ZA200206317B - Partners of PTB1 domain of FE65, preparation and uses. - Google Patents

Description

a v PS

PARTNERS OF THE PTBl1 DOMAIN OF FE65, PREPARATION AND

USES

Alzheimer’s disease (AD) is a neuro- degenerative disease which affects a large proportion of the elderly population. This disease is characterized, clinically speaking, by the loss of cognitive functions, and, neuropathologically speaking, by the presence in the brain of intracellular neurofibrilliary deposits and of extracellular deposits of the P-amyloid (AB) peptide, which form amyloid plaques (Yankner, 1996). Amyloid plaques are mainly composed of the AP peptides having 40 or 42 amino acids, which are generated by a proteolytic process from the precursor protein of the fB-amyloid peptide (APP) (Golde et al., 1992). The extracellular deposits of AB are specific for AD. They are the early and invariable feature of all the forms of AD, including the hereditary forms. These hereditary forms of the disease appear relatively early on (between 40 and 60 years of age) and are due to mutations in the APP gene and in the presenilin 1 (PS1l) and presenilin 2 (PS2) genes. The mutations in these three genes induce changes in the proteolysis of the APP, which lead to an overproduction of AB and to the early appearance of the pathology and of the symptoms, which are similar to those of the sporadic forms of AD.

‘» hg

Internalization of the membrane APP is a step which is required for the process of proteolysis of the

APP (Koo and Sqguazzo, 1994), which depends on its cytoplasmic domain. Specifically, the deletion of this region of the protein, or the presence of point mutations in the sequence Tyr-Glu-Asn-Pro-Thr-Tyr in the cytoplasmic domain of the APP, induces a considerable decrease in the production of the B- amyloid peptide (Perez et al., 1999). Several proteins have been identified as interacting with the cytoplasmic domain of the APP; these proteins might thus participate in the regulation of the proteolytic process of the APP and thus in the production of the f- amyloid peptide. The two protein families FE65 and X11 which interact with the sequence Tyr-Glu-Asn-Pro-Thr-

Tyr of the cytoplasmic domain of the APP (Borg et al., 1996, 1998; Bressler et al., 1996; Duilio et al., 1998;

Fiore et al., 1995; Guenette et al., 1996; McLoughlin and Miller, 1996; Mercken et al., 1998; Tanahashi and

Tabira, 199%a, 1999 and 1999c) should be mentioned.

The FE65 protein family consists of three members which are called FE65, COFE65/FE65L1 and FE65L2. The X11 protein family also consists of three members, which are called Xlla, X11B, and X1lly. These two protein families have opposite effects on the regulation of the production of the P-amyloid peptide. We have shown, as has another laboratory, that the overexpression of FE65 induces an increase in the production of the AP peptide

® : (Mercken et al., 1998; Sabo et al., 1999), whereas the overexpression of X11 induces a decrease in the production of the AP peptide (Borg et al., 1998; Sastre et al., 1998).

Analysis of the primary structure of FE65 indicates that this protein probably plays the role of adapter. Specifically, FE65 contains three protein domains which are involved in protein-protein interactions: a WW domain in the amino-terminal half and two PTB domains (PhosphoTyrosine Binding domain), called PTBl1 and PTB2, in the carboxy-terminal half. The construction of deletions has shown that the PTB2 domain of FE65 is involved in the interaction with the cytoplasmic demain of the APP. The WW domain interacts with at least five proteins, of which two have been identified as being the protein Mena (Mammalian homolog of Enabled) (Ermekova et al., 1997). In addition, the

PTBl domain of FE65 interacts with the transcription factor CP2/LSF/LBP1 (Zambrano et al., 1997) and with the receptor LRP (LDL receptor-Related Protein) (Trommsdorff et al., 1998). The role of these proteins in the physiolecgical function of FE65 is not known to date.

The elucidation of the exact role of the FE65 protein in the process of production of the B-amyloid peptide thus constitutes a major asset for the understanding of, and the therapeutic approach to,

‘» ®

Alzheimer’s disease and more generally neurodegenerative diseases.

The present invention lies in the identification of partners of the FE65 protein which interact with this protein under physiological conditions. These partners represent novel pharmacological targets for the manufacture or the investigation of compounds which are capable of modulating the activity of FE65, in particular its activity on the production of the P-amyloid peptide.

These proteins, the antibodies, the corresponding nucleic acids and the specific probes or primers can also be used for detecting or for assaying the proteins in biological samples, in particular nervous tissue samples. These proteins or nucleic acids can also be used in therapeutic approaches, to modulate the activity of FE65 and any compound according to the invention which is capable of modulating the interaction between FE65 and the polypeptides of the invention.

The present invention results more particularly from the revelation, by the applicant, of two human proteins which interact with the PTB1 domain of FE65 (represented on the sequence SEQ ID No. :1 and 2). Thus, the present invention shows that the central region of the protein hnRNPL interacts with the PTBI1 domain of FE65. It also describes the identification of a novel protein, termed FEBPl (FE65 Binding PTBl domain a ® : protein), which is capable of interacting with the PTBI domain of FE65.

The present invention also results from the identification and from the characterization of specific regions of the hnRNPL and FEBPl proteins above, which are involved in the modulation of the function of the FE65 protein. The demonstration of the existence of these proteins and of regions which are involved in their function makes it possible in particular to prepare novel compounds and/or compositions which can be used as pharmaceutical agents, and to develop industrial methods for screening such compounds.

A first subject of the invention thus relates to compounds which are capable of modulating, at least partially, the interaction of the hnRNPL and/or FEBP1 proteins (or homologs thereof) with the PTBl domain of

FE65, or of interfering with this reaction.

The interference of a compound according to the invention can reveal itself in various ways. The compound according to the invention can slow, inhibit or stimulate, at least partially, the interaction between an hnRNPL and/or FEBPl protein (or homologs thereof) and the PTBl domain of FE65. They are preferably compounds which are capable of modulating this interaction in vitro, for example in a system of double-hybrid type or in any acellular system for detecting an interaction between two polypeptides. The

J 6 compounds according to the invention are preferably compounds which are capable of modulating, at least partially, this interaction, preferably by increasing or inhibiting this interaction by at least 20%, more preferably by at least 50%, with respect to a control in the absence of the compound.

For the purposes of the present invention, the name of the proteins hnRNPL and FEBP1l covers the proteins per se and all homologous forms thereof. “Homologous form” is intended to refer to any proteins which are equivalent to the protein under consideration, of various cellular origin and in particular derived from cells of human origin, or other organisms, and which possess an activity of the same type. Such homologs also comprise the natural variants of the proteins indicated, in particular the polymorphic or splicing variants. The homologous proteins (or polypeptides) can be obtained, for example, by experiments of hybridization between the coding nucleic acids. For the purposes of the invention, a sequence of this type only has to have a significant percentage of identity to lead to a physiological behavior which is comparable to those of the hnRNPL and/or FEBPl proteins as claimed.

According to a particular embodiment, the compounds of the invention are capable of binding at the level of the domain of interaction between the

« ® : hnRNPL and/or FEBP1l proteins and the PTBl domain of

FEGS.

The compounds according to the present invention can be of varied nature and origin. In particular, they can be compounds of peptide, nucleic acid (i.e. comprising a string of bases, in particular a DNA or an RNA molecule), lipid or saccharide type, an antibody, etc. and, more generally, any organic or inorganic molecule.

According to a first variant, the compounds of the invention are of peptide nature. The term . “peptide” refers to any molecule comprising a string of amino acids, such as for example a peptide, a polypeptide, a protein, an antibody (or antibody fragment or derivative), which if necessary is modified or combined with other compounds or chemical groups. In this respect, the term “peptide” refers more specifically to a molecule comprising a string of at most 50 amino acids, more preferably at most 40 amino acids. A polypeptide comprises preferably from 50 to 500 amino acids, or more. A protein is a polypeptide corresponding to a natural molecule.

According to a first preferred embodiment, the peptide compounds of the invention comprise a portion of the peptide sequence of the hnRNPL protein and/or of the FEBPl protein and/or of derivatives thereof. It is more particularly a portion of the sequence of the hnRNPL protein and/or of the FEBPI

‘. ® ; protein, said proteins being characterized, respectively, in that they comprise the sequences SEQ

ID No. 7 and SEQ ID No. 9.

Peptide compounds according to the invention are more preferably compounds comprising a region whose sequence corresponds to all or a functional portion of the site of interaction of the hnRNPL protein and/or the FEBPl protein with the PTB1 domain of FE65. Such compounds, in particular peptides, constitute competitors of hnRNPL and/or FEBP1l, and are capable of modulating, at least partially, the interaction between the hnRNPL protein and/or the FEBP1l protein (and/or homologous forms) and the PTBl domain of FE65. They are more preferably peptide compounds comprising residues 1 to 349 of the sequence SEQ ID No. :7 or residues 1 to 337 of the sequence SEQ ID No. :9. As indicated in the examples, these sequences comprise at least one portion of the central region of the hnRNPL (residues 116 to 464) and FEBPl proteins, and are capable of interacting specifically with the PTBl domain of FE65 and not with the PTB2 domain of FEG65.

In a specific embodiment, the compound is a fragment of the sequence SEQ ID No. :7, of at least 5 amino acids, preferably at least 9 amino acids, comprising the sequence Asn-Pro-Ile-Tyr (residues 55 to 58).

According to another preferred embodiment, the peptide compounds of the invention are compounds which are derived from the hnRNPL protein or from the

FEBP1 protein (and/or from the homologous forms) and which bear an effector region which has been made nonfunctional. Such peptide compounds can be obtained by deletion, mutation or disruption of at least this effector region in the hnRNPL protein and/or the FEBP1 protein and/or the homologous forms. Such modifications can be carried out for example by in vitro mutagenesis, by introduction of additional elements or of synthetic sequences, or by deletions or substitutions of the original elements. These polypeptides thus have the capacity to bind the FE65 protein, but cannot induce a functional signal, at least not to the same degree as the native proteins.

According to a specific embodiment, it is a polypeptide which comprises the sequence SEQ ID No. :7 or SEQ ID No. :9 and which bears at least one mutation in the effector region.

According to a specific embodiment, it is a polypeptide which comprises the sequence SEQ ID No. :7 or SEQ ID No. :9 and which bears at least one deletion in the effector region.

According to a specific embodiment, it is a polypeptide which comprises the sequence SEQ ID No. :7 or SEQ ID No. :9 and which bears at least one insertion in the effector region.

Another specific subject of the invention lies in the FEBPl protein, or any fragment or

® 10 derivative of this protein. It is more particularly any polypeptide comprising the sequence SEQ ID No. :9 or a derivative or fragment of this sequence, even more preferably any polypeptide comprising at least 10 consecutive residues of the sequence SEQ ID No. :9 or of a derivative of this sequence, even more preferably comprising at least the residues which are involved in the binding to the PTBl domain of FE65.

Another subject of the invention lies in a polypeptide comprising the sequence SEQ ID No. :7.

The term “derivative” refers more particularly, for the purposes of the present invention, to any sequences which differ from the sequence under consideration because of a degeneracy of the genetic code, which is obtained by one or more modifications of genetic and/or chemical nature, as well as any peptide which is encoded by a sequence which hybridizes with the nucleic acid sequences SEQ ID

No. :6 or 8, or fragments thereof, and which has the capacity to interfere with the interaction between the hnRNPL protein and/or the FEBPl protein and/or a homolog thereof, and the PTBl domain of FE65. The expression “modification of genetic and/or chemical nature” can be understood to mean any mutation, substitution, deletion, addition and/or modification of one or more residues. The term “derivative” also comprises the sequences which are homologous to the sequence under consideration, which are derived from

® 11 other cellular scurces and in particular from cells of human origin, or from other organisms, and which have an activity of the same type. Such homologous sequences can be obtained by hybridization experiments. The

S hybridizations can be carried out using nucleic acid libraries, using the native sequence or a fragment thereof as probe, under variable conditions of hybridization (Sambrook et al., cf. General techniques of molecular biology). Moreover, the term “fragment” or “portion” refers to any portion of the molecule under consideration, which comprises at least 5 consecutive residues, preferably at least 9 consecutive residues, even more preferably at least 15 consecutive residues.

Such derivatives or fragments can be generated with different aims, such as in particular that of increasing their therapeutic efficacy or of reducing their side effects, or that of conferring on them novel pharmacokinetic and/or biological properties.

When a derivative or fragment as defined above 1s prepared, its biological activity on the binding of the hnRNPL protein and/or of the FEBP1 protein, and/or of the homologous forms, to their binding site on the PTBl domain of FE65 can be demonstrated. Any technique which is known to persons skilled in the art can of course be used for this, as is explained in the experimental section (double-

® 12 hybrid, immobilization on a column, acellular system, cellular system, etc.).

Generally, the compounds of the invention can be any fragment of the hnRNPL or FEBPl proteins or of the peptide compounds indicated above. Such fragments can be generated in various ways. In particular, they can be synthesized chemically, on the basis of the sequences given in the present application, using the peptide synthesizers known to persons skilled in the art. They can also be synthesized genetically, by expression in a host cell of a nucleotide sequence encoding the desired peptide. In this case, the nucleotide sequence can be prepared chemically, using an oligonucleotide synthesizer, on the basis of the peptide sequence given in the present application and of the genetic code. The nucleotide sequence can also be prepared from sequences given in the present application, by enzymatic cleavage, ligation, cloning, etc., according to the techniques known to persons skilled in the art, or by screening of DNA libraries with probes which are developed from these sequences.

Other peptides according to the invention are peptides which are capable of competing with the peptides defined above, for the interaction with their cellular target. Such peptides can be synthesized in particular on the basis of the sequence of the peptide under consideration, and their capacity for competing with the peptides defined above can be determined.

® 13

According to another particular embodiment, the compounds of the invention are antibodies, or antibody fragments or derivatives. Thus, another subject of the invention lies in polyclonal or monoclonal antibodies or fragments of antibodies, which are directed against a peptide compound or a protein as defined above. Such antibodies can be generated by methods known to persons skilled in the art. In particular, these antibodies can be prepared by immunizing an animal against a peptide of the invention, by sampling blood, and by isolating the antibodies. These antibodies can also be generated by preparing hybridomas according to the techniques known to persons skilled in the art.

More preferably, the antibodies or antibody fragments of the invention have the capacity to modulate, at least partially, the interaction of the peptide compounds defined above or of the hnRNPL and/or

FEBP1l proteins with the PTB1 domain of FE65, and can be used to modulate the activity of FE65.

Moreover, these antibodies can also be used to detect and/or to assay the expression of the claimed peptides in bioclogical samples, and consequently to provide information on their activation state.

The antibody fragments or derivatives are, for example, Fab or Fab’2 fragments, single-chain (ScFv) antibodies, etc. They are in particular any fragment or derivative which conserves the antigenic

® y specificity of the antibodies from which they are derived.

The antibodies according to the invention are more preferably capable of binding the hnRNPL and/or

FEBP1l proteins which comprise respectively the sequence

SEQ ID Nos. :7 or 9, in particular the region of these proteins which is involved in the interaction with

FE65. These antibodies (or fragments or derivatives) are more preferably capable of binding an epitope which is present in the sequence between residues 1 and 349 of SEQ ID No. :7 or between residues 1 and 337 of

SEQ ID No. 6.

The invention also relates to compounds which are not peptides or which are not exclusively peptides, which are capable of interfering with the abovementioned interaction, and to their use as pharmaceutical agents. It is in fact possible, from the active protein motifs described in the present application, to produce molecules which are modulators of the activity of the hnRNPL and/or FEBPl proteins, which are not exclusively peptides, and which are compatible with pharmaceutical use, in particular by duplicating the active motifs of the peptides with a structure which is not a peptide, or which is not exclusively peptide in nature.

A subject of the present invention is also any nucleic acid which encodes a peptide compound according to the invention. It can in particular be a

® 15 sequence comprising all or part of the sequences which are presented in SEQ ID No. :6 and in SEQ ID No. :8, or of derivatives thereof. “Derived sequence” is intended to mean, for the purposes of the present invention, any sequence which hybridizes with the sequences which are presented in SEQ ID No. :6 and in SEQ ID No. :8, or with a fragment of these sequences, which encodes a peptide according to the invention, as well as the sequences which result from these latter by degeneracy of the genetic code. The various nucleotide sequences of the invention may or may not be of artificial origin. They can be genomic, cDNA or RNA sequences, hybrid sequences, or synthetic or semi-synthetic sequences. These sequences can be obtained either by screening DNA libraries (cDNA library, genomic DNA library) or by chemical synthesis or by mixed methods which include the chemical or enzymatic modification of sequences which are obtained by screening of libraries.

The abovementioned hybridization is preferably carried out under conditions of high stringency, and in particular at a temperature of 50°C for 1 hour in a solution containing 8.823 gram/l of trisodium citrate- 2H20, 17.532 g/1 of sodium chloride and 1% sodium dodecyl sulfate, or alternatively under the conditions described by Sambrook et al. (1989, pages 9.52-9.55).

A particular nucleic acid, for the purposes of the invention, encodes a polypeptide which comprises the sequence SEQ ID No. :9 or a fragment or derivative

® lo of this sequence, in particular the human FEBPI protein. It is advantageously a nucleic acid which comprises the nucleic sequence SEQ ID No. :8.

The nucleic acids according to the invention can be used for producing the peptide compounds of the invention. The present application thus also relates to a method for preparing a peptide compound, according to which a cell which contains a nucleic acid according to the invention is cultured, under conditions for expression of said nucleic acid, and the peptide compound produced is recovered. In this case, the portion which encodes said polypeptide is generally placed under the control of signals which allow its expression in a host cell. The choice of these signals (promoters, terminators, secretion leader sequence, etc.) can vary as a function of the host cell used.

Moreover, the nucleic acids of the invention can be part of a vector which can replicate autonomously, or which can integrate. More particularly, autonomously- replicating vectors can be prepared using sequences which replicate autonomously in the chosen host. As regards the integrating vectors, these can be prepared for example using sequences which are homologous to certain regions of the host genome, which allows the integration of the vector by homologous recombination.

It can be a vector of plasmid, episomal, chromosomal, viral, etc. type.

® 17

The host cells which can be used for producing the peptides of the invention via the recombinant pathway, are both eukaryotic and prokaryotic hosts. Among the eukaryotic hosts which are suitable, mention may be made of animal cells, yeasts or fungi. In particular, as regards yeasts, mention may be made of yeasts of the genus Saccharomyces,

Kluyveromyces, Pichia, Schwanniomyces or Hansenula. As regards animal cells, mention may be made of COs, CHO,

C127, PCl2 cells, etc. Among the fungi, mention may more particularly be made of Aspergillus ssp. or

Trichoderma ssp. As prokaryotic hosts, use of the following bacteria is preferred: E. coli, Bacillus, or

Streptomyces.

The nucleic acids according to the invention can also be used to prepare genetic antisense or antisense oligonucleotides which can be used as pharmaceutical or diagnostic agents. Antisense sequences are oligonucleotides of short length, which are complementary to the coding strand of a given gene, and which, for this reason, are capable of specifically hybridizing with the mRNA transcript, which inhibits its translation into a protein. A subject of the invention is thus the antisense sequences which are capable of at least partially inhibiting the interaction of the hnRNPL and/or FEBPl proteins on the

PTB1 domain of FE65. Such sequences can consist of all or part of the nucleotide sequences defined above. They

® 18 are generally sequences or fragments of sequences which are complementary to sequences encoding peptides which interact with the PTB1l domain of FE65. Such oligonucleotides can be obtained by fragmentation or by chemical synthesis, etc.

The nucleic acid sequences can also be used in the context of therapies, for transferring and expressing, in vivo, antisense sequences or peptides which are capable of modulating the interaction of the hnRNPL and/or FEBPl proteins with the PTBl domain of

FE65. In this respect, the sequences can be incorporated into viral or nonviral vectors, which allows their administration in vivo (Kahn, A. et al. 1991). As viral vectors in accordance with the invention, mention may be made most particularly of vectors of adenovirus, retrovirus, adenovirus- associated virus (AAV) or herpesvirus type. A subject of the present application is also defective recombinant viruses comprising a nucleic acid which encodes a (poly)peptide according to the invention.

The invention also allows the preparation of nucleotide probes which may or may not be synthetic, and which are capable of hybridizing with the nucleic acids defined above or with their complementary strand.

Such probes can be used in vitro as a diagnostic tool, for detecting the expression or overexpression of the hnRNPL and/or FEBPl proteins, or alternatively for detecting genetic abnormalities (incorrect splicing,

® 19 polymorphisms, point mutations, etc.). These probes can also be used for detecting and isolating homologous nucleic acid sequences which encode peptides as defined above, using other cellular sources and preferably «cells of human origin. The probes of the invention generally comprise at least 10 bases, and they can for example comprise up to the whole of one of the abovementioned sequences or of their complementary strand. Preferably, these probes are labeled prior to their use. For this, various techniques known to a person skilled in the art can be employed (radioactive, fluorescent, enzymatic, chemical labeling etc.).

A subject of the invention is also any pharmaceutical composition which comprises, as an active principle, at least one compound as defined above, in particular a peptide compound.

A subject of the invention is in particular any pharmaceutical composition which comprises, as an active principle, at least one antibody and/or one antibody fragment as defined above, as well as any pharmaceutical composition which comprises, as an active principle, at least one nucleic acid or one vector as defined above.

A subject of the invention is also any pharmaceutical composition which comprises, as an active principle, a chemical molecule which is capable of increasing or of decreasing the interaction between the hnRNPL and/or FEBPl proteins and the FE65 protein.

® 20

Moreover, a subject of the invention is also pharmaceutical compositions in which the peptides, antibodies, chemical molecules and nucleotide sequences defined above are combined, mutually or with other active principles.

The pharmaceutical compositions according to the invention can be used to modulate the activity of the hnRNPL and/or FEBP1 proteins, and consequently, can modify the function of the APP, its intracellular transport, its maturation and its conversion into the

B-amyloid peptide. More particularly, these pharmaceutical compositions are intended for modulating the interaction between the hnRNPL or FEBP1l proteins and the FE65 protein. They are more preterably pharmaceutical compositions which are intended for treating neurodegenerative diseases, such as for example Alzheimer’s disease. The compositions (or compounds) of the invention are more particularly intended for inhibiting, at least partially, the interaction between the FE65 protein and the hnRNPL and/or FEBPl protein.

A subject of the invention is also the use of the molecules described above for modulating the activity of the FE65 protein or for typing neurodegenerative diseases. In particular, the invention relates to the use of these molecules for modulating, at least partially, the activity of the

PTB1 domain of FE65.

° 21

The invention also relates to a method for screening or characterizing molecules which have an action on the function of the FE65 protein, which comprises selecting molecules which are capable of binding the sequence SEQ ID No. :7 or the sequence SEQ

ID No. :9, or a fragment (or derivative) of these sequences. The method advantageously comprises bringing the molecule(s) to be tested into contact, in vitro, with a polypeptide which comprises the sequence SEQ ID

No. :7 or the sequence SEQ ID No. :9, or a fragment (or derivative) of these sequences, and selecting molecules which are capable of binding the sequence SEQ ID No. :7 (in particular the region between residues 1 and 349) or the sequence SEQ ID No. :9 (in particular the region between residues 1 and 337). The molecules tested can be of varied nature (peptide, nucleic acid, lipid, sugar, etc., or mixtures of such molecules, for example combinatory libraries, etc.). As indicated above, the molecules thus identified can be used to modulate the activity of the FE65 protein, and represent potential therapeutic agents for treating neurodegenerative pathologies.

Other advantages of the present invention will appear on reading the following examples. They should be considered as illustrations and nonlimiting.

® 22

MATERIAL AND TECHNIQUES USED

1) Yeast strains used:

The strain L40 of the genus S. cerevisiae 5S (Mata. his3D200, trpl-901, leu2-3,112, ade2, LYS2:: (lexAop)4~HIS3, URA3:: (lexAop)8-LacZ, GAL4, GAL80) was used as a tool for screening the brain fusion library by the two-hybrid system. This strain enables the detection of a protein-protein interaction when one of the protein partners is fused to the LexA protein (Vojtek et al., 1993). It was cultured on the following culture medium:

Minimum YNB medium:-Yeast Nitrogen Base (without amino acids) (6.7 g/l) (Difco) - Glucose (20g/1) (Merck)

This medium can be solidified by addition of g/1 of agar (Difco).

To enable auxotrophic yeast to grow on this medium, it is necessary to add thereto the nitrogen- 20 containing amino acids or bases on which they are dependent, at 50 mg/ml. 100 pg/ml of ampicillin are added to the medium to avoid bacterial contamination. 2) Bacterial strains used:

Strain TGl of Escherichia coli, of genotype supE, hsdA5, thi, A(lacproAB), F’[traD36 pro A'B’ lacI%aczAM15] was used for constructing plasmids and

® 23 for amplifying and isolating plasmids. It was cultured on the following medium:

LB medium: - NaCl (5g/l) (Sigma) ~ Bactotryptone (10g/l) (Difco) - Yeast extract (5g/l1l) (Difco)

This medium can be solidified by addition of 20 g/1 of agar (Difco).

Ampicillin at 100 pg/ml was used to select the bacteria which had received the plasmids bearing the gene for resistance to this antibiotic, as marker. 3) Plasmids used:

The vector pGAD10, supplied by Clontech® allows the expression, in the yeast, of fusion proteins in which fusion is between the transactivating domain of GAL4 and a protein which is encoded by the cDNA originating from a brain library.

The vector plex9 (pBTM1l1l6) (Bartel et al., 1993) allows the expression, in the yeast, of fusion proteins in which fusion is with the protein LexA.

The vector pGAD424 (Clontech®) allows the expression, in the yeast, of fusion proteins in which fusion is with the transactivating domain of GAL4. pLex-FE65PTBl; plasmid pLexS8 which contains the sequence encoding the PTB1 domain of the FE65 protein (amino acids 395 to 543). This plasmid was used for screening protein partners of the PTBl domain of

FE65.

® 24 pLex-FE65PTB2; plasmid pLex9 which contains the sequence encoding the PTB2 domain of the FE65 protein (amino acids 565 to 698) which is known to interact with the cytoplasmic region of the APP (fB- amyloid peptide precursor). This plasmid was used to test the specificity of interaction of the hnRNPL and

FEBPl proteins with the PTB domains of FEG65. plex~HaRasVall2; plasmid pLex9 which contains the sequence encoding the HaRas protein which is mutated at position Vall2, and which is known to interact with the mammalian Raf protein (Vojtek et al., 1993). This plasmid was used to test the specificity of interaction of the hnRNPL and FEBPl proteins with FE65.

PGAD-Raf; plasmid pGAD424 which contains the sequence encoding the Raf protein (Vojtek et al., 1993). This plasmid was used to test the specificity of interaction of the hnRNPL and FEBPl proteins with FE65. pGAD-App:; plasmid pGAD10 which contains the sequence encoding the cytoplasmic domain of the APP protein which is known to interact with the PTB2 domain of FE65 (Mercken et al., 1998). This plasmid was used to test the specificity of interaction of the hnRNPL and FEBPl proteins with FE65. 4) Synthetic oligonucleotides used:

The oligonucleotides are synthesized using an

Applied System ABI 394-08 machine. They are removed from the synthesis matrix with ammonia and precipitated

® 25 twice with 10 volumes of n-butanol, then taken up in water. The quantification is carried out by measuring the optical density,

SEQ ID No. 3 : CTTCCCGGGTCCCCCACGGAATACCAAC

SEQ ID No. 4 : GGGGTCGACGGCATTACGCCGTTCGGC

These oligonucleotides made it possible to obtain the PCR fragment corresponding to the PTB1 domain of FE65 (represented on the sequence SEQ ID No. 1), and to introduce the sites Xmal and Sall at the ends (underlined).

SEQ ID No. 5 : CCACTACAATGGATGATG

This oligonucleotide (GAL4TA) was used to sequence the inserts contained in the plasmids from the brain cDNA double-hybrid library. 5) Preparation of the plasmid DNAs

The preparations of plasmid DNA, in small amounts and in large amounts, were carried out according to the protocols recommended by Quiagen, the manufacturer of the DNA purification kits: - Quiaprep Spin Miniprep kit, ref : 27106 - Quiaprep Plasmid Maxiprep kit, ref : 12163.

® 26 6) Enzymatic amplification of DNA by PCR (Polymerase Chain Reaction)

The PCR reactions are carried out in a final volume of 50ul in the presence of the DNA matrix, of dNTP (0.2 mM), of PCR buffer (10 mM Tris-HCl pH 8.5,

ImM MgCl,, 5 mM KCl, 0.01% gelatin), of 0.5 pg of each one of the oligonucleotides and of 2.5 IU of Ampli Tag

DNA polymerase (Perkin Elmer) with or without formamide (5%). The mixture is covered with 2 drops of liquid petroleum jelly, to limit the evaporation of the sample. The machine used is the “Crocodile II” from

Appligene.

We used a matrix denaturation temperature of 90°C, a hybridization temperature of 50°C and a temperature of elongation by the enzyme of 72°C. 7) Ligations

All the ligation reactions are carried out at +14°C overnight in a final volume of 10 ul, in the presence of 100 to 200 ng of vector, 0.5 to 2 ug of insert, 40 IU of T4 DNA ligase enzyme (Biolabs) and a ligation buffer (50 mM Tris-HCl pH 7.8; 10 mM MgCl;; 10 mM DTT; 1 mM ATP). 8) Transformation of bacteria:

The transformation of bacteria with a plasmid is carried out according to the following protocol: the entire ligation volume (10pl) is used to transform TGl

® 27 bacteria which are made competent by the method of

Chung et al., (1988). 9) Separation and extraction of the DNAs:

The separation and the extraction of the DNA fragments are carried out according to Sambrook et al., (1989). 10) Fluorescent sequencing of the plasmid

DNAs

The sequencing technique used is derived from the method of Sanger et al., (1997), and is adapted for sequencing by fluorescence and developed by Applied

Biosystems. The protocol used is that described by the designers of the system (Perkin Elmer). 11) Preparation of plasmids from the brain library

This preparation was carried out according to the recommendations of the supplier (Clontech®). 12) Transformation of yeast with a plasmid

The yeast are made competent by treating with

LiAC/PEG according to the method described by Gietz et al., (1995).

In the specific case of the transformation of yeast with the brain ¢DNA library, 250 ml, at 107 cells/ml of a culture, in YNB+His+Lys+Ade+Leu

® 28 minimum medium, of yeast containing the plasmid pLex-

FE65PTB1l is used. The yeasts, which are made competent according to the abovementioned protocol, are transformed with 30 pug of cDNA from the brain library.

After the transformation steps, the yeasts are put back into culture in 250 ml of YNB+His+Lys+Ade+Leu at 28°C for 16 hours, then recovered by centrifugation to be plated on a YNB+Lys+Ade medium and incubated for 3 days at 28°C. Determination of the efficacy of transformation and of the level of amplification was carried out according to the Clontech® protocol. 13) Extraction of the DNA (genomic and plasmid) from yeast 3 ml of a yeast culture which has been incubated for 16 h at 30°C are centrifuged and taken up in 200 pl of a lysis buffer (1M Sorbitol, 0.1M

KH,PO4/K;HPOs pH 7.4, 12.5 mg/ml zymolyase) and incubated for 1h at 37°C. The lysate is then treated according to the protocol recommended by Quiagen, the manufacturer of the DNA purification kit; Quiaprep Spin Miniprep kit, ref : 27106. 14) B-galactosidase activity assay

A sheet of nitrocellulose is preplaced on the

Petri dish containing the yeast clones, which are separated from each other. The sheet is then immersed in liquid nitrogen for 30 seconds to rupture the yeasts

® 29 and thus to release the PB-galactosidase activity. After thawing, the sheet of nitrocellulose is placed, colonies facing upwards, in another Petri dish containing a Whatman paper which has been presocaked in 1.5 ml of PBS solution (60 mM NaHPO,, 40 mM NaH,POy, mM KCl, 1 mM MgSO4, pH7) containing 15 pul of X-Gal (5-bromo-4-chloro-3-indoyl-B-D-galactoside) at 40 mg/ml of N,N-dimethylformamide. The dish is then placed in an incubator at 37°C. The test is taken to be positive when 10 the colonies on the membrane turn blue after 6 hours.

EXAMPLE 1 : CONSTRUCTION OF AN EXPRESSION

VECTOR FOR A FUSION PROTEIN IN WHICH FUSION IS BETWEEN

THE PTB1 DOMAIN OF FE65 AND THE LEXA PROTEIN

Screening a library using the double-hybrid system requires the PTBl domain of FE65 (FE65PTB1l) to be fused to the LexA protein. The expression of this fusion protein is carried out using the vector plex9, into which the sequence encoding the PTB1 domain of

FE65 (SEQ ID No. 1 to 2) has been introduced, in the same reading frame as the sequence corresponding to the

LexA protein.

The 448 bp-fragment of DNA which corresponds to amino acids 395 to 543 of the human FE65 protein (SEQ ID No. :2) was obtained by PCR using the oligonucleotides SEQ ID No.3 and SEQ ID No.4, which also allowed us to introduce the sites Xmal and Sall at

® 30 the ends of the sequence. The PCR fragment was introduced between the Xmal and Sall sites of the multiple cloning site of the plasmid plex9, downstream of the sequence which corresponds to LexA, so as to give the vector plex FE65PTBI.

The construct was verified by sequencing the

DNA. This verification allowed us to show that this fragment has no mutations which were generated during the PCR reaction, and that it is fused in the same open reading frame as that of the fragment which corresponds to LexA.

EXAMPLE 2 : SCREENING A BRAIN cDNA FUSION

LIBRARY BY THE TWO-HYBRID TECHNIQUE

We used the double-hybrid method (Fields and

Song, 1988). Screening a fusion library makes it possible to identify clones which produce proteins which are fused with the transactivating domain of

GAL4, and which are able to interact with the PTB1 domain of FE65. This interaction makes it possible to reconstitute a transactivator which will be capable of inducing the expression of the reporter genes His3 and

LacZ in strain L40. To carry out this screening, we chose a fusion library which was prepared from human brain cDNA (Clontech®).

During the screening, it 1s necessary to maintain the probability that each separate plasmid of the fusion library is present in at least one yeast at

® 31 the same time as the plasmid pLex-FE65PTBl. To maintain this probability, it is important to have a good efficiency of transformation of the yeast. For this, we chose a protocol for transforming yeast which gives an efficiency of 10° cells transformed per pug of DNA. In addition, as the cotransformation of the yeast with two different plasmids reduces this efficiency, we preferred to use a yeast which was pretransformed with the plasmid pLex-FE65PTBl. This strain L40-FE65PTB1, of phenotype His-, Lys-, Ade-, Leu-, was transformed with 30 pg of plasmid DNA from the fusion library. This amount of DNA allowed us to obtain, after estimation, 2.8 x 10° transformed cells, which corresponds to a number which is slightly higher than the number of separate plasmids that constitutes the library.

According to this result, we can consider that virtually all of the plasmids of the library were used to transform the yeasts. The selection of the transformed cells which are capable of reconstituting a functional GAL4 transactivator was done on a

YNB+Lys+Ade medium.

At the end of this selection, 97 clones having a His+ phenotype were obtained. A B- galactosidase activity assay was carried out on these transformants to determine the number of clones which express the other reporter gene, LacZ. Out of 97 clones obtained, 27 had the double phenotype His+, BGal+,

® 32 which thus shows that they express proteins which are able to interact with the PTB1 domain of FEG65.

EXAMPLE 3 : ISOLATION OF THE BRAIN LIBRARY

PLASMIDS FROM THE YEAST CLONES SELECTED

To identify the proteins which are able to interact with the PTBl domain of FE65, we extracted the fusion library plasmids contained in the yeast which were selected during the double-hybrid screening. In order to be able to obtain a large amount thereof, this isolation requires a prior transformation of E. coli with an extract of DNA from the positive yeast strains.

As the library plasmid which is contained in this extract 1s a yeast/E.coli shuttle plasmid, it can easily replicate in the bacterium.

The plasmid DNAs from the bacterial colonies obtained after transformation with yeast DNA extracts, were analyzed by digestion with restriction enzymes and separation of the DNA fragments on an agarose gel. Out of the 23 clones analyzed, we obtained 6 different restriction profiles, of which two were highly represented. These results showed that at least © different plasmids were isolated during the screening; we focused more particularly on the DNA fragment originating from the cDNA library which is contained in the two most represented (8 and 4 times) plasmids.

® 33

EXAMPLE 4 : DETERMINATION OF THE SEQUENCE OF

THE INSERTS OF THE PLASMIDS IDENTIFIED

Sequencing was carried out on the 2 most represented plasmids using the oligonucleotide GAL4TA (SEQ ID No.5), which is complementary to the GAL4TA region, close to the insertion site of the brain cDNA library, at 52 bp from the EcoRI site.

Comparison of the sequence of the first plasmid selected with the sequences contained in the databanks GENBank and EMBL (European Molecular Biology

Lab) showed that the sequence of the cDNA which is present in this first plasmid shows more than 99% identity, at the nucleotide level, with the human gene which encodes the hnRNPL protein, having the access number: NP 001524/g4557645. The sequence of this gene, which we cloned by the two-hybrid system, begins at nucleotide 346, which corresponds to the 116%" amino acid, and ends at nucleotide 1392, which corresponds to the 464" amino acid which is located 58 amino acids from the end of the hnRNPL protein (SEQ ID No. :6 and 7). This result shows that the domain of interaction of hnRNPL with the human FE65 protein is contained in the central region of hnRNPL. This region contains a sequence of type NPXY, which is known to be the consensus site of binding of the PTB domains (Borg et al., 1996). The hnRNPL sequence (SEQ ID No. 6 to 7) which we cloned differs from the published sequence (Pinol-Roma et al., 1989) by the substitution of a

® 34 guanine with thymidine at position 748, which leads to the changing of a glycine into cysteine at the 250%" amino acid of the sequence SEQ ID No. :6, which corresponds to nucleotide 1093 and to amino acid 365 of 5S the whole hnRNPL protein.

Comparison of the sequence of the second plasmid selected with the sequences contained in the databanks GENBank and EMBL (European Molecular Biology

Lab) showed that the sequence of the cDNA which is contained in this plasmid shows no significant homology with the sequences contained in these databanks.

Significant homology was found with the protein of sequence SEQ ID No. 4 of application

WO 99/26961. However, the protein of sequence

SEQ ID No. 9 which is the subject of the present application differs from the protein of sequence

SEQ ID No. 4 of application WO 99/26961. In particular, the protein of sequence SEQ ID No. 9 is shorter by 42 amino acids. A certain homology exists over the rest of the sequence, but also notable differences. In particular, the amino acids at positions 1, 48, 61 and 305 are different from the amino acids at the corresponding positions on the part showing a certain homology of the protein of sequence SEQ ID No. 4 of application WO 99/26961.

This sequence (SEQ ID No. 8 to 9) of 1275 nucleotides has a stop codon at position 1012, and encodes a peptide of 337 amino acids. The protein which

® 35 corresponds to this sequence was named FEBP1l for FE65

Binding PTBl domain protein.

EXAMPLE 5 : ANALYSIS OF THE SPECIFICITY OF

INTERACTION BETWEEN THE PTB DOMAINS OF FE65 AND THE hnRNPL AND FEBP1 PROTEINS

To determine the specificity of the interaction between the PTB1 and PTB2 domains of the human FE65 protein and the hnRNPL and FEBPl proteins, we carried out an interaction assay by the two-hybrid method, using the plasmid pLex-FE65PTB2 which encodes the PTB2 domain of FE65 fused to the LexA protein, in place of the plasmid pLex-FE65PTBl. An absence of interaction between the PTB2 domain and these two proteins would make it possible to show a specificity of interaction with the PTB1 domain of FE65.

To carry out this assay, we transformed strain L40 with the plasmids which were isolated during the screening of the brain cDNA library, and with the plasmid pLex-FE65PTB2. Controls for specificity of interaction were also carried out, by transforming this strain with different plasmids as indicated in table

No. 1. A PGal activity assay was carried out on the cells which were transformed with the various plasmids, 1in order to detect a protein-protein interaction. All the plasmid combinations, and thus the interactions, which were tested in the double-hybrid system are reported in Table No. 1. The plasmid combinations and

® 36 the corresponding type of vector (plLex or pGAD) are indicated in the columns “Plasmid Combinations”. The + sign and - sign in the “Interaction” column correspond to the results of the BGal assay, and indicate respectively the detection or the nondetection of protein-protein interaction.

According to the result of the assay (cf.

Table No. 1), only the two yeasts which were transformed with the two plasmids isolated from the brain cDNA library and with the plasmid pLex-FE65PTR] had BGal+ activity, which thus shows that among the two

PTB domains of FE65, only the PTB1 domain interacts with the central portion of hnRNPL or the fragment of the FEBPl protein. The PTBl domain of FE65 appears to interact specifically with hnRNPL and FEBP1l, since we were unable to show interactions with the HaRasVall2 protein or the C-terminal domain of the APP, by the two-hybrids technique.

® 37

BIBLIOGRAPHIC REFERENCES

Bartlet, P.L., C.-T. Chien, R. Strenglanz and

S. Fields, 1993 D.A. Hartley Ed, Oxford University press :153.

Borg, J.-P. Oci, E. Levy and B. Margolis, 1996. The Phosphotyrosine Interaction Domains of X11 and FE65 Bind to Distinct Sites on the YENPTY Motif of

Amyloid Precursor Protein. Mol. Cell. Biol. 16 : 6229-~ 6241.

Borg, J.-P., Y. Yang, M. De Taddeo-Borg, B.

Margolis and R.S. Turner, 1998. The Xlla protein Slows

Cellular Amyloid Precursor Protein Processing and

Reduces AP40 and AP42 Secretion. J. Biol. Chem. 273 : 14761-14766.

Bressler, 5.L., M.D. Gray, B.L. Sopher, OQ.

Hu, M.G. Hearn, D.G. Pham, M.B. Dinulos, K.-I. Fukuchi, $.5. Sisodia, M.A. Miller, C.M. Disteche and G.M.

Martin, 1996. cDNA cloning and chromosome mapping of the human Fe65 gene: interaction of the conserved cytoplasmic domains of the human B-amyloid precursor protein and its homologs with the mouse Fe65 protein.

Hum. Mol. Genet. 5 : 1589-1598.

® 38

Chung, C.T., E.L. Suzann, and R.H. Miller, 1989. One-step preparation of competent Escherichia coli : transformation and storage of bacterial cells in the same solution. Proc. Natl. Acad. Sci. USA, 5S 86 : 2172-2175.

Duilio, A., R. Faraonio, G. Minopoli, N.

Zambrano and T. Russo. 1998. Fe65L2 : a new member of the Fe65 protein family interacting with the intracellular domain of the Alzheimer’s fB-amyloid precursor protein. Biochem. J. 330 : 513-519.

Ermekova, K.S., N. Zambrano, H. Linn, G.

Minopoli, F. Gertler, T. Russo and M. Sudol. 1997. The

WW Domain of Neural Protein FE65 Interacts with

Proline-rich Motifs in Mena, the Mammalian Homolog of

Drosophila Enabled. J. Biol. Chem. 272 : 32869-32877.

Fields, S. and O. Song. 1989. A novel genetic system to detect protein-protein interactions. Nature. 340 : 245-246.

Fiore, F., N. Zambrano, G. Minopoli, V.

Donini, A. Duilio and T. Russo. 1995. The Regions of the Fe65 Protein Homologous to the Phosphotyrosine

Interaction/Phosphotyrosine Binding Domain of Shc Bin the Intracellular Domain of the Alzheimer’s Amyloid

Precursor Protein. J. Biol. Chem. 270 : 30853-30856.

® 39

Gietz, R.D., R.H. Schiestl, A.R. Willems, and

R.A Woods. 1995 Studies on the transformation of intact yeast cells by LiAC/SS-DNA/PEG procedure. Yeast, 11 : 355-360.

Golde, T.E., S. Estus, L.H. Younkin, D.J.

Selkoe and S.G. Younkin, 1992. Processing of the amyloid protein precursor to potentially amyloidogenic derivatives. Science 255 : 728-730.

Guenette, S.Y., J. Chen, P.D. Jondro and R.E.

Tanzi. 1996. Association of a novel human FE65-like protein with the cytoplasmic domain of the b-amyloid precursor protein. Proc. Natl. Acad. Sci. U.S.A. 93 : 10832-10837.

Guenette, S.Y., J. Chen, A. Ferland, C.

Haass, A. Capell and R.E. Tanzi. 1999. HFEG65L.

Influences Amyloid Precursor Protein Maturation and

Secretion. J. Neurochem. 73 : 985-993.

Koo, E.H., and S.L. Squazzo, 1994. Evidence that production and release of amyloid beta-protein involves the endocytic pathway. J. Biol. Chem. 269 : 17386-17389.

® 40

Kahn, A. 1991 Therapie genique: espoirs et limites. Medecine et Sciences. 7 : 705-714.

McLoughlin, D.M. and C.C.J. Miller. 1996. The intracellular cytoplasmic domain of the Alzheimer’s disease amyloid precursor proteins in the yeast two- hybrid system. FEBS Lett. 397 : 197-200.

Mercken, L., M. Bock, J. Menager, X. Franco,

M.-F. Paul, L. Pradier and A. Fournier. 1998. FE65 and

COFE65 : two proteins interacting with the cytoplasmic domain of APP. Neurobiol. Aging 19 : S37.

Perez, R.G., S. Soriano, J.D. Hayes, B.

Ostaszewski, W. Xia, D.J.Selkoe, X. Chen, G.B. Stokin, and E.H. Koo. 1999. Mutagenesis identifies new signals for beta-amyloid precursor protein endocytosis, turnover, and the generation of secreted fragments, including Abetad2. J. Biol. Chem. 274 : 18851-18856.

Pinol-Roma, S., M.S Swanson, J.G. Gall, and

G. Dreyfuss. 1989. A novel heterogeneous nuclear RNP protein with a unique distribution on nascent transcripts. J. Cell Biol. 109 : 2575-2587.

Sabo, S.L., L.M. Lanier, A.F. Ikin, O.

Khorkova, S. Sahasrabudhe, P. Greengard and J.D.

Buxbaum. 1999. Regulation of the PB-Amyloid Secretion by

FE65, an Amyloid Protein Precursor-binding Protein. J.

Biol. Chem. 274 : 7952-7957.

Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular Cloning, A laboratory manual. Second

Edition, Tome I, II, III. Cold Spring Harbor Laboratory

Press.

Sanger, F., S. Nicklen, and A.R. Coulson, 1997. DNA sequencing with Chain Terminating Inhibitors.

Proc. Natl. Acad. Sci. USA, 74 : 5463-5467.

Sastre, M., R.S. Turner, and E. Levy. 1998.

X11 Interaction with P-Amyloid Precursor Protein

Modulates Its Cellular Stabilization and Reduces

Amyloid P-Protein Secretion. J. Biol. Chem. 273 : 22351-22357.

Tanahashi, H. And T. Tabira. 199%a X11L2, a new member of the XII protein family, interacts with

Alzheimer’s beta-amyloid precursor protein. Biochem.

Biophys. Res. Commun. 255 : 663-667.

Tanahashi, H., and T. Tabira. 1999. Genome

Structure and Chromosomal Mapping of the Gene for

FE65L2 Interacting with Alzheimer’s f(-Amyloid Precursor

Protein. Biochem. Biophys. Res. Commun. 258 : 385-389.

® 42

Tanahashi, H., and T. Tabira. 1999c.

Molecular cloning of human Fe65L2 and its interaction with the Alzheimer’s beta-amyloid precursor protein.

Neurosc. Lett. 261 : 143-146.

Trommsdorff, M., J.-P. Borg, B. Margolis and

J. Herz. 1998. Interaction of Cytosolic Adaptor

Proteins with Neuronal Apolipoprotein E. Receptors and the Amyloid Precursor Protein. J. Biol. Chem. 273 : 33556-33560.

Vojtek, A.B., S.M. Hollenberg, and J.A.

Cooper. 1993. Mammalian Ras interacts directly with the serine/threonine kinase Raf. Cell 74 : 205-214.

Zambrano, N., J.D. Buxbaum, G. Minopoli, F.

Fiore, P. De Candia, S. De Renzis, R. Faraonio, 8S.

Sabo, J. Cheetham, M. Sudol and T. Russo. 1997.

Interaction of the Phosphotyrosine

Interaction/Phosphotyrosine Binding-related Domains of

Fe65 with Wild-type and Mutant Alzheimer’s PB-Amyloid

Precursor Proteins. J. Biol. Chem. 272: 6399-6405.

Yankner, B.A. 1996. Mechanisms of neuronal degeneration in Alzheimer’s disease. Neuron 16 : 921-

® 43

TABLE No. 1

I EE eeeserm | hee eens | ome eeersr | ewe eens | eon eens | wee samesvanie | meme demeeverie | ees wemasvanz | eee

Claims (40)

- @ CLAIMS

1. Compound capable of modulating, at least partially, the interaction of the hnRNPL and/or FEBP1 proteins, or a homolog of these proteins, with the PTB1 domain of FE65.

2. Compound according to claim 1, characterized in that it slows, inhibits or stimulates, at least partially, said interaction.

3. Compound according to either of claims 1 and 2, characterized in that it is capable of binding the domain of interaction between the hnRNPL and/or FEBP1l proteins, or a homolog of these proteins, and the PTB1 domain of FE65.

4. Compound according to one of claims 1 to 3, characterized in that it is a compound of peptide, nucleic acid, lipid or saccharide type, or an antibody.

5. Compound according to claim 4, characterized in that it is a peptide compound comprising a portion of the peptide sequence of the hnRNPL protein and/or of the FEBP1l protein and/or of derivatives thereof.

6. Compound according to claim 5, characterized in that it comprises a portion of the sequence SEQ ID No. 7 or SEQ ID No. 9.

7. Compound according to claim 4, characterized in that it is a peptide compound comprising a region whose sequence corresponds to all

® 45 or a functional portion of the site of interaction of the hnRNPL protein and/or the FEBPl protein with the PTB1 domain of FE65.

8. Compound according to claim 4, characterized in that it is a peptide compound which is derived from the hnRNPL protein or from the FEBP1 protein (and/or from the homologous forms) and which bears an effector region which has been made nonfunctional.

9. Polypeptide having the sequence SEQ ID No. :9 or a derivative or fragment of this sequence.

10. Polypeptide derived from the sequence SEQ ID No. :7 bearing an effector region which has been made nonfunctional.

11. Nucleic acid encoding a peptide compound according to one of claims 4 to 10.

12. Nucleic acid according to claim 11, characterized in that it comprises all or part of the sequences SEQ ID No. :6 or 8, or of a sequence which is derived from these sequences.

13. Nucleic acid encoding a polypeptide according to claim 9,

14. Nucleic acid capable of hybridizing with a nucleic acid according to one of claims 11 to 13, or with its complementary strand.

15. Vector comprising a nucleic acid according to one of claims 11 to 14.

A 46

16. Defective recombinant virus comprising a nucleic acid according to one of claims 11 to 14.

17. Antibody or antibody fragment or derivative, characterized in that it is directed against a peptide compound according to one of claims 4 to 10.

18. Nonpeptide compound or a compound which is not exclusively of peptide nature, which is capable of modulating, at least partially, the interaction of the hnRNPL and/or FEBP1l proteins, or a homolog of these proteins, with the PTBl domain of FE65.

19. Compound according to claim 18, characterized in that the active motifs of a peptide according to one of claims 5 to 8 have been duplicated with a structure which is not a peptide or which is not exclusively peptide in nature.

20. Pharmaceutical composition comprising at least one compound according to one of claims 1 to 10, 18 and 19 or an antibody according to claim 17.

21. Pharmaceutical composition comprising at least one nucleic acid according to one of claims 11 to 14 or one vector according to either of claims 15 and

16.

22. Pharmaceutical composition comprising a peptide compound according to one of claims 4 to 10.

23. Composition according to one of claims 20 to 22, intended for modulating, at least partially,

477 PCT/FRO1/00361 the interaction between the FE65 protein and the hnRNPL or FEBPl protein.

24. Composition according to one of claims to 22, intended for treating neurodegenerative pathologies.

25. Method for screening or characterizing active molecules, comprising a step of selecting molecules which are capable of binding the sequence SEQ ID No. 7 or the sequence SEQ ID No. 9, or a fragment of these sequences.

26. Method for producing a peptide compound according to one of claims 4 to 10, comprising the culture of a cell which contains a nucleic acid according to one of claims 11 to 14 or a vector according to either of claims 15 and 16, under conditions for expression of said nucleic acid, and the recovery of the peptide compound produced.

27. Use of a compound as claimed in any one of claims 1 to 10, 18 or 19, an antibody as claimed in claim 17, a vector as claimed in claim 15 or 16, or nucleic acid according to one of claims 11 to 14, in the manufacture of a medicament for treating neurodegenerative pathologies.

28. A substance or composition for use in a method for treating neurodegenerative pathologies, AMENDED SHEET

“ 48 PCT/FR01/00361 said substance or composition comprising a compound as claimed in any one of claims 1 to 10, 18 or 19, an antibody as claimed in claim 17, a vector as claimed in claim 15 or 16, or nucleic acid according to one of claims 11 to 14, and said method comprising administering said substance or composition.

29. A compound according to claim 1, or claim 18, substantially as herein described and illustrated.

30. A polypeptide according to claim 9, or claim 10, substantially as herein described and illustrated.

31. A nucleic acid according to claim 11, or claim 13, or claim 14, substantially as herein described and illustrated.

32. A vector according to claim 15, substantially as herein described and illustrated.

33. A virus according to claim 16, substantially as herein described and illustrated.

34. An antibody according to claim 17, substantially as herein described and illustrated.

35. A composition according to any one of claims 20 to 24, substantially as herein described and illustrated.

36. A method according to claim 25, substantially as herein described and illustrated. AMENDED SHEET

BE 49 PCT/FRO1/00361

37. A method according to claim 26, substantially as herein described and illustrated.

38. Use according to claim 27, substantially as herein described and illustrated.

39. A substance or composition for use in a method of treatment according to claim 28, substantially as herein described and illustrated.

40. A new compound; a new polypeptide; a new nucleic acid; a new vector; a new virus; a new antibody; a new composition; a new method for screening or characterizing molecules; a new method for producing a compound; a new use of a compound as claimed in any one of claims 1 to 10, 18 or 19, an antibody as claimed in claim 17, a vector as claimed in claim 15 or 16, or nucleic acid according to one of claims 11 to 14; or a substance or composition for a new use in a method of treatment, substantially as herein described. AMENDED SHEET