraction of rat brain gangliosides, according to the technique described by Folch (Folch et al. 1951). 1.2. Extraction and purification of glycolipids.

Extraction of total glycolipids from cell pellets.

The glycolipids were extracted from cell pellets according to the method of S. Ladish (Ladish and Li, 2000). The cells are trypsinized, washed twice in PBS pH 7.3 and then incubated at -20 ° C. overnight. They then undergo hypotonic lysis in 1 ml of distilled water. Aliquots are then removed and centrifuged at 15,000 g, 4 ° C, for 15 min to remove insoluble particles, debris cell and DNA. The protein concentration of the aliquots is then determined by absorbance at 280 nm (Nanodrop ND-1000, Labtech, Palaiseau, France) in order to deposit on a thin layer of silica a quantity of glycolipids corresponding to 0.5 - 2.5 mg of protein equivalent per sample. Extraction of the lipids from the lysed cells in distilled water is carried out in chloroform: methanol (1: 1, v / v; 20 ml / 10 ⁹ cells) for 18 hours with mechanical stirring at room temperature. The samples are then centrifuged for 10 min at 750 g, and the cell pellets again extracted for 4 h in an additional volume of chloroform: methanol (1: 1, v / v). After evaporation of the organic solvent containing the glycolipid under nitrogen flow, the total lipid extract is obtained.

Purification of glycolipids.

First step: DIPE / 1-butanol partition. The total lipid extract is redissolved in 5 ml of an organic mixture containing DIPE / 1-butanol (60:40, v / v). Series of vortex and sonication (Diagenode's Bioruptor, Sparta, USA) are alternated and repeated until an opalescent solution is obtained. By adding and mixing 2.5 ml of 0.1% NaCl saline, a partition is obtained after centrifugation at 750 g for 10 min at 4 ° C. The upper organic phase containing neutral lipids and phospholipids is pipetted and can be stored at -20 ° C. The lower aqueous phase and the interface containing the acidic glycolipids and the more hydrophilic neutral glycolipids are again partitioned with a new volume of DIPE / 1-butanol (60:40, v / v). The glycolipids contained in the aqueous phase are evaporated in a speed-vac (Eppendorf, Hamburg, Germany). Second step: desalting by gel filtration. In order to remove salts and contaminants such as proteins that can be co-extracted at this stage, the dry residues are redissolved and vortexed in 500 μΐ of solvent A (chloroform: methanol: H ₂ O (30: 60: 8). , v / v / v)) then introduced into a 10 ml column of LH-20 Sephadex, 1 x 30 cm (Pharmacia Fine Chemicals, Uppsala, Sweden) pre-equilibrated with solvent A. After addition of the glycolipids to be desalinated, the first 3 ml of solvent A are eluted and the next 3 ml are harvested as desalted glycolipid fraction. This fraction is then evaporated under a stream of nitrogen and stored at -20 ° C. 1.3. Glycolipid Analysis by Silica Thin Layer Chromatography

(HPTLC).

The glycolipids recovered are taken up in a small volume of chloroform: methanol (1: 2, v / v) so as to deposit, with the aid of a TLC ATS4 deposition automaton (Camag, Muttenz, Switzerland) 2 to 2, 5 mg of protein equivalent per glycolipid sample. The glycolipids are separated on HPTLC plates consisting of a silica gel 60 covering an aluminum foil (Merck, Darmstadt, Germany). Before deposition, these silica plates are subjected to a first migration in a glass vessel (Camag) containing a mixture of chloroform: methanol (1: 1, v / v), in order to rid them of possible contaminants. After deposition of glycolipids, a first migration system consisting of chloroform: methanol solvent (2: 1, v / v) makes it possible to migrate and eliminate very apolar lipids which can be coextracted and which can interfere with the separation of glycolipids of interest. The latter are then separated during a final development in a saturated tank for 4 h with 70 ml of chloroform: methanol: 0.2% aqueous CaCl ₂ (55: 45: 10, v / v / v). The strips are chemically exposed at 150 ° C. after spraying the plate with a solution of orcinol previously prepared by dissolving 0.2 g of orcinol (Sigma-Al drich, St. Louis, USA) in 40 ml of water. distilled and 10 ml of sulfuric acid. This reagent allows the visualization of purple bands characteristic of the presence of glycosylated residues. The glycolipids can then be quantified by ImageQuant ⁵ ' ² densitometry software (GE Healthcare, Waukesha, USA).

1.4. Immunofixation of glycolipids separated by HPTLC (iTLC).

The specificity of the antibodies can be determined on glycolipids separated by HPTLC. After separation by chromatography of glycolipids for which 2 to 2.5 mg of protein equivalent per sample is exposed to orcinol and 0.5 mg of protein equivalent for the samples which will be labeled by the antibodies, the HPTLC plates are cut into strips and plasticized by incubation for 1 min in 0.1% polyisobutyl methacrylate dissolved in hexane, then saturated at room temperature for 1 hour with 1% PBS-BSA. The strips are then incubated overnight at 4 ° C. with the antibodies (either directly incubated with the hybridoma supernatants or incubated with 5 μg / ml of purified antibody diluted in 0.1% PBS-BSA). After three washes with PBS, the antibody binding is detected by two successive incubation steps, a first incubation step of 1 h at room temperature of the biotinylated secondary antibodies (diluted 1: 2000 in PBS). 0.1% BSA), followed by incubation of the streptavidin-horseradish-peroxidase complex, diluted 1: 2000 for 1 hour at room temperature. After several washes with PBS, the bound peroxidase is revealed by a solution of 4-chloro-1-naphthol (Sigma) which is prepared extemporaneously at the rate of 1 mg of product dissolved in 1 ml of methanol, taken up in 20 ml of PBS. and added 30 μΐ of hydrogen peroxide 30 volumes. 1.5. Analysis of antibody specificity by enzyme immunoassay

(ELISA) on desiccated cells in a 96-well microtiter plate.

The desiccated cell plates were prepared in the following manner. The cells are trypsinized and washed three times with cold PBS, and then diluted to obtain a cell concentration of 2 × 10 ⁶ cells / ml. Is then deposited 50 μΐ of this suspension to the bottom of each of the 96 wells of a microtiter plate Immuno-Plate (Nunc, Maxisorp ^®, Denmark). After desiccation of the cells overnight in an oven at 37 ° C, the plates are either used immediately or stored at room temperature for several months under aluminum.

After washing the desiccated cells with PB S and blocking nonspecific interaction sites with 1% PBS-BSA for 1 h at room temperature with gentle shaking, the antibodies diluted in 0.1% PBS-BSA (10 μg / ml of initial concentration) are deposited in each of the triplicate wells, with a volume of 100 μΐ of antibody in a dilution range of 1: 256. The plates are incubated for 2 hours at room temperature and then washed. three times with PBS. The binding of the antibody is detected successively by a first incubation for 1 h at room temperature with the biotinylated secondary antibodies (diluted 1: 4000 in PBS-BSA 0.1%), then after washing, by incubation. streptavidin-horseradish-peroxidase complex (diluted 1: 4000, 1 hr at room temperature). After several washes with PBS, the fixation of the peroxidase is revealed by a solution of ABTS (Merck) which reveals a progressive green coloration. The reaction can then be blocked by the addition of 10% SDS. The measurement of the optical density is determined by reading the plate with a spectrophotometer at 405 nm (Multiskan Thermo Electron reader, Illkirch, France).

1.6. Immunofluorescence cell staining study in flow cytometry. To detect cell surface antigens by flow cytometry, the cells are trypsinized and washed twice in cold PBS containing 2.5% decomplemented FCS (PBSF). They are then taken up in PBSF and deposited at a rate of 4 × 10 ⁵ cells per well of a plate with a conical bottom (Nunc, Denmark). The plate is centrifuged at 750 g for 1 min in order to eliminate the supernatant, then the nonspecific binding sites are blocked by a 30 min incubation directly on ice with 5% of human serum, previously filtered (0.22 μιη) and decomplemented at 56 ° C for 45 minutes (donors of the EFS, Nantes). The cells are then directly incubated with 100 μl of primary antibodies diluted to 10 μg / ml in PBSF for 45 minutes on ice. After incubation, the cells are washed 3 times with cold PBS and then fixed with 4% paraformaldehyde (Electron Microscopy Science, Washington, USA) in PB S, on ice for 15 minutes, in order to prevent the internalization of the cells. antigen-antibody complexes. After three weeks at PB S, antibody fixation is first detected by incubation for 30 minutes on ice of biotinylated secondary antibodies (diluted 1: 400 in PBSF) followed by washes and then incubation. streptavidin-phycoerythrin complex (diluted 1: 400, 30 minutes on ice). After several washes with PBS, determination of the distribution and intensity of fluorescence of the cells (1 × 10 ⁵ cells) is performed using a FAC SCalibur flow cytometer and Cell Quest runtime software. Pro (BD Biosciences, San Jose, USA). 1.7. Cell localization of Gb3 by immunocytochemistry.

The cells are seeded in their culture medium on glass slides 14 mm in diameter (Thermo Scientific, Hudson, USA) in a 24-well plate, 24 to 48 hours before labeling, at a rate of 2 × 10 ⁵ cells. / cm ² . The non-specific sites are blocked by incubating 5% inactivated human serum on ice and then the cells are incubated with the antibodies diluted to 40 μg / ml in PBSF for 45 minutes on ice. After labeling, the cells are washed 3 times with cold PBS and then fixed with 4% paraformaldehyde on ice for 15 minutes. The cells are again washed 3 times with PBS and the binding of the antibody is detected first by a 30 minute incubation on ice of the biotinylated secondary antibodies (diluted 1: 400 in PBSF) followed by 30 minutes of ice incubation of streptavidin-FITC complex (diluted 1: 400). After staining of the cell membrane, the nuclei are labeled for 15-20 min at room temperature, Drag5 (Biostatus, Leicestershire, UK) diluted 1: 1000 in PBS. The slides are assembled in Fluoromount-G medium (Southernbiotech, Birmingham, AL, USA) and the cell labeling is observed under the confocal microscope TCS-SP1 (Leica, Mannheim, Germany, PICell platform, IFR 26, Inserm, Nantes) at a magnification of 63 X.

1.8. Analysis of the distribution of Gb3 by immunohistochemistry on sections of frozen tumors.

The mice are raised at the animal center of the INSERM U892 unit (under the control of AFSTAL, French association of Sciences and Techniques of Laboratory Animal).

Raji cells and IMR32 cells (2.5 × 10 ⁵ cells) diluted 1: 2 in Matrigel highly concentrated in growth factors (BD Biosciences, Bedford, USA) were injected subcutaneously into the flank of mice. Balb / c @ BYJ Rj 12 weeks old (January, St Berthevin, France). The mice were sacrificed by elongation as soon as small blood capillaries appeared on the xenografted tumors. Directly after excision, the tumors were pre-immersed in the isopentane and then immersed for a few seconds in liquid nitrogen to be stored at -80 ° C. 5 μπι frozen sections made with the Leica CM-1900 cryostat (morphology platform of IFR 26, Inserm, Nantes) were deposited on Starfrost blades (Knittel Glàser, Braunschweig, Germany). They are then treated by heating at room temperature for 30 minutes, fixed in cold acetone for 10 minutes, air-dried for 30 minutes and washed in PBS, before proceeding to the staining procedure. First, non-specific binding sites are blocked with the MOM Kit blocking reagent (Vector Laboratories, Burlingame, CA, USA). The frozen sections are incubated successively with 50 μg / ml of antibody anti-Gb3 or mouse IgM isotype control antibody (clone 1 1E10) for 1 h at room temperature, and then incubated after several washes at PB S, with biotinylated goat anti-mouse IgM secondary antibody, diluted at 1: 100. Detection is performed using a chromogenic substrate. The sections are incubated for 30 minutes with VECTASTAIN ^® Elite ABC reagent (Vector Laboratories). The peroxidase activity is detected with the DAB ^® Menarini Kit substrate (Rungis, France) diluted 1:50, resulting in the deposition of a brown pigment. The sections are then counter-stained with hematoxylin and observed at 40X magnification with a DM IRB (Leica) microscope. 1.9. Obtaining anti-Gb3 antibody 3E2 (IgM.

Mice immunization protocol and somatic hybridization.

Five Balb / c @ BYJ Rj mice, 6 weeks old, were immunized with HMVEC-L cells previously co-cultured with T84 human colonic adenocarcinoma cells via membranes allowing the exchange of soluble factors between the cells. two cell compartments (Gaugler et al., 2007). HMVEC-L cells were seeded in 6-well plates in an amount of 5 10 ⁴ cells / cm ² in their culture medium EGM -2 ^® MV. Three days later, while the endothelial cells were still in proliferation, the T84 cells (1 × 10 ⁵ cells / cm ² ) were seeded in their DMEM: F 12 culture medium, on the porous membrane filter (pores of 0, 4 mm) of a Transwell 6-well Transwell-Clear insert (Corning, The Netherlands) for 72 hours. FDVIVEC-L cells were then trypsinized, washed in PBS, fixed with 4% paraformaldehyde and stored at 4 ° C in incomplete Freund's adjuvant (Sigma, St. Louis, USA). In total, the mice were injected intraperitoneally with five injections at 0, 8, 10, 32, and 60 weeks of 2, 1-2.8 10 ⁶ HMVEC-L cells previously co-cultured, as well as one dose. at 65 weeks of 1.5 × 10 ⁶ cells (stored in PBS without adjuvant), 6 days before sacrifice. The kinetics of the humoral response of the mice was analyzed during the immunization period by ELISA tests carried out on desiccated HMVEC-L cells but also on desiccated HMEC-1 cells in order to envisage the use of these transformed cells. for antibody screening studies. Once the anti-antibody response was positive, mouse splenocytes were fused to SP2 / 0 murine myeloma cells at a 5/1 fusion ratio: 2.5 splenocytes were fused with polyethylene glycol (PEG 1500, Sigma) at 0.5 × 10 ⁸ murine myeloma cells SP2 / 0 (cultured in RPMI 1640). At the end of the fusion, the cells were taken up in 200 ml of RPMI 1640 medium supplemented with 20% inactivated SVF, 2 mM L-glutamine, hypoxanthine-aminopterin-thymidine HAT IX medium (Sigma-Aldrich) and inoculated at 100 μΐ into 35 microplates of 96 wells. After 15 days, the first hybridomas were visible and could be amplified in 12 well microplates in RPMI 1640 medium supplemented with 20% inactivated serum, 2 mM L-glutamine, 100 IU / ml penicillin and 100 μg / ml of streptomycin. Supernatants from these hybridomas were all removed for screening and stored at 4 ° C and all hybridomas were taken up in FCS containing 10% dimethylsulphoxide, to be stored at -80 ° C. The pattern of antibody generation is summarized in Figure 2. Cloning and purification of antibodies.

Cloning of antibodies. The selected hybridomas were cloned by the limiting dilution method (Fig. 3). Each clone obtained was then tested again by ELISA using the anti-IgG + IgM detection antibodies, then the isotype of the positive monoclonal antibodies was determined using anti-isotype detection antibodies (μ, γΐ, y2a). , y2b and γ3) and anti-light chain (κ and λ).

Purification of antibodies by affinity chromatography. The selected antibodies were then purified according to the protocol of the supplier using a 5 ml column of L protein (Pierce, Rockford, USA) which has an affinity for the immunoglobulin "kappa" light chains. The supernatant of the hybridomas was diluted with a solution of 0.1 M Tris-HCl, pH 7.8 (dilution 1: 1), filtered through 0.22 μπι and then slowly passed over the column. Nonspecifically bound proteins were removed with 5 volumes of PBS. The antibodies are collected by elution with 0.1 M glycine-HCl buffer solution, pH 2.5 and the eluate is neutralized with 0.1 M Tris buffer solution, pH 9. The eluates are selected after measurement of the absorbance at 280 nm (Nanodrop) and pooled to be dialysed in PBS, filtered through 0.22 μπι and stored at 4 ° C. The degree of purity of the antibodies is assessed by an electrophoretic analysis under denaturing conditions (SDS-PAGE) that is reducing with 12% acrylamide (375 mM Tris pH 8.8, SDS 0.1%, APS 0.1%, TEMED), which is non-reducing with 6% acrylamide. The migration on the gel is carried out with 5 μg of antibody per condition, at 90 V for 45 h in Tris-Glycine-SDS buffer (Biorad, Hercules, USA). The electrophoretic profile is detected by staining with Coomassie Blue.

1.10. Determination of the affinity constants of the anti-Gb3 antibody 3E2.

The antibodies are first labeled with iodine 125 in the presence of an oxidant, iodogen. In a glass tube whose wall has been treated with 10 μg of iodogen in a volume of 100 μΐ, 200 μg of anti-Gb3 antibody 3E2 is deposited, and the tube is then supplemented with 100 μl of phosphate buffer at 0, 1M, pH 7.2. Iodine 125 (Perkin Elmer, Waltham, USA) is added so as to obtain an activity of 1 μCi / μg of antibody, ie 200 μθ of activity. The reaction mixture is incubated for 30 minutes with gentle stirring at room temperature. After incubation, the purity of the radiolabelled antibody was monitored by thin layer chromatographic analysis. For this, 2 μl of the reaction mixture were deposited on an ITLC-SG strip (PALL Science, Iris Parkway, USA). By migrating in 10% trichloroacetic acid, the free ¹²⁵ I separates from ¹²⁵ I bound to the antibody. It is thus possible to evaluate the level of labeled antibody which must be greater than 90% and which can be reinforced by purification on a NAP-5 TM column (GE Healthcare, Waukesha, USA) and be evaluated again by thin layer chromatography. ITLC-SG.

Once radiolabeled 3E2 antibody, the number of Gb3 antigenic sites was determined according to the Scatchard technique, on Raji lymphoma lines and on HMEC-1 endothelial cells. For HMEC-1 cells, the number of antigenic sites was evaluated in particular after 24 h of incubation in depleted medium (MCDB 131 medium supplemented with 0.1% decomplemented FCS, 2 mg / ml of hydrocortisone, 2 mM of L-glutamine, 100 IU / ml penicillin and 100 mg / ml streptomycin) and after 24 h incubation in complete medium (MCDB 131 medium supplemented with 15% FCS decomplemented, 10 ng / ml EGF, 2 mg ml of hydrocortisone, 2 mM L-glutamine, 100 IU / ml penicillin and 100 mg / ml streptomycin). Briefly, in a 96-well round bottom plate, 2.5 × 10 ⁵ cells were exposed to the antibodies previously radiolabelled, and diluted half to half, from 555 nM to 0, 13 nM (n = 3). In parallel, wells containing a mixture of 500 nM of cold antibody with 5 nM radiolabelled antibody, diluted four half times in half, are used to determine nonspecific binding. The reaction mixture was incubated at 4 ° C. with stirring for 1 h, then 50 μl of this mixture was added to Scatchard tubes (Sarstedt, Numbrecht, Germany) containing 200 μl of a 10% separation solution. paraffin and 90% dibutyl phthalate. After centrifugation at 15,000 g for 3 minutes, the tubes are immersed in liquid nitrogen, cut into two portions and immediately placed in hemolysis tubes for radioactive counting. The lower end containing the cells constitutes the bound fraction and the upper end containing the supernatant constitutes the unbound fraction. Measurements are made at the γ counter (1480 Wizard 3, Perkin Elmer, Finland) and the results are analyzed with GraphPad Prism software (GraphPad Software Inc., San Diego, USA). 1.11. Biological properties of the 3E2 antibody.

Cell viability test at MTT.

 The MTT assay (Mosmann, 1983) is based on the transformation of MTT (3- (4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide) into blue crystals of formazan by a mitochondrial enzyme. succinate dehydrogenase. The formazan crystals formed by the cells are solubilized chemically and detected spectrophotometrically at a wavelength of 570 nm. This test is used to compare the viability of control cells with that of cells treated with molecules.

Briefly, the assay is performed in 24-well plates with cells seeded at a density of 2 × 10 ⁴ cells / cm ² . The adherent cells, such as the HMEC-1 and NXS2 cells, are incubated for 12 hours after their inoculation with the primary antibodies previously diluted in 300 μl of complete medium to give a final concentration of 100, 60, 40, 20, 10 and 0. μg / ml (n = 3). The Raji cells in suspension are inoculated directly into the culture medium containing the antibodies (n = 3). They are then incubated for 24 hours at 37 ° C., 5% of C0 ₂ . 500 μg / well of the stock solution of MTT (Roche, Mannheim, Germany) is then added and the cells are kept at 37 ° C. for 4 hours, in order to allow the synthesis of crystals of formazan. 100 μl / well of the solubilization solution (10% SDS in 10 mM HCl) is added. After overnight incubation at 37 ° C, the plates are read with a Multiskan reader (Thermo Electron, Illkirch, France) by measuring the absorbance at 570 nm and, for the background noise, at 650 nm. Video-kinetic study.

The HMEC-1 cells were seeded in a 24-well microplate at 2 × 10 ⁵ cells / cm ² and then maintained for 12 hours in their complete medium at 37 ° C., 5% CO ₂ . The 3E2 antibodies and the isotype control were then diluted to 20 μg / ml in complete culture medium and incubated with the cells in triplicate. Immediately after incubation, digital images were recorded for 24 to 72 h in order to produce an accelerated film (Leica, DMI6000B / PICell platform). Observations were on the number of cells per time interval, the rate of dividing cells per time interval, and the cell division time (mean of observations from three different wells). Tests of the aortic rings.

C57bl6 mice aged 4 weeks were provided by the Janvier breeding farm (St Berthevin, France). Abdominal aorta were isolated when the mice were 6 to 8 weeks old. They were then cleaned in complete MCDB 131 medium and cut into cross sections 0.5-1 mm wide. These rings were then deposited at the bottom of the wells of a 96-well microplate previously treated with 30 μl of Matrigel (ECM Gel, Sigma-Aldrich, St. Louis, USA) diluted half in PB S. The rings were then were covered with 20 μl of undiluted Matrigel and then kept for 5 days at 37 ° C., 5% of C0 ₂ in complete medium containing 0, 20 and 40 μg / ml of antibody. The culture medium with or without antibodies was renewed every 2 days. After 5 days of incubation, the formation of vascular buds was observed by microscopy (IRB DM microscope, Leica). In order to objectively measure the formation of these buds, we asked a 5-person jury to independently participate in a blind notation of images of the aortic rings recorded at 10X magnification and asked them to assign each image a score, based on the number, length and density of microvessels formed (scale from 0 to 5, the minimum value of 0 corresponding to the absence of buds), in order to evaluate the anti-angiogenic activity of the antibodies.

 Evaluation of the cytotoxic activity in the presence of complement.

The CDC activity is measured by flow cytometry using a solution of propidium iodide (PI) capable of incorporating the DNA of the cells. The cells are trypsinized, taken up in culture medium and then placed in conical bottom 96-well plates (Nunc, Denmark) at a rate of 0.2 10 ⁶ cells / well. The plates are then centrifuged to remove the supernatant, and the cells are incubated with the antibodies to give a final concentration of 0.1 to 10 μg / ml. Fresh filtered serum is collected 24 h after the blood sample in the absence of anti-coagulant. A fraction is decomplemented at 56 ° C for 45 min, filtered again and stored at 4 ° C. Human serum is then added which represents 1: 5 of the final volume, either in its decomplemented form or in its non-decomplemented form. The plates containing the cells, the antibodies and the serum are then incubated for 2 h at 37 ° C., 5% CO ₂ . The cell pellets are then collected by brief centrifugation of the plate and the cells taken up with 0.6 μg of a solution of propidium iodide in a volume of 100 μΐ. After an incubation of 30 min at 37 ° C in the dark, the cells are analyzed directly using the cytometer at a rate of 1.10 ⁵ cells per analysis. 1.12. Nucleotide sequencing of variable regions of heavy and light chains of anti-Gb3 antibodies.

Extraction of the total RNA.

The total RNA of the mouse hybridomas was extracted according to the method described by Chomczynski and Sacchi (Chomczynski and Sacchi, 1987) with the RNAble ^® reagent (Eurobio, Les Ulis, France). The cells are homogenized in a denaturing solution containing guanidinium isothiocyanate and phenol. The nucleic acids are denatured, the proteins dissociated thanks to the formation of complexes between RNA and guanidine isothiocyanate which makes it possible to break the hydrophilic interactions between the DNA and the proteins. Thus the DNA and proteins are extracted from the aqueous phase while the RNA remains in this phase. The cell pellet of hybridomas (10 10 ⁶ cells) is washed twice in PB S and then taken up in 2 ml of RNAble. The cells are lysed by repeated pipetting to dissociate the nucleoprotein complexes. After the addition of 0.2 ml of chloroform, the homogenate is mixed vigorously for 15-30 seconds. This step makes it possible to separate, by differential solubilization, the nucleic acids (the RNAs being insoluble in phenol) from the proteins. The mixture is incubated on ice for 5 minutes and then centrifuged for 15 minutes at 12,000 g at 4 ° C. The aqueous phase containing the RNA is removed and transferred to a microtube. When precipitated with 2 ml of isopropanol for 10 minutes at room temperature, they are recovered by centrifugation at 12,000 g for 5 min at 4 ° C. and then washed twice with 1 ml of a 75% ethanol solution (v). / v), air-dried and taken up in 15 μl of H ₂ O-DEPC 0.1% (water previously treated against ribonucleases by diethyl pyrocarbonate). The RNA concentration is determined spectrophotometrically by measuring the absorbance at 260 nm of an aliquot as well as its purity at 280 nm by the ratio of the absorbances at 260 nm / 280 nm for a value of about 2. The RNA thus extracted are either stored at -80 ° C or directly treated with DNAse.

Treatment with DNAse 1.

In order to eliminate all traces of genomic DNA, the DNAse is treated according to the protocol of the supplier. The reaction is carried out in a final volume of 10 μl comprising 2 μg of total RNA, a 20 mM Tris-HCl buffer solution, pH 8.4, 2 mM MgCl ₂ , 50 mM KCl, and 1 U of DNAse 1. (Sigma, St-Quentin-Favier, France). After incubation for 15 minutes at room temperature, the DNAse is inactivated by the addition of 1 μl of a 25 mM EDTA solution, followed by incubation at 65 ° C. for 10 minutes. The treated RNAs can also be stored at -80 ° C, or be directly subjected to a reverse transcription step.

Synthesis of the complementary DNA.

Retro-transcription followed by polymerization makes it possible to stabilize the single-stranded RNA in complementary DNA (cDNA). The cDNAs are synthesized from 0.2 μg of RNA in a final volume of 40 μΐ. According to the supplier's protocol (Invitrogen), the total RNAs are denatured by incubation at 65 ° C. for 5 min. 500 ng of poly- (dT) i2-is oligonucleotides and 1 nM of each dNTP in 12 μl of H _{2 0} wg QSP. After incubation, the RNAs are placed rapidly on ice to prevent refolding. their secondary structures, and then are added 14 μΐ reaction medium consisting of reaction buffer (250 mM Tris-HCl, pH 8.3, 375 mM KC1, 15 mM MgCl _2), 0.2 μΜ DTT, 40 U of RNaseOUT ^® and 200 U of M-MLV (Moloney Murine Leukemia Virus Reverse Transcriptase). After incubating for 50 minutes at 37 ° C., the reaction is stopped by incubation for 15 minutes at 70 ° C. The cDNAs thus obtained can be stored at -20 ° C. or directly amplified by PCR.

Amplification of V _H and V _L DNAs by PCR.

The PCR gene amplifications of the V _H and V _L DNAs are carried out using specific oligonucleotide pairs described in Table 4 (Clackson et al., 1991, Lefranc and Lefranc, 1997). On the 5 'side, the oligonucleotides hybridize to the FR1 region and to the 3' side, to the different ½ or J _L segments.

Table 4. Primers used for the amplification of the variable regions of the V _H and V _L gene segments of the mouse monoclonal antibodies.

 A - VH primers

 VH back primers (22-sea)

 VH1 A back 5 '- AGG TGC AGC TTC AGG AGT CAG G - 3' SEQ ID NO: 71 VH1 B back 5 '- AGG TGC AGC TAG AGG AGT AGG G - 3' SEQ ID NO: 72 VH2 A back 5 '- AGG TCC AGC TGC AGC AGT CAT G - 3 'SEQ ID NO: 73 VH2 B back 5' - AGG TCC ACC TGC AGC AGC CTG G - 3 'SEQ ID NO: 74 VH2 C back 5' - AGG TCC AGC TGC AGC AGT CTG G - 3 'SEQ ID NO: 75 VLB A back 5' - AGG TGA AGC TGG TGG AGT CTG G - 3 'SEQ ID NO: 76 VLB B back 5' - AGG TGA AGC TTC TCG AGT CTG G - 3 'SEQ ID NO: 77 VLB C back 5 '- AAG TGA AGC TTG AGG AGT CTG G - 3' SEQ ID NO: 78 VLB D back 5 '- AAG TGA TGC TGG TGG AGT CGT G - 3' SEQ ID NO: 79 VH5 back 5 '- AGG TTC AGC AGC TCC AGT CTG G - 3' SEQ ID NO: 80

 JH forks (24-mer)

JH1 for 5 '- TGA GGA GAC GGT GAC CGT GGT CCC - 3' SEQ ID NO: 81 JH2 for 5 '- TGA GGA GAC TGT GAG AGT GGT GCC - 3' SEQ ID NO: 82 JH3 for 5 '- TGC AGA GAC AGT GAC CAG AGT CCC - 3 'SEQ ID NO .: 83 JH4 for 5' - TGA GGA GAC GGT GAC TGA GGT TCC - 3 'SEQ ID NO: 84 B - Primers VK

 VK back primers (24-sea)

 VKA back 5 '- GAT GTT TTG ATG ACC CAA ACT CCA - 3' SEQ ID NO: 85 VKB back 5 '- GAT ATT GTG ATA ACC CAG GAT GAA - 3' SEQ ID NO: 86 VKC back 5 '- GAC ATT GTG CTA / G ACC CAA / G TCT CCA - 3 'SEQ ID NO: 87 VKD back 5' - GAC ATC CAG ATG AC (N) CAG TCT CCA - 3 'SEQ ID NO: 88 VKE back 5' - CAA ATT GTT CTC ACC CAG TCT CCA - 3 'SEQ ID NO: 89 VKF back 5' - GAA AAT GTG CTC ACC CAG TCT CCA - 3 'SEQ ID NO: 90

 Primers JK for (24-sea)

 JKI for 5 '- CCG TT GAT CAG CTT GGT GCC - 3' SEQ ID NO: 91

 JK2 for 5 '- CCG TTT TAT TTC CTG CTT GGT CCC - 3' SEQ ID NO: 92

 JK3 for 5 '- CCG TTT TAT TTC CAA CTT TGT CCC - 3' SEQ ID NO: 93

 JK4 for 5 '- CCG TTT CAG CTC CAG CTT GGT CCC - 3' SEQ ID NO: 94

The PCR amplification conditions specific to each hybridoma are described in Table 5 (Lefranc and Lefranc, 1997).

Table 5. Experimental Conditions for PCR Amplification of V _H and V DNAs

V _L of mice.

Type Primers Primers Amplification conditions VH or VK JH or JK PCR

 back for

 VH to VH1A, VH1 JH1, JH2, 94 ° C, 50s

 B JH3, JH4 60 ° C, 90s

 72 ° C, 90s b VH2A, VH2 JH1, JH2, 94 ° C, 50s

 B, JH3, JH4 60 ° C, 120s

VH2 C, VH5 72 ° C, 120s c VH3A, VH3 JH1, JH2, 94 ° C, 50s

 B JH3, JH4 65 ° C, 90s

 72 ° C, 90s d VH3 C, VH3 JH1, JH2, 94 ° C, 60s

 D JH3, JH4 55 ° C, 120s

 72 ° C, 120s

 VK e VKA, VKB, JKI, JK2, 94 ° C, 90s

 VKE, VKF JK3, JK4 60 ° C, 90s

 72 ° C, 90s f VKC, VKD JKI, JK2, 94 ° C, 50s

 JK3, JK4 65 ° C, 90s

72 ° C, 90s Table 5 proposes four different amplification conditions for the amplification of the heavy chain variable regions (a, b, c and d) and two different conditions for the amplification of the light chain variable regions (e and f). . For each of the conditions, there are primer mixtures that hybridize on the 5 'side (primers back) and 3' (forward primers). In order to carry out the sequencing of the variable regions of the heavy and light chains, we first identified by PCR the condition (a, b, c and d, e and f) which allowed an optimal amplification of the V _H and V _L DNAs. then, we have, in a second time, identified by PCR the primer back which allowed the best amplification.

All the PCR reactions were carried out in a final volume of 25 μl containing 1 μl of cDNA, each primer (10 μM), each dNTP (10 nM), MgCl ₂ (62.5 nM), reaction buffer, 5 U GoTaq ^® DNA polymerase (Promega, Charbonnières-les-Bains, France) and VH ₂ mQ 25 μΐ qs. Thirty cycles of amplification (Gene Amp ^® PCR-System 2700, Applied Biosystems, Courtaboeuf, France) were performed after a denaturation step of 94 ° C for 5 min. Each of the cycles included a step of 10 minutes at 72 ° C to complete the synthesis of the DNA strands. The amplification products were then visualized on a 1% agarose gel (QA-Agarose-TM, MP Biomedicals, Strasbourg, France) containing 0.1% GelRed ™ (FluoProbes, Montlucon, France).

 For each hybridoma, three independent RNA extractions and three independent PCR amplifications were performed in order to perform three independent sequencing of the heavy and light chain variable regions. The amplification conditions chosen for each hybridoma are listed in Table 5.

Nucleotide sequencing

The nucleotide sequencing was performed by the platform of "Sequencing DNA and Genotyping" of the IFR 26 of the University of Nantes, according to the method of Sanger (Sanger et al., 1977) with an AB3730 sequencer of 48 capillaries (Applied Biosystems). ). For each sequencing, 1 μl of appropriate primer back (Table 6) and 5 μl of PCR product V _H OR V _{L were used} . The PCR products were purified by the ExoSAP-IT process (Amersham GE Healthcare) which allows, under the action of a enzyme degrade single strand fragments less than 100 bp and eliminate nucleotide primers and excess dNTPs.

Table 6. Back primers used for nucleotide sequencing.

The three independent sequencing made from three different RNA extractions enabled us to compare and determine the exact sequences of the heavy and light chain variable regions of each hybridoma. The nucleotide sequences obtained were then aligned with data from the International Immunogenetics Information System® (IMGT) database, http: //www.imgt.org, in order to study the repertoire of V _H and V _L genes used in mice. .

EXAMPLE 2. Search for glycolipidic markers of endothelial cells during angiogenesis

In order to look for glycolipidic markers of intratumoral endothelial cells, we studied the glycolipid polymorphism of these cells according to different stimuli. The HMEC-1 cells were transformed with SV40 virus, so as to make them more easily used in culture while allowing them to retain their microvascular endothelial cell characteristics. They were then placed under different culture conditions close to the stimuli that the endothelial cells can undergo in a tumor microenvironment, which involves the sending of chemical signals by the tumor as growth factors that will induce their proliferation. We have thus compared the glycolipid expression profiles of proliferating or confluent HMEC-1 cells, incubated in complete medium or depleted medium (in serum and growth factor), or incubated or not with a pro-angiogenic factor, sphingosine-1-phosphate. The estimation of the proportions of the different glycolipids could then be carried out by densitometry (ImageQuant ⁵ ' ² ).

2.1. Results: Demonstration of a glycolipid expression polymorphism of endothelial cells HMEC-1

We sought to demonstrate an expression polymorphism of endothelial cells according to different conditions, to look for the presence of new glycolipid markers and / or the presence of overexpressed glycolipids. We thus compared the glycolipid expression of HMEC-1 cells according to their culture state (glycolipids extracted from cells in growth phase or at confluence), according to their physiological state (quiescent cells in depleted culture medium, and proliferative cells in complete culture medium), as well as under the effect of pro-angiogenic factors (sphingosine-1-phosphate). Indeed, if there are glycolipidic markers of proliferating endothelial cells, then these markers are targets of interest in a perspective of anti-angiogenic intratumoral therapy.

Glycolipid expression profile of proliferating or confluent HMEC-1 cells (by HPTLC).

 Figure 1A shows the HPTLC analysis of glycolipids extracted when the HMEC-1 cells were in the growth phase (lane 1), or at confluence (lane 2). The cells were also analyzed by densitometry.

Depending on whether the HMEC-1 endothelial cells are in the proliferating phase two or three days after seeding, or having reached confluence after five days, a difference in glycolipid expression is observed, especially for the more apolar complex gangliosides. (No. 1 and 2) but also for glycolipid Gb3 (No. 12, Fig. 1). Densitometric analysis showed that the Gb3 content is almost twice as high for proliferating cells as for confluent cells (value equal to 1.716 ± 0.069, two independent analyzes). Glycolipid expression profile of HMEC-1 cells following the culture medium composition.

FIG. 1B shows the HPTLC analysis of HMEC-1 cells which have been seeded at 20,000 cells / cm ² and then incubated for 24 hours, as soon as the subconfluence state is reached, in depleted culture medium (lane 1) or in complete culture medium (lane 2). The cells were also analyzed by densitometry.

 Depending on whether the cells are cultured in MCDB-131 medium with 15% FCS (complete medium) or 0. 1% FCS, without EGF (depleted medium), it is observed by FIPTLC that there is a difference of expression for the most apolar complex gangliosides (No. 1, 2 and 3, Fig. 2) and for the neutral glycolipid Gb3 (No. 12). Densitometric analysis shows that the Gb3 content of the cells incubated in complete medium is almost twice that of the cells incubated in depleted medium (value equal to 1.512 ± 0.061, two independent analyzes).

Glycolipid expression profile of HMEC-1 cells treated or not treated for 24 h with sphingosine-1-phosphate (by HPTLC).

Figure 1C shows the FIPTLC analysis of HMEC-1 cells that were seeded at 20,000 cells / cm ² and cultured in complete medium, until the subconfluency state is reached, and then incubated for 24 h in depleted medium supplemented with either sphingosine-1-phosphate 1 μl (lane 2) or its diluent, PET (lane 1).

 This HPTLC analysis shows that there is a difference of expression for Gb3 (No. 12, Fig. 1C), the Gb3 content of cells incubated 24 h with S 1P is almost 1.5 times that of cells incubated with polyethylene glycol (value equal to 1.271 ± 0.020, two independent analyzes).

2.2. Conclusion

For human microvascular endothelial cells HMEC-1 transformed with SV40, we were able to develop a reproducible glycolipid analysis method that allowed to consider their characterization as well as the study of glycolipid expression polymorphism. The glycolipids are derived from the aqueous phase of a partition DIPE / 1-butanol, 60:40 v / v, NaCl 0, 1% and revealed by orcinol, which revealed 14 bands of variable intensity that correspond to at least 7 molecular species. We have thus been able to identify the presence of a neutral glycolipid which is not revealed by resorcinol / HCl but which is recognized by a commercially available anti-Gb3 rat antibody. This glycolipid Gb3 is in particular present in the form of a doublet corresponding to two types of ceramide fatty acid chains.

 Analysis of the expression of glycolipids of HMEC-1 cells according to different stimuli made it possible to highlight glycolipid expression modifications of HMEC-1 endothelial cells. An increase in the content of complex gangliosides, and in neutral glycolipid Gb3, is observed when the cells are in the exponential phase of growth (FIG 1 A), in their complete medium (FIG 1B) but also an increase in the content of Gb3 when stimulated by the presence of a proangiogenic factor, sphingosine-1-phosphate (Fig. 1C). The Gb3 content can be twice as low when the cells are incubated in depleted medium (0, 1% FCS, without EGF). In the presence of depleted media, the cells enter the GO phase of the cell cycle, which is a quiescent non-divisional stage. We previously verified by cell counting that the number of FDVIEC-1 cells after 24 h incubation in depleted medium was almost identical to the number of cells before incubation, while the number of cells FDVIEC-1 after incubation for 24 h in complete medium, could almost double. This stage of non-division is reversible because under the influence of growth factors that order the cell to divide, they can then return to the Gl phase cell cycle to continue a normal progression of the cell cycle and proliferate.

In other words, the Gb3 content of the proliferating cells is greater than that of the quiescent cells. In the same way, when a proangiogenic factor such as sphingosine-1-phosphate is added to the depleted culture medium, the Gb3 content of the cells increases by 50%, whereas in complete medium it is twice as much. higher. We had also previously demonstrated in the laboratory by thymidine incorporation tests, that after 24 hours of incubation of FDVIEC-1 cells with depleted medium supplemented with 1 μl of S IP, the proliferation rate of the cells was 1 , 2 times higher compared to untreated cells, confirming the pro-angiogenic action of SIP (Bonnaud et al., 2007). The increase of the content in Gb3 more important in the presence of complete medium can be explained in particular by a synergistic action of growth factors present in the fetal calf serum of the complete medium.

EXAMPLE 3 Preparation of antibodies directed against endothelial cells stimulated by tumor cells

To generate mouse monoclonal antibodies against glycolipids of intratumoral endothelial cells, we chose to immunize Balb / c mice with endothelial cells that had previously been cultured in the presence of tumor cells in a co-culture model. mimicking the tumor microenvironment. In HMEC-1 endothelial cells, HPTLC revealed the presence of at least seven more or less expressed glycolipid molecular species, including a neutral glycolipid present in the form of a doublet corresponding to glycolipid Gb3. We have chosen to immunize the mice with whole cells, and not with a purified glycolipid, because we do not wish to privilege a particular glycolipid, or rule out the possibility of generating antibodies directed against antigens of a glycoprotein nature. Following immunization, all generated hybridomas were frozen, all supernatants stored at 4 ° C, and then two screening strategies were used to selectively select for antibodies against glycolipid antigens (ELISA). on desiccated endothelial cells).

3.1. Results

Obtaining antibodies

 Implementation of the co-culture model.

In order to generate new antibodies directed against glycolipid markers of endothelial cells, we first set up a co-culture model mimicking the tumor microenvironment (Fig. 2). HMVEC-L human primary lung endothelial cells were co-cultured in the presence of T84 human colonic adenocarcinoma cells through porous membranes allowing the exchange of soluble factors between the two compartments (6-well Transwell-Clear, 0.4 mm pores, Corning, the Netherlands). Once Isolated and fixed with paraformaldehyde, these endothelial cells were used to immunize five Balb / c mice.

 This co-culture system has been developed in the laboratory and has been previously described to study the influence of endothelial cell irradiation on non-irradiated T84 cells co-cultured with irradiated endothelial cells (Gaugler et al., 2007). . In order to simulate as much as possible the conditions of the tumor microenvironment, the endothelial cells were proliferating at the time when the adenocarcinoma cells were implanted three days later and were still proliferating at the time they were trypsinized and then 4% paraformaldehyde, for immunization.

Immunization of mice and somatic hybridization.

 Five Balb / c mice were immunized with these endothelial cells co-cultured and fixed with 4% PFA. Sera were collected before each injection and the kinetics of the mouse humoral response was analyzed by desiccated ELISA on both HMVEC-L primary endothelial cells and SV40-transformed HMEC-1 endothelial cells. . For the anti-IgM response, we used polyclonal biotinylated goat antibodies directed against the μ chain of mouse antibodies and for the anti-IgG response, polyclonal biotinylated goat antibodies directed against the (H + L) antibody fragments mouse.

 The immunoglobulin isotype analysis shows that this is an early μ response followed by a lower γ response. Indeed, for the anti-IgM response, antibody levels are detectable from the first immunization (week 8) and for the anti-IgG response, antibody levels are detectable after the second injection (week 10) . We also observe a cross-reaction of these antibodies on HMEC-1 endothelial cells that allows us to consider the use of these SV40-transformed cells for screening studies.

Screening of antibodies on HMEC-1 cells after somatic hybridization

Somatic hybridization yielded 1086 hybridomas. Their supernatants were removed for screening and stored at 4 ° C and these 1086 hybridomas were taken up in SVF containing 10% dimethylsulphoxide, to be stored at -80 ° C.

 Due to the cross-reaction observed with the antibodies present in the serum of the immunized mice, on SV40-transformed HMEC-1 cells, we could consider the use of these transformed cells for the screening studies. The generated hybridomas were selected using goat detection antibodies directed against both mouse IgM (μ) and IgG (H + L) and in two approaches.

 For the first approach, the hybridoma supernatants were tested by ELISA on desiccated HMEC-1 cells in order to preferentially screen for antibodies directed against antigens of glycolipidic nature. The antibodies selected by ELISA were then screened by flow cytometry and the nature of the antigens was determined by immunolabeling glycolipids of HMEC-1 cells separated by HPTLC (iTLC). For the second approach, all the antibodies preselected by the ELISA were analyzed by immunolabeling of glycolipids of HMEC-1 cells separated by HPTLC (iTLC).

First antibody screening strategy

From the 1086 hybridomas, we pre-selected 139 hybridomas by ELISA on desiccated HMEC-1 cells. In addition to being simple, rapid, and having the advantage of selecting glycolipidic patterns that are resistant to desiccation, this technique also makes it possible not to favor a particular molecular species, since the antibodies are screened on whole cells. instead of purified glycolipids. The 139 hybridomas pre-selected by ELISA were then screened by flow cytometry. This method makes it possible to select the antibodies directed against membrane antigens on the surface of living cells. Indeed, the ELISA technique may have certain disadvantages. During desiccation, the cell membranes are damaged and the antibodies selected by ELISA can be directed against intracellular antigens while we are looking for antigens present on the surface of the membranes. By flow cytometry we were able to isolate 13 hybridomas and in order to determine the nature of the recognized antigen, we immunolabeled HPTLC-separated glycolipids with these hybridoma supernatants (data not shown).

 The supernatants of the 13 hybridomas recognize for 6 of them (25C10 (1), 1E10 (3), 3E2 (4), 15C1 1 (5), 16G8 (10) and 22F6 (12)), the same antigen of glycolipidic nature present in the form of a doublet that migrates above GM1, previously identified as Gb3.

 Immunolabeling shows that Gb3 is present as a doublet corresponding to two types of ceramide fatty acid chains. It is found that some antibodies recognize both bands of Gb3 (antibodies 3E2 and 15C11), others recognize rather the upper band (antibody 22F6) and others the lower band (antibodies 25C10, 1 1E10 and 16G8). It has already been shown that the composition of the ceramide fatty acid chain can influence the conformation of a glycolipid and thus modify the reactivity of the antigen for its antibody.

 In order to verify the specificity of the antibodies for Gb3, we performed immunostaining of a standard mixture of neutral glycolipids comprising GalCer, LacCer, Gb3 and Gb4, with the supernatant of the hybridoma antibody 3E2 which present, among all anti-Gb3 antibodies, the best iTLC immunostaining chromatographic profile and the best flow cytometry fixation profile. The results show that this antibody recognizes only the Gb3 present in the form of a doublet without cross reaction with the Gb4 globoside, or with GalCer and LacCer.

The 13 selected hybridomas were cloned by limit dilution, again tested by ELISA on desiccated HMEC-1 cells, and their isotype was determined by a new ELISA using anti-μ, γΐ, y2a, y2b and γ3 antibodies. and their type of light chain with anti-κ and λ antibodies. The hybridomas selected by this first screening approach are listed in the following Table 7. This table notably presents the isotypes of each of the selected antibodies as well as the name of the corresponding monoclonal antibody. Table 7. Hybridomas selected by the first screening strategy.

Second strategy of screening antibodies.

 For the second screening strategy, we performed immunolabelings on the glycolipids of HMEC-1 cells separated by HPTLC with the 139 hybridoma supernatants pre-selected in ELISA with anti-IgG (H + L) + detection antibodies. anti-IgM (μ). New hybridomas (15 in number) could be selected by this second approach, including 7 anti-Gb3 antibodies and other antibodies directed against antigens of a glycolipidic nature undergoing characterization.

Immunolabeling on HPTLC is a sensitive detection method that can directly detect purified glycolipids extracted from about 10 ⁶ to 30 6 ⁶ cells while in flow cytometry, whole cells are analyzed at a rate of 0. , 4 10 ⁶ cells. Immunostaining on HPTLC would thus make it possible to detect antibodies that bind to minority glycolipids which, due to their low content, could be undetectable by flow cytometry. In addition, at this stage of the screening, we use uncloned hybridoma supernatants whose antibody concentration is not known. These hybridomas can only contain very low concentrations of antibodies of interest (if the hybridoma is a bad producer, if there is a mixture of hybridomas). Immunolabeling on HPTLC would thus detect anti-Gb3 antibodies present in small amounts in the supernatants tested.

 The hybridomas were cloned by limiting dilution and isotyped by ELISA using anti-isotype detection antibodies (μ, γΐ, y2a, y2b and γ3) and detection antibodies directed against the light chains (K and λ ).

 Apart from the antibodies whose isotype could be determined by this technique

(like 12E10 isotype γΐ antibody, or 21D4 and 14C11 isotype μ antibodies), some of them were not detectable with anti-isotype antibodies (μ, γΐ, y2a, y2b, γ3). We assume that these antibodies could be light chain dimers known to be secreted by myeloma cells since they can only be detected with secondary anti-IgG (H + L) antibodies and secondary antibodies specific for the chain. slight κ.

 The hybridomas selected by this second screening approach, their isotype and the name of the corresponding monoclonal antibodies are listed in Table 8 below.

 Table 8. Antibodies selected by the second screening strategy.

Purification of the antibodies.

Outside the 7G10 antibody. 1 e of light chain λ, directed against a glycolipid undergoing characterization and obtained during the second screening, all the κ light chain antibodies could potentially be purified by affinity chromatography on a protein L column. The mouse monoclonal antibody 1A4 specific for Gb3 _; supplied as ascites by J. Wiels, does not bind to the L-protein column and was purified with an MBP mannan column (Pierce).

After purification, the purity of the antibodies was analyzed by SDS-PAGE. This technique makes it possible in particular to check whether the antibodies have undergone denaturation during the elution step of the antibodies in acid buffer consisting of 0.1 M glycine-HCl, pH 2.5. The SDS-PAGE profile of the purified antibody 3E2 is shown in Figure 4. The monoclonal antibody 3E2 is pure and undenatured. The migration of the monoclonal antibody 3E2 on a polyacrylamide gel under denaturing conditions allows us to visualize, under reducing conditions, three bands corresponding to the heavy (H) and light (L) chains and to the characteristic J segment of μ-type immunoglobulins. Due to its high molecular weight, equal to 900,000 Da, IgM can not migrate in the 6% acrylamide gel under non-reducing conditions. Analysis of the specificity of purified anti-Gb3 antibodies by affinity chromatography.

Anti-Gb3 3E2 Antibody. C. 25C10.a. 16G8.h. 14Cl l. ld and 15C7.2e.

 Fig. 5 shows the cell labeling in flow cytometry of the purified anti-Gb3 monoclonal antibodies selected by the first screening strategy (3E2.c, 25C 10.a, 16G8.h) or by the second screening strategy (HC ll. and 15C7.2e). Tagging was performed on HMEC-1 cells, Raji lymphoma cells for which Gb3 is a marker, and NXS2 cells of murine neuroblastoma that do not express Gb3, which we had previously verified with an ELISA on NXS2 cells desiccated using anti-Gb3 antibody 38.13 (data not shown). For 3E2 antibodies. C, 25C10.a, 16G8.h and 14Cl. ld (A, B, C and D) we used secondary antibodies specific for the μ chain of mouse antibodies and for the 15C7.2e (E) antibody which would be a light chain dimer we used secondary antibodies directed against IgG (H + L).

 There is no binding on NXS2 cells regardless of the anti-Gb3 antibodies tested. Antibodies from the second screen appear to have a lower affinity for Gb3 than antibodies from the first screening as shown by the percentages of positive cells obtained with the 14C1 antibody. ld and 15C7.2e on HMEC-1 cells (respectively 25.3 and 11.7%) and with the 14C1 1 antibody. ld on Raji cells (10.0%).

When the labeling is carried out with the antibody 3E2.c, 68.2% of the HMEC-1 cells and 87.9% of the Raji cells are positive. Unlike Raji cells, the labeling of HMEC-1 cells is heterogeneous and shows that some cells strongly express Gb3 and others express it more weakly. For the 25C10.a and 16G8.h antibodies resulting from the first screening, the HMEC-1 cells are more weakly homogeneously labeled (respectively 50.7% and 15.4%) and the Raji cells are labeled at 87, 1% and 51, 1%. Immunolabeling on HPTLC of HMEC-1 cells showed that the 16G8.h and 25C 10.a antibodies, instead, recognized the lower band of the Gb3 doublet (data not shown) with a lower content than the upper band, which could explain the lower binding of these antibodies by cytometry. Fixing purified antibody 13D2.2d (data not shown) against a glycolipid antigen undergoing characterization, could not be detected in flow cytometry probably because of the small amount of this glycolipid in HMEC-1 cells, or because of poor affinity of the antibody.

 12E10.1c Antibody.

 In the second screening procedure, we selected the 12E10.1c antibody of IgG1 isotype, which was fixed on the Gb3 HMEC-1 cells by immunostaining on HPTLC. In fact, the IgG1 isotype antibodies are capable of significantly activating the complement pathway, which gives them interesting cytotoxic properties. The hybridoma was therefore cloned and purified and the antibodies were tested by flow cytometry on HMEC-1, NXS2 cells and Daudi cells, which are human Burkitt lymphoma cells expressing Gb3. We also compared the flow cytometric fixation of mAb 3E2 and 12E10. on human cells of mantle lymphoma JEKO-1.

 The results (not shown) indicate that the 12E10.1c antibody does not bind to HMEC-1 cells and binds very weakly on Daudi cells, probably because of its low affinity for Gb3. Furthermore, it binds more significantly on JEKO-1 cells whereas the monoclonal antibody 3E2 having a good affinity for Gb3 does not bind to these cells (data not shown). The 12E10.1c antibody probably recognizes a different Gb3 antigen that is present on JEKO-1 coat lymphoma cells and cross-reacts with Gb3, which would explain why it binds to Gb3 by HPTLC immunostaining of cells. HMEC-1.

3.2. Conclusion Five Balb / c mice were immunized with primary endothelial cells

HMVEC-L in proliferation that had previously been co-cultured in the presence of tumor cells through a membrane system allowing the passage and exchange of growth factors between cells, in order to mimic the tumor microenvironment. Since the immunogenic power of lipids is low, we first verified the serum kinetics of the antibody response by an ELISA test on desiccated cells, which allows antibodies to be screened preferentially. directed against desiccation-resistant antigens such as glycolipids. We obtained an antibody response against HMVEC-L primary cells and then a cross-reaction on HMEC-1 cells that allowed the use of these SV40 transformed cells as a cell model for further work.

This co-culture model had already been developed in the laboratory (Gaugler et al., 2007) and it had previously been demonstrated in a similar co-culture system that the presence of tumor cells amplified cell proliferation and migration. primary endothelial cells by acquiring phenotypic and genotypic modifications (Khodarev et al., 2003). The ELIS A technique, which uses desiccated cells, was developed in the laboratory and is based on preliminary studies with other cell types, such as the screening of monoclonal antibodies against membrane components of human mononuclear cells. Also, because of the fragility of the pancreatic tissue which is rapidly degraded, hence the need to stabilize it by fixation, mouse pancreatic cells have been used in a desiccated form for the detection and quantification of antibodies present in ELISA. in the serum of insulin-dependent diabetics. In our antibody generation strategy, we chose to immunize Balb / c mice with whole endothelial cells. Many anti-glycolipid antibodies have already been generated as a result of the immunization of whole tumor cells and in this way new molecular species of glycolipidic nature have been identified. Initially, fixation of HMVEC-L cells with 4% paraformaldehyde was chosen to screen hybridoma supernatants on cells treated under the same conditions as those used for immunization, but in this way also The plasma membrane of HMVEC-L cells could be stabilized and the cells for the different immunizations during the 65 weeks could be from the same co-culture. On the other hand, the glycolipids being weakly immunogenic, they can be assimilated to molecules such as haptens incapable of triggering an immune reaction by themselves. To make them more immunogenic, they can be coupled to a "carrier" protein molecule whose role will be to expose, more importantly, the hapten molecule. By fixing glycolipids with paraformaldehyde on endothelial cell membranes, can simulate the formation of "hapten-carrier" complexes in order to induce a greater immune response. In order to verify that fixation with paraformaldehyde did not modify antibody binding, we performed an ELISA on desiccated HMEC-1 cells, fixed or not fixed with 4% PFA, before or after desiccation (data not shown). Analyzes of about fifty hybridomas from the 139 hybridomas selected in the first screening strategy, show that there was no change in the attachment of antibodies on the dried cells whether fixed or not. The somatic hybridization made it possible, from the 1086 hybridomas obtained, to preselect the first 139 hybridomas by ELISA on dried cells, which were then screened according to two strategies. The first, which was to screen hybridomas on living cells by flow cytometry, allowed 13 of them to be selected and by HPTLC immunolabeling of glycolipid extracts from cells, we determined that 6 of them recognized the glycolipid Gb3. A second screening strategy, which consisted of screening by immunostaining the 139 hybridomas obtained by ELISA, made it possible to select many other antibodies, mostly directed against Gb3 but also against glycolipids being characterized. After purification, it was found that the antibodies generated by the second screen had a lower affinity. Also, we chose to retain three monoclonal antibodies (IgM, κ) directed against Gb3: monoclonal antibodies 25C10 and 16G8 which, by immunostaining on HPTLC, instead recognize the lower band of the doublet Gb3 _; which is shown in a minority manner in HMEC-1 cells, and the monoclonal antibody 3E2 which has the best immunostaining binding profile on HPTLC and in flow cytometry.

Gb3 is a globotriaosylceramide, also called Gb3 / CD77, or CTH (ceramide trihexoside). He first identified as an antigen of a rare blood group, the Pk group on the surface of erythrocytes and is also known as a marker of Burkitt lymphoma (BLA, Burkitt Lymphoma Antigen). On the surface of endothelial cells, its expression is regulated by pro-inflammatory cytokines such as T F-α involved in tumorigenesis processes. In addition, this glycolipid is known as a specific receptor for bacterial toxins and Preliminary studies have shown that verotoxin binding to Gb3 could inhibit in vitro angiogenesis (Heath-Engel and Lingwood, 2003). These results support us in the interest of developing therapeutic antibodies targeting Gb3, especially since the content of this glycolipid seems to be modulated when the endothelial cells are proliferating (see Example 2). In addition, the vast majority of anti-glycolipid antibodies generated as a result of somatic hybridization were directed against Gb3, suggesting that this glycolipid is quite immunogenic in Balb / c mice. Regarding the other seven antibodies generated during the first screening strategy and whose antigens are being characterized, it is possible that they are directed against other glycolipids of HMEC-1 cells that are not extracted according to our method of detection. purification by partition (as neutral and apolar glycolipids) or that it is antigens of glycoproteic nature, because it is conceivable that glycosylated units carried by glycoproteins are resistant to desiccation. EXAMPLE 4. Characterization of V murine monoclonal antibody 3E2 anti-Gb3

4.1. Results

Determination of the affinity constant.

 Analysis of the saturation curve obtained by the Scatchard technique allowed us to evaluate the affinity constant of the 3E2 antibody for Gb3 of Raji cells and HMEC-1 cells, as well as the number of sites to their surface. The results are listed in the following Table 9.

Table 9. Affinity constant of antibody 3E2 for Raji cells and

 HMEC-1.

The affinity of the antibody 3E2 was evaluated at 30 nM. The number of Gb3 sites is of the order of 2 × ⁶ sites for both cell types but there are slightly more sites on Raji cells than on HMEC-1 cells. Previously, we had shown by flow cytometry (FIG 9, Example 3) that the distribution of Gb3 was homogeneous for Raji cells for which 87.9% of the cells are positive, and heterogeneous for the HMEC-1 cells for which 68 2% of the cells are positive. Indeed, in the population of HMEC-1 cells, some cells weakly express Gb3 while others express it more strongly, unlike Raji cells.

Study of the specificity of the 3E2 antibody.

By an immunoenzymatic test (ELISA) on desiccated cells.

 Firstly, the specificity of mAb 3E2 was evaluated by an ELISA test on Raji, HMEC-1 desiccated cells and IMR32 cells of human neuroblastoma that do not express Gb3, using an anti-detection antibody. Mouse μ chain (Fig. 6).

 The 3E2 antibody does not bind to the IMR32 cells. It binds to Gb3-positive HMEC-1 and Raji cells but their ELISA binding profile is different (A). Indeed, the antibody binds more weakly on HMEC-1 cells. This difference could be explained by the average number of Gb3 sites which is slightly larger for Raji cells. Moreover, in the population of HMEC-1 cells, it is observed by flow cytometry that there is a heterogeneity of labeling with the 3E2 antibody and that some cells express Gb3 more weakly (Fig. 5, Example 3). It could also be envisaged that the Gb3 of HMEC-1 cells is less well recognized when it is present in a desiccated form, since it is recognized that the conformation of a glycolipid can influence its recognition by an antibody.

Despite their low immunogenicity, anti-glycolipids monoclonal antibodies have been obtained from other laboratories. We therefore compared the binding of the antibody 3E2 with that of the rat antibody 38.13. We did not include data for isotype control antibody and another mouse antibody monoclonal anti-Gb3 1 A4, as they bind non-specifically to the bottom of microtiter plate wells, even after the well saturation step (1 hour incubation at room temperature with 1% PBS-BSA) (data not shown). Antibody 38.13 recognizes Gb3 from HMEC-1 cells. The binding of antibodies 38, 13 and 3E2 is not entirely identical because these antibodies are revealed by two detection systems. Indeed, the binding of the antibody 3E2 is revealed by the successive incubation of a biotinylated anti-mouse secondary antibody followed by the incubation of a streptavidin-peroxidase complex, whereas that of the antibody 38.13 is revealed by the incubation of a secondary antibody anti-μ rat directly coupled to peroxidase.

Immunolabeling of glycolipids separated by HPTLC (iTLC).

 We compared the binding chromatographic profile of antibody 38.13 with that of antibody 3E2 (Fig. 7).

 Antibody 38.13 recognizes Gb3 of HMEC-1 cells as a doublet (lane 3), whereas the 3E2 antibody recognizes only the upper lane (lane 4). When screening the antibodies, that is to say before cloning the hybridomas, it is found that when immunolabeling with the supernatant of the antibody 3E2, it also recognized the Gb3 in the form a doublet (track 5). It is thus assumed that cloning has made it possible to select, in a mixture of several hybridomas of different specificities, a monoclonal antibody which binds only to the upper band of the Gb3 doublet.

 We then compared the two mouse anti-Gb3 monoclonal antibodies: the 3E2 antibody generated in the laboratory and the A4 antibody (Fig. 8).

The mouse monoclonal antibodies 3E2 and 1A4 have different chromatographic binding profiles: the 3E2 antibody recognizes the Gb3 upper band of the HMEC-1 cells whereas the 1A4 antibody recognizes the Gb3 in the form of a doublet (B) . In contrast, both the 3E2 and the A4 antibodies recognize the porcine source purified Gb3 (Matreya) as a doublet. Moreover, it is found that the binding of the antibody 3E2 is weaker than that of the antibody A4, probably because the antibody 1A4 has a better affinity for Gb3 than the antibody 3E2. Immunofluorescence cell staining study in flow cytometry.

In order to compare the binding of mouse anti-Gb3 antibodies, we performed flow cytometric labeling of HMEC-1, Raji and NXS2 cells using monoclonal antibodies 3E2, 1A4 and isotype control (IgM, κ). and revealed their binding by mouse anti-mouse detection antibodies (Fig 9 and Table 10).

Table 10. Mean fluorescence intensity after analysis in flow cytometry.

For HMEC-1 cells, the binding profile of the 3E2 antibody is different from that of the 1A4 antibody: 68.2% of the cells are positive with the 3E2 antibody (A) and 90.8% of the cells are positive with the antibody 1A4 (B). The average fluorescence intensity value is 568.3 for the 3E2 antibody and 984.2 for the 1A4 antibody (Table 10). There are therefore more positive cells when the cells are labeled with the 1A4 antibody, but for both antibodies, the labeling of HMEC-1 cells is heterogeneous. In fact, there are two cell populations: for the 3E2 antibody, the presence of a predominantly less marked population and for the 1A4 antibody, a predominantly strong population. The labeling of Raji cells is homogeneous and the percentage of positive cells is comparable between the two antibodies (87.9% of positive cells and 423.8 of average fluorescence intensity for antibody 3E2, and 93.6% positive cells and 498, 1 mean fluorescence intensity value for the antibody 1A4). Thus Raji cells strongly express Gb3 homogeneously within the cell population and both antibodies seem to recognize the glycolipid in comparable affinities. We then tried to compare the Gb3 content of HMEC-1, Raji and NXS2 cells by HPTLC analysis (Fig. 10).

 In HMEC-1 cells, the upper band of Gb3 is major. The antibody 3E2 preferentially recognizes this form with respect to the antibodies 38, 13 and 1 A4 which recognize Gb3 as a doublet (Fig. 10). In flow cytometry, after labeling the HMEC-1 cells with the antibody 3E2 (Fig. 10, A), the presence of a predominantly less marked population was observed. We can then suggest that in HMEC-1 cells, the molecular species corresponding to the upper band of Gb3, is found on a high proportion of cells that express it more weakly than a more marked minority population. In flow cytometry, after labeling HMEC-1 cells with the 1A4 antibody (Fig. 10, B), the presence of a predominantly more marked population was observed. There would then be a small proportion of HMEC-1 cells that strongly express the lower form of Gb3 or both forms undifferentiated.

 When the glycolipid expression profiles of HMEC-1 cells with Raji cells are compared by HPTLC (Fig. 10, lanes 4 and 5), it is found that Raji cells do not contain many charged glycolipids such as gangliosides. they are mostly neutral glycolipids (lane 5). The upper band of Gb3 is very much in the majority. Because of the low content of the species corresponding to the lower band, the binding of the antibody A4 is limited to the higher form, which explains why the antibodies 3E2 and 1A4 recognize the Gb3 of Raji cells in an equivalent manner. flow cytometry.

NXS2 cells of murine neuroblastoma, and T84 cells derived from human colonic adenocarcinoma, used in the co-culture system with HMVEC-L primary cells, do not express Gb3. This result is important because it has already been suggested that micronvironmental effects could be incorporated into the endothelial cells as a result of spontaneous release of glycolipids. thus making them more susceptible to VEGF. Since the T84 cells do not express Gb3, the antibodies generated by the Balb / c mice after immunization recognized an antigen expressed on the surface of the HMVEC-L cells and not a glycolipid antigen originating from the T84 cells and incorporated into the HMVEC cells. L. We thus analyzed the Gb3 content of HMVEC-L cells by comparing them by flow cytometry (Fig. 1 1) and by HPTLC (Fig. 12) to the expression profiles of FDVIEC-1 cells and human HUVEC macrovascular cells. for which the presence of Gb3 has already been demonstrated in the literature (Obrig et al 1993, Muthing et al 1999, Kanda et al 2004).

Table 11. Mean fluorescence intensity after analysis in flow cytometry.

Whether by flow cytometry (Fig. 11), or by HPTLC analysis (Fig. 12), it is found that the primary endothelial cells HMVEC-L and HUVEC express less Gb3 than the transformed endothelial cells HMEC-1.

 In flow cytometry, the 3E2 antibody does not recognize non-co-cultured HMVEC-L cells (3.4% positive cells). This lack of fixation is due to an overall decrease in the Gb3 doublet observed by HPTLC, because even if the overall Gb3 content is low, the upper band remains predominant in HMVEC-L cells. Since the 1A4 antibody binds to the Gb3 of HMVEC-L cells (26.1% positive cells) and since the upper band is predominant, the difference in binding is due to a difference in affinity between the two antibodies. The affinity of the 1A4 antibody for Gb3 seems superior to that of the 3E2 antibody, since the low Gb3 content that can be visualized in HPTLC can be detected by flow cytometry only with the 1A4 antibody. . In HUVEC cells, almost no Gb3 is detected by HPTLC. In flow cytometry, only 7.5% of the cells are positive with the A4 antibody instead of 2.5% positive cells with the 3E2 antibody.

HPTLC analysis shows that the overall content of neutral glycolipids is limited in HUVEC cells (lane 6). For the HMVEC-L cells, it is the proportion of Gb3 within the total neutral glycolipid fraction which is limited (lane 5) since these cells strongly express a glycolipid present in the form of a doublet probably corresponding to glycolipid Gb4. . For cells HMVEC-L and HUVEC, it is the upper band of the globoside Gb4 which is the majority whereas for the cells HMEC-1, it is the lower band of Gb4 which is maj oritaire. Primary cells have higher levels of complex gangliosides migrating below GDi _a . Moreover, for the three cell types, ganglioside migrating above GM1 would be the most abundant of the total ganglioside fraction. Since the preponderance of GM ₃ in the total ganglioside fraction has already been demonstrated in human endothelial cells (Obrig et al., 1993, Muthing et al., 1999, Kanda et al., 2004), it is likely that this predominant ganglioside HMEC-1, HMVEC-L and HUVEC cells are GM ₃ .

Immunofluorescence on cells in culture.

 We analyzed by immunofluorescence the cellular localization of Gb3. The results show that the 3E2 antibody is suitable for the study of immunofluorescent cells. The advantage of this technique is that one can observe the antigenic distribution of living cells directly on their culture medium, without trypsinization before labeling as is the case in flow cytometry. According to the results obtained with isotype control, the signal obtained does not come from non-specific interactions. It is observed that the distribution of Gb3 is localized to the cell membrane and that its membrane distribution is rather homogeneous within the cell (data not shown).

Immunohistochemistry on sections of frozen tumors.

 We analyzed immunohistochemically the distribution of Gb3 on frozen sections of Raji tumors expressing Gb3, and IMR32 tumors not expressing Gb3, which we had previously verified by ELISA with the antibody 3E2 and by flow cytometry . These tumors were implanted with Matrigel and removed after 3 to 4 weeks.

The results (not shown) indicate that the 3E2 antibody is an immunohistochemical tool that seems to meet expectations. It recognizes the presence of Gb3 which is localized at the membrane level of Gb3-positive tumors (Raji) and shows no non-specific binding on non-expressing tumors (IMR32). The antiparasitic antibody is also suitable for immunohistochemical studies because it does not exhibit specificity.

Nucleotide sequences of the VH and VL variable regions of the 3E2 antibody.

 In order to finalize the characterization of the 3E2 antibody, we analyzed after

RT-PCR, the nucleotide sequences of the variable regions VH and VL in order to align them with the data of the IMGT (International Immunogenetics Information System®) bank. The primers used for heavy and light chain sequencing are listed in Table 12. The nucleotide sequences of the heavy and light chains are shown in FIG. 13 and the genes used for these variable regions are listed in Table 13.

Table 12. Back primers used for amplification of the variable regions of the VH and VL gene segments of the 3E2 antibody.

Table 13. Genes used for variable regions of light chains.

Monoclonal antibody 3E2 has more than 99% homology with the germ-line IGHV4 gene (genomic DNA). There are two nucleotides that are mutated in the FR1 region. The antibody has 65.96% homology with the IGHJ2 gene and 78.57% with the IGHD1 gene. The VK of the antibody has a homology of 90.91% on 220 nucleotides with the IGKV14 gene and has 100% homology with the IGKJ5 gene. Due to the differences observed with the germinal configuration genes, it is thus demonstrated that the antibody has been subjected to a maturation process revealed by several somatic mutations. 4.2. Conclusion

Several hybridoma lines were obtained by somatic hybridization, including the antibody 3E2 (IgM, κ) directed against the neutral glycolipid Gb3. We continued its characterization using Raji human lymphoma cells for which Gb3 is a marker, IMR32 human and murine NXS2 neuroblastoma cells that do not express Gb3, and primary endothelial cells, HMVEC- microvascular lung cells. L, and macrovascular umbilical cord cells HUVEC.

We have determined from the Scatchard representation that the antibody has a good affinity of the order of 30 nM. On HMEC-1 cells, the membrane localization of Gb3 was demonstrated by immunocytochemistry, followed by immunohistochemistry, its ability to specifically bind to Gb3 on sections of frozen Raji tumors. We analyzed its specificity by ELISA on desiccated cells, by flow cytometry and immunostaining on glycolipids separated by HPTLC. The 3E2 antibody thus constitutes a new tool for analyzing Gb3, in addition to the antibodies that have already been generated previously, such as the rat IgM antibody 38.13 and the mouse IgM antibody 1A4 that were obtained in the team. Pr. S. Hakomori. These antibodies were used to establish Gb3 as a marker of Burkitt's lymphoma, to show that Gb3 was involved in the apopototic phenomena of Gb3-positive lymphoma B cells and that Gb3 targeting by Gb3 antibodies (Taga et al 1997, Tetaud et al., 2003) or recombinant verotoxin B subunit (Mangeney et al., 1993) could also induce an apoptotic response. A third antibody, BGR-23 obtained more recently (Kotani et al., 1994) was used to study the tissue distribution of Gb3 in Fabry disease (Askari et al., 2007). These antibodies are specific for the Gakx1 → 4Gai terminal motif and can not recognize (1) the Gakx1 → 3Gai motif present at the terminal end of the oligosaccharide chain of isoglobotriosylceramide (iGb3) and (2) the Galal → 4Gai motif present at the of the oligosaccharide chain of the Gb ₄ globoside (GalNac 1 → 3Gakx1 → 4Gal 1 → 4Glc 1 → Cer).

In the analysis of the speci fi city, we have shown the following characteristics. The 3E2 antibody recognized on HPTLC the band of the Gb3 doublet of HMEC-1 cells and the porcine Gb3 as a doublet, unlike the anti-Gb3 antibodies of rat 38. 13 and mouse A4, which recognized the two molecular species of Gb3 cells HMEC-1 and porcine Gb3. The molecular species corresponding to this upper band is expressed heterogeneously on HMEC-1 cells. There are two cell populations: a predominant population that expresses more weakly this Gb3 and a minority population that strongly expresses this Gb3. Antibodies 3E2 and 1A4 equally recognized Gb3 of Raji cells, which express on their surface the superior form in a largely major and homogeneous manner because all the cells are marked extensively. Moreover, in an in vitro model, it has been very clearly shown that incorporation of GDi _a into HUVEC cells makes these cells more sensitive to low levels of VEGF. They thus propose that, in the tumor microenvironment, glycolipids of tumor cells released and incorporated into the endothelial cells by a shedding phenomenon, would promote tumor progression. Since T84 cells do not express Gb3, the antibodies generated as a result of the immunization were directed against Gb3 expressed by HMVEC-L endothelial cells derived from co-culture.

Concerning the primary endothelial cells, it is noted that the presence of Gb3 is globally diminished, regardless of the shape of the ceramide. This difference in expression with HMEC-1 cells may be due to the tissue origin of the cells. Human microvascular endothelial cells of the kidney (HRMEC) express 48 times more than Gb3 macrovascular endothelial cells of umbilical cord (HUVEC) (3.2 nmol / 10 ⁶ cells HRMEC and 0.067 nmol / 10 ⁶ HUVEC) (Obrig et al 1993). Human microvascular brain endothelial cells (HBMECs) express 2 times more Gb3 than umbilical cord endothelial cells (HUVEC) (612 ± 185 and 269 ± 62 ng / mg protein) (Kanda et al., 2004).

Tumor transformation may also influence glycolipid content, as evidenced by the abnormal overexpression of GD ₂ disialoganglioside acid in neuroblastomas, most melanomas and some other tumors, or the overexpression of Gb3 on several tumor cell lines such as lymphoma. Burkitt (Wiels et al., 1981), breast cancer cells, ovarian cancer cells astrocytic tumors (Gariepy 2001, LaCasse et al., 1999), epithelial carcinoma cells of the upper digestive tract (Marques Filho et al., 2006) or in human tumor tissues such as breast cancer (Johansson et al., 2009). ), colorectal tumors or metastases (F aiguières et al., 2008). The fact that primary endothelial cells express less Gb3 than transformed cells is an interesting result because we want to target pro-angiogenic endothelial cells while sparing endothelial cells from healthy tissues. In addition, we analyzed the glycolipid content of HMVEC-L cells grown alone in the absence of co-culture with other cell types, and the presence of tumor cells in co-culture was shown to induce a phenotypic change. and genotypic primary endothelial cells (Khodarev et al., 2003).

 The biochemical changes observed during the malignant transformation of cells also affect the biosynthesis of ceramide. These modifications are manifested as well by variations in the length of the aliphatic chain, the number of substitutions or the degree of unsaturation. If in healthy tissues, the chains of fatty acids are shorter (C14: 0 to C18: 0), gangliosides of tumor cells have longer chains of fatty acids (C22: 0, C22: 1 and C24: 1 ) (Hakomori and Kannagi, 1983). Aberrant α-hydroxylation of the fatty acids of tumor gangliosides can also be observed.

It is usual to find glycosphingolipids present in the form of a doublet on chromatograms stained with orcinol. These doublets reflect the presence of various substitutions, the length of the fatty acid chain, the differences in hydroxylation and saturation, in the glycolipid ceramide which can thus migrate differentially by HPTLC. The porcine Gb3 obtained from Matreya consists of a mixture of two types of ceramide chains (C: 16 and C: 20) which are composed of 70% saturated fatty acids and 30% unsaturated fatty acids. . These two forms are recognized by the antibody 3E2 and it is very likely that the Gb3 doublet found in HMEC-1 cells is itself composed of two types of fatty acid chains and that the lower form which is not recognized by the antibody has a yet different fatty acid chain. It has already been demonstrated by HPTLC HUVEC cells that Gb3 was present in the form of a doublet that corresponded to two types of ceramide (C24: 0) and (Cl6: 0) chains determined by mass spectrometry (Miithing et al., 1998). In addition to the tissue origin of vascular cells, it appears that the glycolipid profile varies from one species to another. Only a few traces of Gb3 could be detected in BAEC primary bovine aortic endothelial cells, in which different types of fatty acid chains ranging from C24: 0 or C24: 1 to C16 could be detected.

The presence of these hydroxyl groups as well as variations in the length of the ceramide modify the conformation of the antigen and thus its accessibility to the antibody. The affinity of the anti-glycolipid antibodies would increase as a function of the length of the ceramide fatty acid chain, which could be the case for the higher form of Gb3 recognized by the antibody 3E2 in the HMEC-1 cells, then that the antibody has no affinity for the lower form. An anti- GD ₃ (mAb 4.2) obtained after immunization of mice with human melanoma cells containing a high content of GD ₃ C: 24 binds only slightly to cells having low levels of GD ₃ C: 24. In addition, it has been shown that monoclonal antibodies against glycolipids are able to recognize high density antigens and that these same antibodies would not be able to react with the same low density antigens.

The data concerning antibody binding differences are to be correlated with studies that show that density, and not just the presence of Gb3, is required for binding Shiga-like toxin. Other studies show that the Shiga-toxin (BxB) subunit B, induced by binding a "clustering" of Gb3 to the membrane that is responsible for the membrane invagination and that allows the internalization of the subunit and its routing into the intracellular compartment. This is made possible because StxB is homopentameric, which is also the case of verotoxin. The vast majority of the antibodies we selected belonged to the IgM iotype. Because of their pentavalence, M immunoglobulins would mimic the pentameric structure of these toxins and collect naturally occurring glycolipids in the form of rafts lipid levels, based on synthetic plasma membrane models. These data notably explain why an IgG1 isotype antibody such as the 12E10.1c antibody selected by the second screening strategy had a bad affinity for Gb3, as shown by the weak binding of this antibody to Gb3 of the HMEC-cells. 1.

 It has already been demonstrated that the molecular structure of Gb3, ie its fatty acid composition, determines lipid localization in lipid bilayers, and contributes significantly to the formation of Gb3 clusters. In the case of synthetic bilayers composed of phosphatidylcholine and Gb3, this finding is corroborated by the fact that the STxB affinity would increase if the length of the fatty acid chain of phosphatidylcholines decreases, resulting in higher Gb3 exposure. on the surface of this bilayer. Thus, in the same way as antibodies, molecular modeling studies have shown that the fatty acid composition influences the Gb3 conformation at the membrane interface and that this has a direct impact on the affinity of STxB. Verotoxin binding also depends on the length of the fatty acid chain of Gb3 ceramide. Glycolipids with medium or long fatty acid chains (Cl 6 and C24) are preferentially recognized, while short chains (Cl 2 and Cl 4) have minimal binding. C20: 0 and C22: 1 fatty acid chains have greater binding capacity and the presence of unsaturated fatty acids significantly increases verotoxin binding. Hydroxylation of fatty acids also increases binding. The binding of the verotoxin to the terminal Gal (al-4) Gal motif of glycolipids thus depends on the entire structure of the molecule and its molecular environment at the level of the plasma membrane of the target cells. These results highlight that glycolipids are involved in a cellular tropism closely related to their structure.

The antibody 3E2 thus targets an original glycolipid present on the endothelial cells. It would be important to determine the structure of the two molecular species of Gb3 present in the HMEC-1 and HMVEC-L cells, in order to highlight the structural variations that are responsible for the differential specificity of the 3E2 and 1 A4 antibodies. We previously showed that Gb3 expression was heterogeneous in HMEC-1 cells and appeared to be modulated according to the state of proliferation of these cells. In particular, it should be investigated whether these results can be corroborated with biological activity, in order to determine the therapeutic potentialities of the antibody.

EXAMPLE 5 In Vitro Biological Properties of VAcM 3E2 Anti-Gb3 We also wanted to determine whether this antibody may have biological properties and whether it may constitute a new therapeutic tool.

5.1. Results

Study of the expression polymorphism of Gb3 by flow cytometry with the antibody 3E2.

 We had previously demonstrated by HPTLC (see Example 2) that the glycolipid Gb3 content could be modulated when the cells were proliferating, either incubated in complete medium, or incubated in depleted media supplemented with a pro-proliferative factor, sphingosine -1-phosphate (S1P). Since the 3E2 antibody is a new analytical tool, we have again analyzed the expression of Gb3 on HMEC-1 cells by two different methods that differ from the HPTLC analysis allowing to analyze the total glycolipids of the cells independently of their location in the cell. By flow cytometry, we can analyze the homogeneity of Gb3 expression on the cell surface and by the Scatchard technique, we can determine the number of Gb3 sites on the cell surface.

 We first analyzed the expression of Gb3 after 24 h of incubation in depleted or complete culture medium (Fig. 14).

We confirm by cytometry that the Gb3 content decreases when the HMEC-1 cells are incubated for 24 h in depleted media. Indeed, in complete medium, there is the presence of two cell populations after labeling with the antibody 3E2. When the cells are incubated in a depleted medium, there are fewer labeled cells (66.4 ± 5.5% in complete medium and 42.0 ± 2.0% in depleted medium (n = 3)) until that there is almost only one cell population that expresses weakly Gb3. The average fluorescence intensity is decreased by more than 50% (Table 14). Table 14. Percentage of positive cells and average fluorescence intensity of the cells after flow cytometry analysis.

The decrease in Gb3 content is less important when the cells are labeled with the A4 antibody. Although this result is preliminary, it could notably suggest that the upper band of the doublet Gb3 is particularly overexpressed in complete medium, that is to say when the endothelial cells proliferate. This would show that the 3E2 antibody, which specifically recognizes this upper band, and any other antibody with this same specificity, would primarily recognize proliferating endothelial cells, unlike antibodies recognizing the two bands of the doublet.

We also analyzed by immunocytochemistry with the antibody 3E2, 24 hours after their seeding, the distribution of the membrane Gb3 of two interacting HMEC-1 cells grown in their complete medium. It is observed that the distribution is not homogeneous as can be observed when the cell is isolated (data not shown). There is a membrane polarization of Gb3 at the level of the contact zone between the cells and also the presence of more isolated cells more weakly marked (indicated by the white arrows). Thus, in addition to the heterogeneity of the labeling of HMEC-1 cells within the cell population, there is heterogeneity in the distribution of Gb3 at the level of the membrane of interacting cells. The presence of isolated cells indicated by the white arrows, which are more weakly labeled by the 3E2 antibody, confirm the presence of cell populations observed in flow cytometry, which express Gb3 differently on their surface. We then analyzed by flow cytometry with the 3E2 antibody, the expression of Gb3 of HMEC-1 cells incubated for 24 h and 72 h in depleted medium supplemented with sphingosine-1-phosphate or its diluent, PET (FIG. 15).

 Table 15. Percentage of positive cells and average fluorescence intensity of the cells after flow cytometry analysis.

When cells are incubated in the presence of S 1P, the Gb3 content at the surface of HMEC-1 cells increases slightly. In flow cytometry, this content is detectable only after 72 hours whereas this increase was visualized as early as 24 hours in HPTLC (FIG. The cells are positive at 42.0 + 0.3% after incubation with S1P, whereas only 24.4 + 1.6% are positive in the presence of depleted medium supplemented with PET diluent. When the cells are incubated in complete medium, the increase in Gb3 content is greater (Fig. 14). The fetal calf serum present in the complete medium comprises a mixture of growth factors and this increase may be due to a synergistic action of the growth factors.

We then evaluated by the Scatchard technique, the number of Gb3 sites on the surface of HMEC-1 cells incubated for 24 h in depleted media or in complete medium. Analysis of the saturation curve according to the Scatchard equation allows us to precisely determine the number of Gb3 sites of HMEC-1 cells, recognized by the antibody 3E2 (Table 16). Table 16. Number of Gb3 sites evaluated by antibody 3E2 (Kd = 30 nM) for HMEC-1 cells after incubation in depleted culture medium (MA) and in complete medium (MC).

There are 8.5 times more Gb3 sites when the HMEC-1 cells are incubated for 24 h with complete culture medium than when the cells are incubated with depleted culture medium.

 It is clear that the Gb3 content of proliferating cells is modulated: the average fluorescence intensity of the cells is increased by more than 50% (FIG 14), a polarization of Gb 3 is observed when the HMEC-1 cells in culture are in contact while isolated HMEC-1 cells are more weakly labeled (data not shown) and there are 8.5 times more Gb3 sites on the surface of proliferating HMEC-1 cells (Table 15). These results confirm once again the interest of generating anti-Gb3 antibodies, it remains to determine the biological properties of the antibody.

In vitro biological properties of murine monoclonal antibody 3E2.

Cytostatic properties of the 3E2 antibody.

Cell viability study (MTT) of cells treated with the 3E2 antibody.

We measured on the one hand the inhibition of the cell viability of HMEC-1 and Raji cells after 24 hours of incubation with the antibodies 3E2 and 1 A4 at different concentrations (Fig. 16) and on the other hand the kinetics of inhibition of cell viability of HMEC-1 cells after 24 h, 48 h and 72 h incubation with 20 μg / ml of 3E2 antibody (Fig. 17).

At 24 h (Fig. 16, A), the inhibition of cell viability induced by the antibodies 3E2 and 1A4 is comparable and leads to a saturation plateau of 20 to 40 μg / ml of antibody, for which values equal to 14.6 ± 1.6% and 16.3 ± 3, 1% inhibition. No inhibition of cell viability is observed with the isotypic control antibody on these HMEC-1 cells. This inhibition is more important for Raji cells (B) which have more Gb3 sites (2.0 10 ⁶ sites against 1.7 10 ⁶ sites for HMEC-1 cells) and which express Gb3 homogeneously. Within the HMEC-1 cell population, some cells express Gb3 more weakly, which may explain why the inhibition of viability is lower. No inhibition is observed for NXS2 cells that do not express Gb3. Inhibition of cell viability is therefore dependent on Gb3.

 Inhibition of cell viability was observed for 72 hours at a single dose of 20 μg / ml of antibody 3E2 (Fig. 17). The maximum value of inhibition of cell viability is almost reached as early as 24 hours of incubation. At 72 h, the inhibition of cell viability is observed to be slightly higher (22.2 ± 9.1%). For the 1A4 antibody, the kinetics of the inhibition is comparable to that of the 3E2 antibody (data not shown). No inhibition of cell viability was observed with the 3E2 antibody on NXS2 cells that did not express Gb3 at 48 h and 72 h (data not shown).

Video-kinetic study of HMEC-1 cells treated with the 3E2 antibody.

 We sought to determine what could be the nature of the inhibition of cell viability observed and for this we carried out a video-kinetic accelerated study of HMEC-1 cells for 24 h, in the presence of 20 μg / ml of 3E2 antibody or 20 μg / ml isotype control antibody. For each condition, we compared the number of HMEC-1 cells per time interval (Fig. 18), the rate of dividing cells per time interval and the cell division time (Fig. 19).

Fig. 18 represents the number of HMEC-1 cells as a function of the incubation time with 20 μg / ml of 3E2 antibody or isotype control antibody. It is observed that there are fewer HMEC-1 cells in the presence of the 3E2 antibody at 24 h of incubation. This decrease is significant from 6 hours of incubation (A). After 24 hours, there is a decrease in the number of cells that can be evaluated at about 18%, which is close to the 14.6% inhibition of cell viability that was observed at 24 hours in the morning. MTT (Fig. 16). Figure 19 shows the proportion of cells that enter cell division per time interval (A) and their division time during these 24 hours of cumulative incubation (B). In fact, the dividing cells can be detected and counted: the cells lose their adhesion, become detached from the support, become more refractive and adopt a circular morphology whereas they exhibit cytoplasmic expansions when they are adherent. It is thus possible to identify dividing cells and also their dividing time until both daughter cells adhere back to the support.

 It is observed that the number of dividing HMEC-1 cells is smaller when the cells are incubated in the presence of the antibody 3E2 (A). This decrease is significant from 12 o'clock. When we analyze their cell division time over all 24 hours of incubation, we see that this time is globally increased (B). In the presence of the antibody 3E2, there are fewer cells that take between 0 and 40 minutes or between 41 and 80 minutes to divide, in favor of cells that take more than 80 minutes to divide. There is also the presence of cells that take more than 120 minutes to divide ("sup 120", B), cells that fail to complete their division before the end of 24 hours of incubation and cells that fail. not to enter division ("recolle", B). We classified these cells as mitotic aberrations (Fig. 19, B).

 The analysis of the recorded images corresponding to the division times of cells incubated in the presence of the isotype control and the 3E2 antibody shows that the cell indicated by the arrow, takes about 40 minutes to divide in the presence of the isotype control and another cell. incubated with antibody 3E2 puts 130 min (data not shown).

 The analysis of the recorded images of the cells classified as mitotic aberrations shows the presence of cells that fail to complete their division within 24 h of incubation, and the fate of such a cell observed at 27 h and which finally burst in 120 minutes.

Among the mitotic aberrations we observed, there is also a significantly greater presence of cells that attempt to enter cell division without being able to do so. These cells detach, become refractive and then adhere again, successively. Anti-angiogenic properties of the 3E2 antibody.

 We evaluated the anti-angiogenic properties of the 3E2 antibody by the aortic ring test which is the ex vivo reference test for measuring the pro-angiogenic or anti-angiogenic activities of molecules (Fig. 20).

 After five days of incubation rings in complete medium without antibodies (A, photo 2), the formation of a large number of vascular buds characterized by their length and density is observed. When the rings are incubated with the antibody 3E2 (Fig. 20, B), inhibition of bud formation is observed as early as 20 μg / ml, which is significant at 40 μg / ml. This inhibition is Gb3-dependent, the A4 antibody has anti-angiogenic activity comparable to the 3E2 antibody, while the isotypic control antibody does not induce a significant decrease in bud formation. Cytotoxic properties of the 3E2 antibody in the presence of human complement

(CDC).

 We have assessed above that the 3E2 antibody exhibits cytostatic activity: it induces an inhibition of cell viability at 24 h, a decrease in the number of cell divisions with an extension of the dividing time. It also induces inhibition of vascular bud formation on explants of aortic cross sections in culture. These observations were obtained in the absence of complement. We sought to measure the CDC (complement dependent cytotoxicity) activity of the antibody by flow cytometry according to a method developed in the laboratory using the ability of propidium iodide to bind to cell DNA. If cell membranes have been damaged by CDC activity, their DNA will be accessible and propidium iodide can be visualized in flow cytometry.

Fig. 21 presents the CDC activity of the antibodies 3E2, 1A4 and the isotype control (10 μg / ml of antibody) on the FDVIEC-1, Raji and NXS2 cells, in the presence of decomplemented human serum and not decomplemented. Table 17 lists the percentage of cells lysed for antibody concentrations ranging from 0.1 to 10 μg / ml (average of 3 independent experiments). Table 17. Percentage of lysed cells (n = 3) after antibody fixation

 3E2 and 1 A4 on HMEC-1, Raji and NXS2 cells.

 ND: not determined. ISO: Isotype control.

In the absence of non-decomplemented serum, the cells are incubated with decomplemented serum. At 10 μ / ηι1, the 3Ε2 antibody exhibits CDC activity on HMEC-1 cells (Table 17, 51.3 ± 9.2% lysed cells) and Raji cells (71.8 ± 5.5%). lysed cells). The activity can be detected at low concentrations of the order of 0.5 μ ηύ of antibodies (20.6 ± 3.2 of lysed HMEC-1 cells and 36.4 ± 6.8% of lysed Raji cells). ). The 1A4 antibody also has a significant CDC activity at 10 μg / ml (65.2 ± 13.2% of HMEC-1 cells positive and 82.9 ± 7.7% of Raji positive cells). The CDC activity can be detected as soon as 0.1 μg / ml of antibody (37.2 ± 12.6% of HMEC-1 positive cells and 25.8 ± 3.9% of Raji positive cells).

 NXS2 cells, which do not express Gb3, are not lysed in the presence of the 3E2 and 1 A4 antibodies and the HMEC-1 and Raji cells, which express Gb3, are not lysed in the presence of isotype control: thus the observed cytotoxic activity is clearly dependent on Gb3. In the absence of antibodies, the NXS2, HMEC-1 and Raji cells are not lysed in the presence of non-decomplemented serum, so there is no activation of the alternative complement pathway which plays a role. important in innate nonspecific immune defenses. The HMEC-1 and Raji cells are not lysed in the presence of the antibodies A4 or 3E2 when they are incubated in the presence of decomplemented serum or in the absence of human serum: they do not observe apoptosis-type cell death or necrosis. The cytolytic activity observed is therefore very dependent on complement activation, via the binding of antibodies to Gb3.

In parallel with the propidium iodide markings, which reflect the CDC activity of the antibodies, we measured the expression of Gb3 by flow cytometry, in order to evaluate the ratios "CDC activity / cell labeling" for 10 μg / ml of antibodies, which are listed in Table 18.

Table 18. Ratios "CDC activity / cell labeling".

 Marking the

 Iodide marking

 Gb3 Propidium Ratios (CDC)

 membrane

 3E2 cells 51.3 + 9.2% 75.0 + 1 1.3% 68.2 + 6.2% HMEC-1 1A4 65.2 ± 13.2% 92.7 + 5.9% 69.9 + 10.2%

 3E2 71.8 ± 5.5% 91.6 + 5.3% 78.7 + 8.2%

Raji cells

1A4 82.9 ± 7.7% 99.3 + 0.2%, 83.4 + 7.8% For HMEC-1 cells, the ratio is evaluated at 68.2 ± 6.2% for the 3E2 antibody and 69.9 ± 10.2% for the 1A4 antibody. For Raji cells, the ratio is evaluated at 78.7 ± 8.2% for the 3E2 antibody and 83.4 ± 7.8% for the 1A4 antibody. Thus both antibodies have comparable CDC activities. The 1A4 antibody has higher CDC activity values than the 3E2 antibody because its affinity for Gb3 is greater.

5.2. Conclusion

Antibodies act primarily in three ways that can be cumulative. By direct fixation, they can induce cytostatic activities (cell cycle blocking type) or cytotoxic activities (apoptosis type, necrosis). By CDC, they can induce the activation of complement proteins by binding of the Clq protein complex to the Fc region of antibody bound to their target, resulting in the formation of the membrane attack complex and cell lysis. By ADCC, the FcγR receptors bind to the Fc region of antibody bound to their target thereby leading to lysis or phagocytosis induced by an effector cell of the immune system.

The various biological tests used show that the antibody 3E2 initially has a complement-independent cytostatic activity. MTT is observed to inhibit the cell viability of FDVIEC-1 cells equal to 14.6 ± 1.6% for 20 μg / ml of antibody as early as 24 hours (Fig. 16). This result is confirmed by a video-kinetic study of HMEC-1 cells. When the cells are incubated with 20 μg / ml of antibody, there is a decrease in the overall number of cells close to the inhibition value obtained in MTT (FIG 18). There are fewer cells dividing and their dividing time is generally longer. At 72 h, the cell viability inhibition of HMEC-1 cells is 22.2 ± 9.1%. This value is close to the maximum value obtained at 24 h with 40 μg / ml of antibody 3E2 (16.3 ± 3, 1%, Fig. 16). This can be explained by the presence of HMEC-1 cells expressing Gb3 more weakly, which are then less subject to the inhibition of cell viability induced by the antibody 3E2 and by the very fast growth of these cells transformed by the SV40 virus. This is why we reduced our video-kinetic study over the first 24 hours because, beyond this, HMEC-1 cells invade the fields of analysis very quickly and it is no longer possible to count them.

In the presence of the antibody 3E2, there is also an increase in the frequency of mitotic aberrations associated with the presence of cells blocked in division for several hours and which do not break out, as well as the presence of cells that can not enter in division and which alternate repeated cycles of cellular attachment and detachment. Such mitotic aberrations can be attributed to death mechanisms involving adhesion, such as death by anoikis, as has already been demonstrated for an antibody targeting GD ₂ ganglioside in small cell lung cancers. Anoikis is a phenomenon of death that was first identified in 1994 and characterizes the loss of contact of cells with their extracellular matrix. It has been shown that an anti-GD ₂ antibody induces apoptosis via the ubiquitous cytoplasmic protein FAK (Focal Adhesion Kinase) found within the adhesion complex and which can be activated by integrins but also by different growth factors. cytokines or hormones. Regarding Gb3, it has already been shown for microvascular endothelial cells from patients with Fabry disease, that Gb3 present in large amounts was associated with an increased expression of molecules involved in cell adhesion such as ICAM proteins. -1, VCAM-1 and selectin E. Zemunic et al. (2004) compared the distribution of Gb3 with that of E-selectin in H FUVEC cells stimulated by T F-α. Activated cells were characterized by an increase in E-selectin expression associated with an increase in Gb3 expression. These results suggest that Gb3 may have a potential role in adhesion mechanisms within the endothelium (Zemunik et al., 2004).

We have also demonstrated by immunocytochemistry on proliferating cells, that there was a polarization of membrane Gb3 at the contact zone of two interacting HMEC-1 cells (data not shown). When the cell is isolated, the membrane distribution of Gb3 is rather homogeneous (data not shown). It has recently been described that the increase in glycosphingolipid content in tumor tissue, such as Gb3, which is 2 to 3 times higher in epithelial carcinomas, may result in morphological levels at membrane microdomains (Marques Filho et al., 2006). These modifications would be responsible for a better interaction between cells and cells with their extracellular matrix, which would increase their infiltration capacity and their metastatic power (Marques Filho et al., 2006). The migration of the cells requires the establishment of a finely regulated system of organization of the cytoskeleton and adhesion complexes. In endothelial cells, treatment with IFN-γ increases the relative proportion of intracellular Gb4 that is associated with the cytoskeleton of the cell. The specific modulation of glycosphingolipids by IFN-γ suggests that they may play a role in the adhesion mechanisms of activated endothelial cells. In particular, the most abundant glycolipids HUVEC, the Gb4 and GM ₃ are located at the surface and intracellularly where they are associated with the intermediate filament vimentin cytoskeleton could play a role in the transport of glycosphingolipids. It was then shown that after binding to Gb3, the verotoxin is internalized to the rough endoplasmic reticulum and the nuclear envelope, suggesting that Gb3 may play a role in cell growth since its expression varies with the cycle. cell and that the cells in phase G1 / S are more sensitive to the binding of the toxin whereas the cells in the stationary phase are refractory there. It would appear that variations in Gb3 expression are associated with cell cycle, cell migration and adhesion.

The strong membrane expression of an antigen, especially if it is not subject to internalization phenomena, can make it possible to obtain a high density of antigen-antibody complexes at the membrane, favoring the recruitment of the effectors of the antigen. immunity, and in particular complement. The classical pathway is activated by the antigen-antibody complex and among immunoglobulins, IgM are effective activators of complement-dependent cytotoxicity. The 3E2 antibody is capable of activating CDC as early as 0.5 μg / ml on HMEC-1 cells (20.6 +/- 3.2% of lysed cells) which have 1.7-10 ⁶ Gb3 sites and 0.1 g / ml on Raji cells (15.7 ± 5.2%) which have 2.0 June ¹⁰ sites. For the 1A4 antibody, the CDC activity is observed as soon as 0.1 g / l on the HMEC-1 (22.2 ± 6.9%) and Raji (25.8 ± 3.9%) cells. Both antibodies are thus able to activate the complement pathway for weak antibody concentrations. This complement-dependent activity allows us to consider future preclinical studies in mice.

EXAMPLE 6 Sequences of the anti-Gb3 antibodies according to the invention The seven selected monoclonal antibodies (3E2, 14C11, 15C11, 22F6,

25C10, 11E10, 16G8) were obtained from a single somatic hybridization and a single cloning. They are all IgM isotype and all have a κ light chain. To analyze the mechanisms of the Gb3 immune response and their structure-function relationship, we have sequenced the V _H and V _L gene segments of the antibodies directed specifically against glycolipid Gb3. The problem was to determine what this specificity could be due to: either the restricted expression of the repertoire of VDJ genes, or the presence of somatic mutations.

Genes used for the variable regions of the heavy chain of anti-Gb3 antibodies.

 The nucleotide sequences of the heavy chains of the seven anti-Gb3 antibodies are shown in FIG. 22.

The genes used for the heavy chain variable regions are listed in the following Table 19.

Table 19. Genes used for variable regions of heavy chains.

 Gene J Sequence Length in

Name Gene V Gene D

 and his sequence AA and his Homology Homology allele and CDR FR allele

 allele of the junction

IIGGHHVV11-- 99 66., 88 33 %% 1 0 0%

5C11 VH2A IGHJ3 * 01 IGHD3-3 * 01 8.8.10 [2.17.38.7] CA GD AW

 82 * 01 (214/221 nt) (35/35 nt) FAYW

IGHV1S22 9 8, 7 1% 9 3, 3 3% CTRKFLFDY2F6 VH2B IGHJ2 * 01 IGHD5-1 * 01 8.8.8 [6.17.38.10]

 * 01 (229/232 nt) (42/45 nt) W

IGHV1-

74 * 01, or 9 3, 0 9% 8 4, 7 8% CAGGDRYG5C10 VH2B IGHJ2 * 01 IGHD3-3 * 01 8.8.8 [6.17.38.10]

 IGHV1- (216/232 nt) (39/46 nt) YW

74 * 04

 IGHV2-6-

7 * 01, or 1 0 0% 81.65% CARNGNYL4C11 VH1A IGHJ2 * 01 IGHD2-1 * 01 8.7.11 [9.17.38.11]

 IGHV2-6- (238/238 nt) (39/48 nt) AFDYW

7 * 02

 IGHV4-9 9, 1 5% 6 5, 9 6% CARGDYYGE2 VH1A IGHJ2 * 01 IGHD1-1 * 01 8.8.12 [7.17.38.11]

 1 * 02 (233/235 nt) (31/47 nt) SRCDYW

IGHV5- 9 8, 7 3% 9 3, 7 5% CARPTYGYF6G8 VH3A IGHJ2 * 01 IGHD2-11 * 01 8.8.10 [8.17.38.11]

 12 * 02 (234/237 nt) (45/48 nt) DYW

IGHV9-3- 9 8, 7 3% 8 4, 0 9%

1E10 VH2C IGHJ2 * 01 IGHD2-3 * 01 8.8.5 [7.17.38.11] CASGVYW

 1 * 01 (233/236 nt) (37/44 nt)

From the above table, the seven Gp3 glycolipid-specific antibodies each use a different V gene with a germline homology degree of greater than 98% except for the 15C1 1 and 25C10 MABs which are respectively distinguishable. by 96.8% and 93, 1% homology. Note that for mAb 14C1 1 a homology of the V gene identical to the IGHV2-6-7 * 01 or 02 allele. Moreover except for the mAb 15C11 which has 100% homology with the IGHJ3 allele. * 01, the other 6 mAbs all use the same gene J (IGHJ2 * 01) which combines with a D gene, different products VH sequences, different for each hybridoma. The length of the hypervariable region CDR3H is between 5 and 12 amino acids.

Genes used for the variable regions of the light chain of anti-Gb3 antibodies.

The nucleotide sequences of the heavy chains of the seven anti-Gb35 antibodies are shown in FIG. 23. The genes used for the variable regions of the light chains are listed in the following table.

 Table 20. Genes used for variable regions of light chains.

Sequence length in

Name Gene V Gene J

 AA

 de and his and his CDR CDR CDR

 Homology FR homology of the sequence allele allele 1 2 3

 junction

11E10 VK IGKV1- 1 0 0% 1 0 0% CVQGTH ##

 IGKJ1 * 01 11 3 X [6.17.36.7]

At 133 * 01 (234/234 nt) (28/28 nt) TF

25C10_V IGKV3- 1 0 0% 1 0 0% CQQSKEVP

 IGKJ1 * 01 10 3 9 [6.17.36.7]

KA 2 * 01 (233/233 nt) (26/26 nt) RTF

16G8 VK IGKV4- 9 9, 5 3% 1 0 0% CQQWSSNP

 IGKJ5 * 01 5 3 9 [5.17.36.7]

F 72 * 01 (214/215 nt) (25/25 nt) PTF

15C11_V IGKV4- 1 0 0% 1 0 0% CQQGSSIP

 IGKJ2 * 01 7 3 9 [5.17.36.8]

KF 91 * 01 (221/221 nt) (23/23 nt) RTF

 14C11_V IGKV5- 1 0 0% 1 0 0% CQNGHSFP

 IGKJ5 * 01 6 3 9 [7.17.36.10]

KA 39 * 01 (222/222 nt) (38/38 nt) LTF

 IGKV5- 9 9, 5 5% 1 0 0% CQQSNSWP

22F6_VKF IGKJ5 * 01 6 3 9 [6.17.36.10]

 45 * 01 (220/221 nt) (35/35 nt) HTF

IGKV14- 9 0, 9 1% 1 0 0% CLQHGESP

3E2_VKF IGKJ5 * 01 6 3 9 [6.17.36.10]

111 * 01 (200/220 nt) (38/38 nt) LTF The alignment of the nucleotide sequences coding for the variable region VK of the seven monoclonal antibodies firstly shows a use of V genes that are all different but which, however, are almost identical. all in germinal configuration with the exception of the MAb 3E2 that we have selected as the most affine of all. Its specificity is most likely associated with the variability of the J segment of VH (65.9%) combined with that of the V gene segment of VK (90.9%). It is also interesting to note that the MAb 3E2 is encoded by the same J gene (IGKJ5 * 01) as the 14C 11 and 22F6 mAbs with the same degree of homology but with a different amino acid junction expression. one hybridoma to another. The seven anti-Gb3 antibodies were generated in a single somatic hybridization and the similarities observed in the heavy and light chains suggest that the antibodies are cloned. Our results indicate that Variety of V _H and V _L chains can encode for anti-Gb3 antibodies and some are subject to a maturation process revealed by several somatic mutations. The set of nucleotide data collected for the expression of the seven specific Gb3 mAbs confirms after cloning of the hybridomas the existence of distinct combinations of the VDJ and VJ genes for the expression of the mAbs without apparently any restriction.

 It could indeed be very interesting to follow at the same time the measurement of the affinity of the monoclonal antibody and the state of the respective nucleotide sequences to understand the nature of the interactions of the mAbs and glycolipid Gb3. By site-directed mutagenesis it may be easier to define the critical amino acids of the variable regions and more particularly the hypervariable regions which classically confer the specificity of the antibodies.

 We are currently evaluating the affinities of each of the seven antibodies by the Scatchard technique in order to more precisely highlight the relationship between the nucleotide structure of the antibody and its affinity.

Even so, the sequencing of the V _H and V _{L regions} of mAb is essential for designing and optimizing for human therapeutic purposes chimeric or even humanized antibodies failing to obtain human antibodies. This approach requires at least a three-dimensional modeling of the variable regions V _H and V _L of the antibody adapted to the specific recognition of Gb3 to substitute the hypervariable regions of a human antibody by those of the mAb 3E2 that we have defined as the leader of the seven antibodies that we have characterized with respect to glycolipid Gb3. EXAMPLE 7 In Vivo Confirmation of the Anti-Angiogenic Properties of the 3E2 Antibody in Two Tumor Models

The therapeutic potential of the 3E2 antibody in the treatment of solid tumors was confirmed in vivo in two tumor models based on the murine neuroblastoma NXS2 line, and compared to that of an anti-GD2 antibody.

These two models based on murine neuroblastoma NXS2 were chosen for the following reasons: - Since NXS2 cells do not express Gb3 (see Example 3 and results below), these models make it possible to accurately evaluate the in vivo effects of the 3E2 antibody on the tumor vessels (ie ie its anti-angiogenic properties), and not a combination of the effects on the tumor vessels and on the tumor cells themselves, as would be the case if the selected tumor cells also expressed Gb3,

 In this model, there exists a reference antibody treatment corresponding to the anti-GD2 antibodies, the GD2 being another glycolipid which is, unlike Gb3, expressed by the NXS2 tumor cells and not expressed by the cells of the blood vessels.

These two models therefore make it possible to evaluate the in vivo anti-angiogenic properties of the 3E2 antibody, and to compare its efficacy with that of an antibody considered as reference treatment in the chosen pathology.

7.1. Materials and methods

NSX2 tumors established subcutaneously

Female AJ mice were inoculated subcutaneously in the right flank with 10 ^{6 live} NSX2 cells (neuroblastoma line) (viability> 95%) resuspended in 150 μl. of PBS. Tumor progression was measured daily. The long (D) and short (d) diameters of each tumor were measured with calipers. The tumor volume V was calculated using the following formula: V = 0.5 x D xd ² . When the tumors reached a volume V of 100-150 mm ³ , the mice were treated with the antibody 3E2 directed against the upper band of the doublet Gb3 expressed by the HMEC-1 cells, the anti-GD2 14G2a antibody, IgM isotype control antibody or PBS alone. The treatment consisted of a single injection into 150 μl of PB S. The tumor progression was measured as above and expressed as a percentage increase with respect to the tumor volume at the time of injection. Metastatic NSX2 tumor model

 Mouse

 Male A / J mice (6-8 weeks old) were obtained from Harlan Sprague-Dawley (Sulzfeld, Germany). Animal studies were carried out in accordance with Directive 86/609 / EEC.

Tumor model

Experimental hepatic metastases were induced by injection into the tail vein at day 0 of 2.5 × 10 ⁵ NSX2 tumor cells (viability> 95%) in 200 PBS. The mice were treated with intravenous injections of monoclonal antibodies at day 1, 2, 7 and 8 for the antibody 3E2 (IgM, 200 μg in 150 μl of PBS), or at days 1 and 2 for the antibody 14G2a ( IgG, 100 μg in 100 μl) or IgG isotype control antibody (100 μg in 100 μl), respectively. The molar amounts of antibody of each injection are similar for all antibodies, the larger mass used for the antibody 3E2 (IgM) relative to the other antibodies (IgG) being due to the higher molecular weight of IgM compared to IgG. . Similarly, the new 3E2 antibody injections at days 7 and 8 are necessitated by the limited half-life of IgM in vivo (approximately 1 week), which is not the case for the other antibodies. isotype IgG, IgG having an in vivo half-life of about 3 weeks.

 Control mice received only PBS. The animals were sacrificed at day 28, their liver was weighed and the number of metastases evaluated.

Histology and immunohistochemical staining

Frozen tumors were cut and fixed with cold acetone for 10 minutes. The samples were labeled for 90 minutes with the following monoclonal antibodies: a rat antibody against CD31 Murin (1: 50, Millipore, Molsheim, France) and a biotinylated antibody 3E2, 14G2a (anti-GD2) or its isotype control (Beckmann Coulter, Fullerton, CA, USA) (40 μg / ml). Antibody 3E2 and isotype control were biotinylated with an EZ-Link Sulfo-HS-LC-Biotinylation kit (Thermo Scientific, Courtaboeuf, France). The labeling was revealed by 90 minutes of incubation with a goat anti-rat antibody conjugated with AF 488 (1: 400, Invitrogen) for CD31 and with streptavidin conjugated with AF 568 (1: 200, Invitrogen) for the 3E2 antibody or isotype control thereof. The samples were finally coated with a ProLong gold antifade reagent with DAPI. Healthy animal organs have also been labeled with the 3E2 antibody or its isotype control using the same protocol. The markings were observed under a confocal fluorescence microscope (Nikon, Champigny sur Marne, France). Muscle sections were used as a normal tissue control.

7.2. Results

NSX2 tumors established subcutaneously

Tumor volumes ranged from 120 to 130 mm ³ and did not differ significantly upon injection of treatment (3E2, 14G2a, IgM control, or PB S).

 The evolution of the tumor volume after treatment is shown in FIG. 24. At 24 hours after treatment, a single injection with the 3E2 antibody significantly reduces the tumor volume at 24H compared with treatment with PB S alone (tumor volume of 178.3 ± 13.19 mm 3 with PBS (n = 11) vs 137.9 ± 11.08 mm 3 for 3E2 antibody (n = 8)). This is not the case for other treatments (14G2a antibody or IgM control).

 At 48 and 72 hours, treatment with the 3E2 antibody leads to a lower tumor volume, although the difference is not statistically significant. This is probably due to the fact that too few animals were tested at 48 and 72 hours after treatment.

Metastatic NSX2 tumor model

 The antitumor efficacy of the 3E2 antibody was then determined in the model of experimental liver metastases of mouse neuroblastoma NSX2 developed by Lode et al.

It was first verified that the 3E2 antibody specifically targets the vessels within the tumor mass by histology and immunohistochemical staining. The results are shown in Table 21 below, and show that the antibody 3E2 generates an intense colocalized labeling with that of the anti-CD31 antibody (marker cells of the blood vessels), and no markings in the tumor mass. The muscle sections showed no Gb3 staining, only the endothelial cells within the tumor vessels being labeled with the 3E2 antibody. On the other hand, NXS2 tumor cells showed a high level of GD2.

cells

 Cells of

 Endothelial Antigen Muscle Neuroblastoma

 tumor

 GD2 +

 CD31 +

 Gb3 +

 Antibody

 control

 Table 21. Distribution of Gb3 within the tumor mass of NXS2 metastases. The distribution of GD2, Gb3 and CD31 distribution within the tumor mass of untreated mouse NXS2 metastases was detected by immunohistochemistry. -, no edge lines; +, positive marking. Muscle sections were used as a control.

To evaluate the therapeutic benefit of injections of the monoclonal antibody 3E2 and to compare it with immunotherapy with the anti-GD2 monoclonal antibody 14G2a, the number of liver metastases of NSX2 and the weight of the liver were evaluated after these different treatments ( Figure 25).

Treatment of mice with 3E2 and 14G2a antibodies was highly effective in reducing hepatic neuroblastoma metastases, as indicated by decreased liver weight from 39 ± 6.23 g (PBS-treated mice) to 17 ± 5.03 g (3E2 antibody treated mice) and 6.75 ± 2.06 g (14G2a antibody treated mice) (p> 0.05). These last two values are not significantly different from those found in healthy control animals (p> 0.1). The effect of treatment with monoclonal antibody 3E2 is not significantly different from treatment with monoclonal antibody 14G2a (p> 0.5). These data further confirm the specificity of 3E2 antibody therapy, since treatment with a control isotype antibody is completely ineffective.

 Because the 3E2 antibody is an IgM, whose distribution is limited to the intravascular compartment and because it recognizes blood vessels in solid tumors but not in healthy tissues, the results suggest that its therapeutic effects are related to its action. on these blood vessels.

7.3. Conclusion

These results show for the ^st time a passive immunotherapy with anti-Gb3 3E2 antibody is effective in suppressing the growth of tumor metastases in a relevant model of neuroblastoma syngeneic mice, which is similar to the human disease (Lode et al). This tumor model expresses the disialoganglioside GD2, a well-known neuroblastoma-associated antigen (Lode et al), but not Gb3, as demonstrated in vitro and in vivo.

 Similarly, the 3E2 antibody injection is also effective for inhibiting subcutaneous NXS2 neuroblastoma tumors in the A / J mouse.

The distribution of Gb3 within NXS2 metastases, demonstrated by labeling with 3E2 antibody and validated by CD31 staining of tissue sections, is clearly limited to the endothelial cells of the blood vessels of the tumor compartment. Therefore, the results obtained demonstrate that the anti-vascular activity of anti-Gb3 3E2 antibodies is sufficient to inhibit tumor growth in vivo. The results obtained also demonstrate that the anti-tumor efficacy of the anti-Gb3 3E2 antibody is comparable to that of the anti6GD2 14G2a antibody, which directly targets neuroblastoma NXS2 cells and has been clinically evaluated. with positive results (Murray et al). REFERENCES

- Ades, et al (1992). J Invest Dermatol 99, 683-690

 Almagro et al. Frontiers in Bioscience 13, 1619-1633, January 1, 2008, - Askari et al. 2007. Virchows Arch 451, 823-834

 - Bast DJ et al. Infect Immun. 1997 Jun; 65 (6): 2019-28,

 - Bonnaud et al. 2007. Cancer Res 67, 1803-1811

 - Bordron A et al. J Clin Invest. 1998 May 15; 101 (10): 2029-35,

 - Pike, X. et al. Nucl. Acids Res. 36, W503-508 (2008)

Chark D et al. Eur J Biochem. 2004 Jan; 271 (2): 405-17,

 - Chomczynski and Sacchi, (1987). Anal Biochem 162, 156-159.

 - Clackson et al. 1991, Nature 352, 624-628

 EP1176195A1,

 EP1331266A1,

- Falguieres et al. 2008. Mol Cancer Ther 7, 2498-2508

 Folch, J., et al. (1951). J Biol Chem 191, 833-841

 - Gariepy, 2001. Crit Rev Oncol Hematol 39, 99- 106

 - Gaugler et al. 2007. Radiat Res 167, 185-193

 - Hakomori and Kannagi, 1983. J Natl Cancer Inst 71, 231-251

Heath-Engel HM et al. Angiogenesis. 2003; 6 (2): 129-41,

 Janssen KP et al. Cancer Res. 2006 Jul 15; 66 (14): 7230-6,

 - Johansson et al. 2009, BMC Cancer 9, 67

 - Jones et al. Nature, 321: 522-525, 1986;

Kabat et al. Sequences of the proteins of immunological interest, 5 ^th ed., US Department of Health and Human Services, NIH, 1991, and later editions,

- Kanda et al. 2004. J Neurosci Res 78, 141-150

 - Khodarev et al. 2003. J Cell Sci 116, 1013-1022

 - Kotani et al. 1994. Arch Biochem Biophys 310, 89-96

 - LaCasse et al. 1999. Blood 94, 2901-2910

Ladisch and Li. Methods Enzymol. 2000; 312: 135-45; Lefranc and Lefranc, 1997 Combinatorial library antibody engineering. eds. Lefranc, MP and Lefranc, G. INSERM editions, Paris

 - Lefranc, M.-P. et al. Dev. Comp. Immunol, 27, 55-77 (2003),

 Lingwood et al. Bioscience Reports, Vol. 19, No. 5, 1999,

- Lode et al. Proc Natl Acad Sci U S A 1999 Feb 16; 96 (4): 1591-6,

 - Mangeney et al. 1993. Cancer Res 53, 5314-5319

 - Brands Filho et al. 2006. Otorhinolaryngol 72, 25-30

 - Morita, Y., et al. (2000). Nat Med 6, 1109-1114,

 Murray JL, et al. J Clin Oncol. 1994; 12: 184-93,

- Muthing et al. 1998. Carbohydr Res 307, 147-157

 - Muthing et al. 1999. Glycobiology 9, 459-468

 Nikolayenko IV et al. Ukrainica Bioorganica Acta 2 (2005) 3- 11

 Obrig TG et al. Infect Immun. 1988 Sep; 56 (9): 2373-8,

 - Obrig et al. 1993. J Biol Chem 268, 15484-15488

- Riechmann et al. Nature, 332: 323-327, 1988,

 - Sanger et al. 1977, Proc Natl Acad Sci U S A 74, 5463-5467

 Shields et al. Journal of Biological Chemistry. Flight. 276, No. 9, Issue of March 2, pp. 6591-6604, 2001,

 - Taga et al. 1997. Blood 90, 2757-2767

- Tétaud C et al. J Biol Chem. 2003 Nov 14; 278 (46): 45200-8,

 - US7432357B2,

 Verhoeyn et al. BioEssays, 8:74, 1988,

 - Verhoeyen et al. Science, 239: 1534-1536, 1988;

 - Viel et al. Mol Imaging. Nov-Dec; 7 (6): 239-47;

- Wiels et al. 1981. Proc Natl Acad Sci U S A 78, 6485-6488,

 - W098 / 51326,

 - WO01 / 77181,

 - Zemunik et al. 2004. Biochemistry (Mosc) 69, 513-519

Claims

An antibody directed against the glycosphingolipid membrane globotriaosylceramide (Gb3), or a functional fragment or a derivative thereof, characterized in that it has at least one complementarity determining region (CDR) having an acid sequence amino acids selected from SEQ ID NO: 1 to 42, or having an amino acid sequence of at least 80%, preferably at least 85%>, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 1 to 42.

2. Antibody, functional fragment or derivative thereof according to claim 1, characterized in that it has a heavy chain comprising at least one complementarity determining region (CDR) having an amino acid sequence selected from SEQ ID NO: 1 to 3, 7 to 9, 13 to 15, 19 to 21, 25 to 27, 31 to 33, and 37 to 39, or having an amino acid sequence of at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 1 to 3, 7 to 9, 13 to 15, 19 to 21, 25 to 27, 31 to 33, and 37 to 39.

3. Antibody, functional fragment or derivative thereof according to claim 1 or 2, characterized in that it has a light chain comprising at least one complementarity determining region (CDR) having a chosen amino acid sequence. from SEQ ID NO: 4 to 6, 10 to 12, 16 to 18, 22 to 24, 28 to 30, 34 to 36, and 40 to 42, or having an amino acid sequence of at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with one of SEQ ID NO: 4 to 6, 10 to 12, 16 to 18, 22 to 24, 28 to 30, 34 to 36, and 40 to 42.

4. Antibody, functional fragment or derivative thereof according to any one of claims 1 to 3, characterized in that it has a heavy chain comprising three CDR-H (heavy chain CDR) having the following amino acid sequences, or sequences having at least 80%, preferably at least 85%>, at least 90%, at least 95%), at least 96% , at least 97%, at least 98%>, at least 99% identity with the following sequences:

 a) CDR1-H-3E2: SEQ ID NO: 1, CDR2-H-3E2: SEQ ID NO: 2, CDR3-H-3E2:

 SEQ ID NO: 3,

 b) CDR1-H-14C11: SEQ ID NO: 7, CDR2-H-14C11: SEQ ID NO: 8, CDR3-H-14C11: SEQ ID NO: 9,

 c) CDR1-H-15C11: SEQ ID NO: 13, CDR2-H-15C11: SEQ ID NO: 14, CDR3-H-15C11: SEQ ID NO: 15,

 d) CDR1-H-22F6: SEQ ID NO: 19, CDR2-H-22F6: SEQ ID NO: 20, CDR3-H-22F6: SEQ ID NO: 21,

 e) CDR1-H-25C10: SEQ ID NO: 25, CDR2-H-25C10: SEQ ID NO: 26, CDR3-H-25C10: SEQ ID NO: 27,

 f) CDR1-H-11E10: SEQ ID NO: 31, CDR2-H-11E10: SEQ ID NO: 32, CDR3-H-1E10: SEQ ID NO: 33, or

 g) CDR1-H-16G8: SEQ ID NO: 37, CDR2-H-16G8: SEQ ID NO: 38, CDR3-H-16G8: SEQ ID NO: 39.

Antibody, functional fragment or derivative thereof according to any one of claims 1 to 4, characterized in that it has a light chain comprising three CDR-L (light chain CDR) having the acid sequences. amino acids, or sequences having at least 80%, preferably at least 85%, at least 90%, at least 95%), at least 96%, at least 97%, at least 98%, at least 99% by weight, identity with the following sequences:

 a) CDR1-L-3E2: SEQ ID NO: 4, CDR2-L-3E2: SEQ ID NO: 5, CDR3-L-3E2:

 SEQ ID NO: 6,

 b) CDR1-L-14C11: SEQ ID NO: 10, CDR2-L-14C11: SEQ ID NO: 11, CDR3-L-14C11: SEQ ID NO: 12,

c) CDR1-L-15C11: SEQ ID NO: 16, CDR2-L-15C11: SEQ ID NO: 17, CDR3-L-15C11: SEQ ID NO: 18, d) CDR1-L-22F6: SEQ ID NO: 22, CDR2-L-22F6: SEQ ID NO: 23, CDR3-L-22F6: SEQ ID NO: 24,

 e) CDR1-L-25C10: SEQ ID NO: 28, CDR2-L-25C10: SEQ ID NO: 29, CDR3-L-25C10: SEQ ID NO: 30,

 f) CDR1-L-11E10: SEQ ID NO: 34, CDR2-L-11E10: SEQ ID NO: 35, CDR3-

L-1E10: SEQ ID NO: 36, or

 g) CDR1-L-16G8: SEQ ID NO: 40, CDR2-L-16G8: SEQ ID NO: 41, CDR3-L-16G8: SEQ ID NO: 42.

6. Antibody, functional fragment or derivative thereof according to any one of claims 1 to 5, characterized in that it has a heavy chain comprising a variable region having a sequence chosen from SEQ ID NO: 43 to 49 or a sequence having at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the one of SEQ ID NO: 43-49.

7. Antibody, functional fragment or derivative thereof according to any one of claims 1 to 6, characterized in that it has a light chain comprising a variable region having a sequence selected from SEQ ID NO: 50 to 56 or a sequence having at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the one of SEQ ID NO: 50-56.

8. Antibody, functional fragment or derivative thereof according to any one of claims 1 to 7, characterized in that it has heavy and light chains whose variable regions have the following amino acid sequences or sequences having at least 80%, preferably at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the following sequences:

 a) Antibody 3E2: heavy chain: SEQ ID NO: 43, light chain: SEQ ID NO:

50 b) 14C11 antibody: heavy chain: SEQ ID NO: 44, light chain: SEQ ID NO: 51,

 c) 15C11 Antibody: Heavy Chain: SEQ ID NO: 45 Light Chain: SEQ ID NO:

 52

 d) 22F6 antibody: heavy chain: SEQ ID NO: 46, light chain: SEQ ID NO:

 53

 e) 25C10 antibody: heavy chain: SEQ ID NO: 47, light chain: SEQ ID NO: 54,

 f) 11E10 antibody: heavy chain: SEQ ID NO: 48 light chain: SEQ ID NO:

 55, and

 g) 16G8 antibody: heavy chain: SEQ ID NO: 49, light chain: SEQ ID NO:

 56.

9. Antibody, functional fragment or derivative thereof according to claim 8, characterized in that it has heavy and light chains whose variable regions have the following amino acid sequences or sequences having at least 80%, preferably at least 85%>, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% identity with the following sequences:

 a) Antibody 3E2: heavy chain: SEQ ID NO: 43, light chain: SEQ ID NO:

 50, and

 b) 22F6 antibody: heavy chain: SEQ ID NO: 46, light chain: SEQ ID NO:

 53.

Antibody, functional fragment or derivative thereof according to any one of claims 1 to 9, characterized in that it is of IgG or IgM isotype.

11. Antibody, functional fragment or derivative thereof according to any one of claims 1 to 10, characterized in that it is a chimeric or humanized antibody.

12. Functional fragment of antibody according to any one of claims 1 to 11, characterized in that it is selected from Fv, ScFv, Fab, F (ab ') 2, Fab', scFv-Fc fragments or Diabodies ("diabodies").

13. Antibody, functional fragment or derivative thereof according to any one of claims 1 to 12, characterized in that it is not coupled to a cytotoxic molecule.

14. Nucleic acid encoding an antibody according to any one of claims 1 to 13.

A vector comprising a nucleic acid according to claim 14.

16. Host cells comprising a nucleic acid according to claim 14 or a vector according to claim 15.

An antibody, functional fragment or derivative thereof, directed against the Gb3 membrane glycosphingolipid and not coupled to a therapeutic molecule, for use as a medicament in the treatment of angiogenesis-associated diseases.

An antibody, functional fragment or derivative thereof according to claim 17 for the treatment of solid tumors; psoriasis; angiomas; proliferative eye diseases, including age-related macular degeneration (AMD), diabetic retinopathy, or neovascular glaucoma; autoimmune diseases, including lupus or rheumatoid arthritis as well as atherosclerosis-related diseases.

19. Antibody, functional fragment or derivative thereof according to claim 18 for the treatment of adenomas and sarcomas; and carcinomas, including adenocarcinomas, ovarian, breast, pancreatic, skin, lung, brain, kidney, liver, nasopharyngeal cavity, thyroid, central nervous system, prostate, colon, rectum, cervix, testes, or bladder.

Antibody, functional fragment or derivative thereof according to any one of claims 17 to 19, characterized in that the antibody is as defined in claims 1 to 13.

21. Antibody, functional fragment or derivative thereof according to any one of claims 17 to 20, characterized in that it recognizes the upper band but not the lower band of the doublet Gb3 expressed by the HMEC-1 cells.