WO2000043779A1

WO2000043779A1 - Method for identifying the ligands of a receptor capable of being internalized

Info

Publication number: WO2000043779A1
Application number: PCT/FR2000/000113
Authority: WO
Inventors: Hubert Vaudry; Nicolas Chartrel; Alain Beaudet; Zsolt Lenkei; Catherine Llorens-Cortes
Original assignee: Institut National De La Sante Et De La Recherche Medicale (Inserm); Universite Mc Gill
Priority date: 1999-01-20
Filing date: 2000-01-19
Publication date: 2000-07-27
Also published as: FR2788602A1; FR2788602B1; EP1145003A1; CA2359213A1; JP2002535654A

Abstract

The invention concerns a method useful for detecting and/or identifying in a bank of peptides, pseudopeptides and non-peptide compounds, a biological extract or a purified fraction of a tissue extract, a ligand of a receptor of interest capable of being subjected to internalization induced by immobilising said ligand. The invention is characterised in that it comprises at least steps which consist in: expressing at the surface of a cell said receptor in a marked form; bringing in contact said cell and said bank in conditions suitable for cell internalization of said ligand-receptor complex; and displaying said internalization via the detection of the marker associated with said receptor.

Description

IDENTIFICATION OF THE LIGANDS OF A RECEIVER CAPABLE OF INTERNALIZING

The present invention relates to a screening method applicable to the identification of potential ligands for a receptor capable of internalizing.

The invention relates more particularly, but not exclusively, to receptors belonging to the family of receptors with 7 transmembrane domains coupled to G proteins as well as those of the family of tyrosine kinase receptors to a single transmembrane domain.

To date, around 800 receptors with 7 transmembrane domains coupled to G proteins (GPCR) have been cloned from different eukaryotic species. In humans, 240 GPCRs have been isolated. For only 140 of them the endogenous ligand is known, for the other 100 remaining which constitute the pool of orphan receptors, they are to be identified.

At the same time, several hundred new drugs acting on GPCRs have been registered over the past twenty years.

Consequently, the newly cloned orphan receptors are of great interest for pharmaceutical research since they are capable of representing potential therapeutic targets.

However, their valorization on a therapeutic level implies beforehand to isolate their endogenous ligand and, by the same token, to elucidate their function.

On all of these orphan receptors coupled to proteins

G, only 5 endogenous ligands have so far been isolated, all belonging to families of new peptides: orexins / hypocretins, nociceptin / orphanine, PrRP, prolactin-releasing peptide, apeline and leucokinin-type peptide ( leucokinin-like).

To identify them, the authors screened in all cases the fractions of purified tissue on different tests. For the choice of tests, they took advantage of the fact that when the agonist fixes itself on its receiver, there is formation of a second messenger which will produce, according to the signaling channel used by the receiver, different products and which will in all cases lead to a change in intracellular pH.

In the case of nociceptin, the authors measured the accumulation of cAMP, resulting from the stimulation of adenylate cyclase.

In the case of orexins, the authors measured the accumulation of intracytoplasmic Ca2 + resulting from the stimulation of phospholipase C. In the case of PrRP, the authors measured the formation of arachidonic acid resulting from the stimulation of phospholipase A2.

In the case of apeline, they measured the changes in acidification of the extracellular medium using a "cytosensor".

However, all of these identification routes are not completely satisfactory for the following reasons:

The identification process consisting in evaluating the production of second messengers requires knowing the signaling pathway of the receptor. In the presence of an orphan receptor, having no idea of the path involved, it is necessary to test all of the known signaling pathways with the hope that the receptor of interest is not coupled to a path still unknown. This therefore requires having a large amount of sample. This approach also implies that the receptor of interest is coupled to a cascade of second messengers in the heterologous transfection systems in which they are expressed, which is not necessarily the case.

As for the approach consisting in measuring the acidification of the extracellular medium, following the production of protons by the activated cell, it comes up against the following problem: if one puts in the presence of peptide banks or purified fractions of extracts of tissues with an orphan receptor expressed on the surface of a cell, we measure a change in extracellular pH which results from the stimulation not only of the orphan receptor but also of all the endogenous receptors present. The ligand responsible for activating the orphan receptor cannot therefore be easily demarcated.

The object of the present invention is precisely to propose a new method for detecting and / or identifying ligands for orphan receptors which proves to be more reliable than those mentioned above, feasible on samples of low volumes and usable in the presence of other endogenous receptors and a large number of ligands, peptide or non-peptide, for these annexed endogenous receptors. The method developed in the context of the present invention takes advantage of the properties of receptors with seven transmembrane domains, coupled to G proteins, or tyrosine kinase receptors, of internalizing in the cells which express them under the action of agonist ligands. Internalization is a fairly universal phenomenon that affects a large number of receptors. This internalization can thus be visualized by confocal, optical or even electronic microscopy when the ligand is labeled with a fluorescent molecule or an epitopic marker. The ligand-receptor complexes then follow a characteristic intracellular path, which has already been well studied. For example, this fluorescent or epitopic labeling technique has already been recommended to track intracellular traffic, either:

- a labeled ligand, complexed to its respective receptor, said receptor undergoing internalization induced by its activation,

- a protein involved in signal transduction, in this case the labeled protein kinase C,

- Or a protein involved in the desensitization of a receptor after the activation of the latter, in this case the labeled β-arestine 2.

However, the option of labeling either a ligand or a protein involved in signal transduction or in the desensitization of a receptor after activation, such as labeled β-arestine 2, is not completely reliable. There may indeed remain an ambiguity as to the identity of the receiver whose internalization has been followed. Moreover recent data suggest that different proteins may be involved in the internalization and desensitization mechanisms of various receptors. Β-arestine 2, in particular, does not seem to play a universal role. For example, when monitoring the mobilization of β-arestine coupled to EGFP, during stimulation of the orphan receptor by its ligand, there is phosphorylation of the receptor which can then fix the mobilized β-arestine2-GFP from the membrane cytoplasm. Then there is sequestration of the receptor and internalization. However, in the case of the search for an endogenous ligand for an orphan receptor overexpressed on the surface of eukaryotic cells, contact with a library of peptides or purified fractions of tissue results in the activation not only of the receptor orphan but also of all endogenous receptors coupled to G proteins. This therefore results in a mobilization of β-arestine 2-GFP not only by the orphan receptor but also by endogenous receptors. This therefore does not make it possible to identify the ligand responsible for the activation of the orphan receptor.

Likewise, in the case of a library of purified peptides or tissue fractions containing the endogenous ligand sought, if all the peptides are homogeneously labeled, it is not possible to identify the one that is sought.

The claimed screening method precisely has the advantage of removing this indeterminacy.

More specifically, the present invention relates to a method useful for detecting and / or identifying in a library of peptide, pseudopeptide or non-peptide compounds, a biological extract and / or a purified fraction of a tissue extract, a ligand of a receptor of interest and capable of undergoing internalization induced by the binding of said ligand, characterized in that it comprises at least the steps consisting in: - expressing said receptor in a form marked on the surface of a cell,

- bringing said cell into contact with said bank, the biological extract and / or a purified fraction of a tissue extract and containing less a peptide, pseudopeptide or non-peptide compound capable of being a ligand of said receptor, under conditions sufficient to allow cellular internalization of said receptor-ligand complex and

- visualize this internalization via the detection of the marker linked to said receptor.

The present invention therefore involves labeling the orphan receptor for which, in particular, potential agonists are sought.

The fact of marking in itself the orphan receptor is clearly advantageous with regard to the detection techniques mentioned above. Indeed, this approach offers the possibility of directly visualizing the target studied. When the receptor is brought into contact with the endogenous ligand, it is the ligand-receptor-marker complex which is internalized. There is no ambiguity about the identity of the internalized receptor. In addition, in the claimed method, the cell constitutively overexpresses the labeled receptor. The labeling is intracellular, on the cytoplasmic tail of the receptor and therefore has the advantage of not hindering the binding of the ligand.

As regards the markers suitable for the invention, it may be either an autofluorescent protein or an epitopic marker capable of being detected by immunohistochemistry.

The fluorescent proteins which can be used in the claimed process preferably belong to the family of wild fluorescent protein GFP and its mutants (Ex: EGFP, EBFP and EYFP). Wang S. & Haizeirigg T. (1994) Nature, 369: 400-403; Yang TT et al. (1996) Nucleic Acid Res, 24: 4592-4593; Heim R & Tsien RY (1994) Curr. Biol., 6: 1, 178-182; Ormô M et al., (1996) Science, 273: 1392-1395.

By way of illustration of the non-fluorescent markers capable of also being used according to the invention, mention may be made most particularly of hemaglutinin, polyhistidine, myc and flag proteins and viral epitopes. For all these highly immunogenic compounds, there are already commercially highly selective antibodies (e.g. monoclonal antimyc antibodies, Clontech; antihemaglutinin influenza antibodies, Boehringer; anti-vesicular somatostatitis virus antibody, Clontech; anti-polyhistidine antibody, In Vitrogen). Their detection involves recognition of the antigenic group by one of these antibodies (called primary), followed by visualization of the primary antibody by a secondary antibody from another species and labeled either with a fluorophore or with the horseradish peroxidase which will be reacted consecutively with an appropriate substrate.

This type of labeling is particularly advantageous insofar as it allows a high measurement sensitivity which results in the detection of a ligand concentration of the order of 10 ^"8 M, that is to say 100 fmoles in 10 The internalization of the complex is visible by confocal or even optical microscopy when the receptor is strongly expressed and if the ligand is in sufficient concentration (minimum 10 ^"8 M).

This internalization is preferably detected by confocal and / or optical microscopy.

Finally, advantageously, the other endogenous, unlabeled receptors present on the surface of the host cells, even if they are also internalized, are not visible and do not interfere with the measurement. No background noise is therefore noted, which is particularly interesting in view of the high sensitivity of the measurement method.

In general, the claimed method turns out to have sufficient sensitivity to allow visualization of an internalization of a receptor-ligand conjugate within the fraction tested notwithstanding the presence of other endogenous receptors and a large number of ligands, peptide or non-peptide for these additional endogenous receptors.

Advantageously, the claimed screening method proves to be suitable for the study of any receptor provided that it has the capacity to internalize. According to a preferred mode of the invention, the receptor of interest is coupled to a protein G.

The claimed method is particularly interesting for characterizing receptor ligands belonging to the receptor family with 7 transmembrane domains coupled to the G protein, the GPCRs, to the family of tyrosine kinase receptors with a single transmembrane domain or to the family of cytokine receptors.

In the case of GPCRs, labeling receptors at C-terminal has the following advantages:

- the post-translational maturation of the receptor is not modified,

- the binding of agonists and antagonists as well as intracellular signaling are the same as with the native receptor,

- the internalization of the ligand-receptor complexes is not modified and

- the expression detection method is rapid and can be carried out on non-fixed living cells.

The receptor to be studied is coupled to a marker and then expressed on the surface of a host cell. With regard to these two operations, namely the labeling and expression of said receptor at the level of a host cell, they of course both fall within the competence of those skilled in the art. Generally, the procedure is as follows: The sequence of the coding part of the receptor of interest is inserted in phase, in an expression vector, upstream or downstream of the coding sequence of the fluorescent protein (GFP, EGFP, EBFP, EYFP) or an epitopic marker. These expression vectors (of the pGFP-N1 or pGFP-C1, N1 or C1 type indicate the position of the receptor relative to the protein either at the N-terminal or at the C-terminal) already contain the sequence of these markers. The cells are transfected by a conventional method (such as the liposome method, calcium phosphate), then selected for their resistance to an antibiotic. In the case of proteins of the GFP family, it is possible to sort the cells expressing the receptor coupled to the fluorescent protein by flow cytometry. As regards the transformed cells, these are eukaryotic cells such as, for example, epithelial cells from monkey kidneys: COS-7; hamster ovaries: CHO; and human embryonic kidney cells: HEK 293). In a variant of the claimed method, it is possible to envisage expressing on the surface of a cell, two or more distinct receptors marked respectively by different markers. The two receivers are of course capable of internalizing. This option thus offers the possibility of screening potential ligands for several receptors simultaneously on the same sample.

In accordance with the claimed method, the host cells are brought into contact with a potential ligand. To do this, the cells are therefore brought into contact with either a bank of peptide, pseudopeptide or non-peptide compounds, a biological extract or even purified fractions of a tissue extract.

This contacting is carried out under conditions sufficient to allow cellular internalization of the receptor-marker-ligand complex (s). These sufficient conditions are of course appreciated by those skilled in the art, in terms of temperature, duration and concentration. It may also be necessary to carry out repeated experiments.

Generally, internalization of ligand-receptor complexes in mammalian cells is optimal at 37 ° C. The average half-life of the internalization process (time required for 50% of occupied surface receptors to be internalized) is around 10 minutes. Thus, exposures to the ligand of 20 to 40 minutes are usually optimal for the identification of internalized receptors, the optimal ligand concentrations are of the order of 10 to 20 times the affinity of the ligand for its receptor (Kd). We follow by confocal microscopy or in the case of a strong expression of this labeled receptor, by optical microscopy, the internalization of the ligand-receptor complex which takes the form of the displacement of the labeling (fluorescent or immunohistochemical) of the cell membrane (whose marking intensity decreases) towards the interior thereof in the form of vesicles.

In the case of a receptor marked with a fluorescent label, the visualization is carried out directly by observing the displacement of the fluorescence. In the case where the receptor is labeled with a non-fluorescent antigen, the internalization is visualized by confocal microscopy after fixation of the cells and detection of the antigenic epitopes by secondary antibodies coupled to a fluorophore. Alternatively, the antigenic epitopes can be revealed by an enzymatic reaction, or with the help of a radioactive antibody which is concretized in both cases by the accumulation of opaque deposits visible both by light and electron microscopy.

As mentioned above, internalization can be viewed on ten cells and for a very low incubation volume, namely of the order of μl. These two qualities are particularly valuable when only a very small amount of sample is available.

Consequently, the claimed method proves to be particularly advantageous for detecting and / or identifying the endogenous ligand (s) of an orphan receptor or else identifying new agonists or antagonists for a known receptor, that is to say which knows the endogenous ligand. Being based on the direct observation of a biological phenomenon which affects most of the GPCR or tyrosine kinase receptors, namely internalization, this method proves to be particularly useful for the search for ligands for orphan receptors and more particularly receptors for peptides, using biological preparations such as tissue extracts.

It is also possible to envisage using the claimed process with a labeled receptor, already identified and characterized, as a "biosensor", which would make it possible to detect in a biological fluid the presence even in the state of traces of a toxic or non-toxic substance. able to bind to this receiver.

The present invention also extends to any ligand of a receptor of interest, identified using the claimed method. The examples and figures submitted below are presented by way of non-limiting illustration of the present invention.

FIGURES Figures 1: Figure 1a: Visualization by confocal microscopy of the neurotensin type 1 receptor coupled to the fluorescent protein EGFP (NTR1-

EGFP) on the surface of CHO cells, incubated with buffer alone or with a low concentration (1 nM) of rat or frog neurotensin which does not induce internalization. Figure 1b: Visualization of the internalization of the NTR1-EGFP receptor characterized by the formation of numerous fluorescent cytoplasmic vesicles of 0.6 μm in diameter inside CHO cells incubated with neurotensin concentrations of 10 and 100 nM. Figures 2: Visualization of the internalization of the NTR1- receptor

EGFP with or without acid washing, in the presence of: 10 nM of neurotensin (FIG. 2a) or of a purified fraction of extract of frog brains (FIG. 2b). The conditions are the same as in FIG. 1, and also include an acid wash of the endogenous ligand still bound to the surface of the cells at the end of the charge period.

Figure 3: Internalization of the NTR1-EGFP receptor in the presence of 10 prepurified fractions of an extract of frog brains.

Figure 4: Radioimmunoassay of the prepurified fractions of an extract of frog brains using an antibody directed against the conserved region of neurotensin. Materials and methods

1) Construction of the NTR1 -EGFP receptor The entire coding sequence of the NT1 receptor (K. Tanaka, M.

Masu and S. Nakanishi (1990) Structure and functional expression of the cloned rat neurotensin receptor. Neuron 4: 847-54) was amplified by PCR using two primers directed against the 5 'and 3' ends of the coding sequence of this cDNA and also containing Hindlll or BamH1 restriction sites 5'-CTT AAG CTT ATG CAC CTC AAC AGC TCC GTG-3 '(SEQ ID N ° 1) and 5'-TTT GGA TCC GCG TAC AGG GTC TCC CGG GT-3' (SEQ ID N ° 2). After digestion with the enzymes Hindlll and BamH1 and purification, the amplified sequence was inserted into the expression vector pEGFP-N1 (Clontech) at the Hindlll and BamH1 sites. The construction was verified on an AbiPrism 377 automatic sequencer (Perkin-Elmer) using fluorescent ddNTPs.

2) Stable transfection in CHO cells

The CHO-K1 cells (American Type Culture Collection: ATCC) were cultured in a humid atmosphere at 5% CO2, in F12 medium supplemented with 7.5% fetal calf serum, 1 mM glutamine, 100 units / ml of penicillin and 100 units / ml streptomycin (Boehringer

Mannheim). To establish the stable line expressing this receptor, the cells

CHO-K1 (approximately 2.6x106 cells) were transfected with 8 μg of plasmid using cationic liposomes (Dosper, Boehringer). The transfected cells are selected for their resistance to geneticin. The clones obtained were sorted using flow cytometry which takes into account the intensity of the fluorescence of each cell. A second selection was made using a fluorescence microscope, observing the level of expression of the fluorescent NTR1-EGFP receptor, located on the membrane surface of the cells of each of the clones. 3) Pre-purification of frog brain extract

- Collection of fabrics

2541 brains of male green frog (species Rana ridibunda), corresponding to a weight of fresh tissue of 215 g, were collected in the laboratory from freshly sacrificed animals. The brains were frozen on dry ice immediately after being removed and stored at -80 ° C.

- Tissue extraction

The brains were immersed for 15 min in 0.5 M boiling acetic acid (2 liters), then homogenized in a mixer. The homogenate was centrifuged at 4000 x g for 30 min at 4 ° C. The supernatant was then prefiltered on an assembly of 10 columns of Sep-Pak C18 (Waters Associates, Milford, MA) mounted in series at a flow rate of 2 ml / min. The material fixed on the Sep-Pak columns was eluted with 20 ml of a 70% acetonitrile solution. This operation was repeated 2 times.

- Purification of frog brain extract by semi-preparative HPLC

The material eluted from the Sep-Pak columns was partially evaporated in order to remove the acetonitrile, then centrifuged at 13,000 xg for 5 min. The supernatant was removed and divided into 2. Each of the 2 pools was then injected into a Vydac C18 218TP1010 semi-preparative column (1 x 25 cm) (The Séparations Group, Hesperia, CA) balanced with a solution of water / trifluoroacetic acid (99.9: 0.1; vo vol) at a flow rate of 2 ml / min. Peptide material attached to the matrix was eluted using an acetonitrile gradient rising from 14 to 42% in 40 min at a flow rate of 2 ml / min, then rising from 42 to 56% for 60 min at a flow rate 1 ml / min. The eluate was collected in 1 min fraction and the absorbance measured at 215 and 280 nm. The elution fractions were stored at -20 ° C. Before the experiments internalisation, each of the fractions was diluted with water, partially evaporated in order to remove the acetonitrile, and made up to a final volume of 50 μl.

4) Internalization The cells are distributed (50% confluence, 20,000 cells per well) then cultured overnight on multi-well slides (LabTek, Nunc) pretreated with polyallylamine (0.1 mg / ml, Aldrich) . On these slides, the volume of incubations (except for the loading period) and washes is 250 μl per well. 90 min before the start of the experiment, the cell medium is changed to a medium supplemented with cycloheximide (70 μM, Sigma). The cells are then preincubated for 15 min on ice in cold Earle's buffer (pH 7.4, containing 140 mM NaCI, 5 mM KG, 1.8 mM CaCI ₂ , 3.6 mM MgCI ₂ , 0.1% albumin bovine serum, 0.01% glucose and 0.8 mM 1.10-phenanthroline). The cells are then incubated for 30 min with the ligand diluted in 50 μl of Earle's buffer at 4 ° C (loading period). At this stage, a batch of cells is subjected to a hypertonic acid wash (0.2 M acetic acid and 0.5 mM NaCl in Earle's buffer, pH 4 for 2 min at 4 ° C.) in order to dissociate the ligand of its surface receptors. Internalization is caused by replacing the medium with Earle's buffer at 37 ° C and incubation of the slides at 37 ° C for 20 to 30 min (hunting period). At the end of the incubation, the cells are rinsed with cold Earle's buffer, fixed with 4% paraformaldehyde dissolved in 0.1 M phosphate buffer at pH 7.4, rinsed again in cold Earle's buffer and then assembled. using Vectashield (Vector).

5) Confocal microscopy

The cells are examined with a Leica TCS NT confocal microscope configured with an inverted microscope (Leica DM IRBE) equipped with an argon / krypton laser with excitation and emission filters of 488 and 530-600 nm respectively. 1024-1024 pixel images of individual cells are obtained using a 63x oil immersion objective. 6) Radioimmunoassay of neurotensin on prepurified fractions of an extract of frog brains

A volume of 5 μl of each fraction collected at the outlet of semi-preparative HPLC is subjected to a radioimmunoassay of neurotensin. The neurotensin assay was carried out using an antibody directed against the C-terminal fragment of the pig neurotensin

(Marcos et al., Peptides 1996, 17: 139-146).

The antibody is used at a final dilution of 1: 50,000, and the sensitivity of the assay is 10 pg. The radioimmunoassay is carried out at 4 ° C in a 0.02 M veronal buffer (pH 8.6) containing 4% bovine serum albumin and 7000 cpm of (3- [125i] iodotyrosyl) neurotensin (Amersham, Buckingamshire, UK ). The samples and the antibody were incubated at 4 ° C for 48 h. The separation of the tracer fraction linked to the antibody was carried out by precipitation by adding to each sample a solution of γ-globulins (1% in 0.02 M veronal buffer) and a solution of polyethylene glycol (20% in 0.02 M veronal buffer containing 0.1% bovine serum albumin and 0.1% newt X-100). After an incubation of 20 min at room temperature, the samples are centrifuged (3000 x g, 30 min) and the pellets counted in a gamma counter.

EXAMPLE 1 Internalization of the NTR1-EGFP receptor in the presence of frog or rat neurotensin

1) Characterization of the NTR1 -EGFP receptor The NTR1-EGFP receptor is stably expressed in the CHO cell line and its binding and intracellular signal transmission properties have been determined. The affinity of neurotensin was found to be of the same order (0.3 nM) for the NTR1-EGFP receptor and the wild-type NT1 receptor. In the same way, the labeled receptor leads to the production of phosphate inositols with an EC50 (1 nM) identical to that obtained for the wild-type NT1 receptor (results not shown). 2) Internalization of the NTR1-EGFP receptor in the presence of frog or rat neurotensin

Cells incubated with buffer alone or with low concentrations (concentrations below 1 nM) of rat or frog neurotensin show marked fluorescence of the NTR1-EGFP receptor localized to the cell membrane (FIG. 1a).

Incubation of CHO-NTR1-EGFP cells with increasing concentrations of neurotensin (range starting at 10 nM) results in internalization of the NTR1-EGFP receptors, indicated by a decrease in fluorescent membrane labeling and by the formation of numerous intracytoplasmic fluorescent vesicles of 0.6 μm in diameter (Fig. 1 b). This type of labeling is not detected when the ligand is previously dissociated from the surface receptors following an acid wash carried out at the end of the charge period (Fig. 2a). These results indicate that the internalization observed results from the binding of neurotensin to membrane receptors during the charge period.

EXAMPLE 2 Internalization of the NTR1-EGFP receptor in the presence of prepurified fractions of an extract of frog brains

In a first series of experiments, 0.5 μl of each of the 120 elution fractions of a frog brain extract (prepurified on a semi-preparative column) were pooled by 10, giving rise to 12 pools of 10 fractions, each made up to 50 μl with Ear's buffer. Only pool 2 containing fractions 11 to 20 leads to the internalization of the NTR1-EGFP receptor.

In a second series of experiments, the fractions in pool 2, that is to say fractions 11 to 20 (0.5 μl of volume of original fraction diluted to 50 μl with Ear's buffer) were tested individually. Only fractions 15,16,17,18 induce the internalization of the NTR1-EGFP receptor (Fig. 3). This internalization is not detected if the cells are subjected to an acid wash at the end of the charge period (Fig. 2b) indicating that the internalization observed is the result of the binding of a specific ligand to the NTR1-EGFP receptors during the loading period.

EXAMPLE 3 Radioimmunoassay of Neurotensin on the Pre-Purified Fractions of an Extract of Frog Brains

The radioimmunoassay of the prepurified fractions of a frog brain extract using an antibody directed against the conserved region of neurotensin shows that the immunoreactive material is exclusively contained in the fractions 15,16,17,18 (Fig. 4). The total apparent amount of neurotensin measured in the elution fractions of the semi-preparative HPLC is 894 ng in 16 ml, which corresponds to a value of 352 pg of peptide per frog brain. Consequently, the material causing the internalization of the NTR1-EGFP receptor, contained in the fractions 15,16,17,18 corresponds to the endogenous neurotensin of the frog brain.

In conclusion, the internalization process described above is a direct, simple, specific and reliable means, making it possible to detect, from the observation of a single cell, an amount of neurotensin as low as 500 fmoles in 50 μl. It appears that this measurement can be carried out both on a pure neurotensin solution and on a fraction of tissue extract containing not only neurotensin but also a large number of other neuropeptides (the minimum estimated number per fraction being 50 peptides) , with a similar sensitivity since we manage to detect, according to the RIA assay, about 250 fmoles.

Claims

1. Method useful for detecting and / or identifying in a library of peptide, pseudopeptide or non-peptide compounds, a biological extract or a purified fraction of a tissue extract, a ligand of a receptor of interest and capable of undergoing internalization induced by the fixation of said ligand characterized in that it comprises at least the steps consisting in:

- expressing on the surface of a cell said receptor in a marked form, - bringing said cell into contact with said bank, the biological extract and / or a purified fraction of a tissue extract and containing at least one peptide compound, pseudopeptide or non-peptide capable of being a ligand of said receptor, under conditions sufficient to allow cellular internalization of said receptor-ligand complex and - visualizing this internalization via the detection of the marker associated with said receptor.

2. Method according to claim 1, characterized in that it allows the detection of a ligand concentration of the order of 10 ^"® M.

3. Method according to claim 1 or 2, characterized in that the expressed receptor is an orphan receptor.

4. Method according to claim 1 or 2, characterized in that the expressed receptor is a receptor of which the endogenous ligand is known.

5. Method according to one of the preceding claims, characterized in that the receptor is labeled with an autofluorescent protein.

6. Method according to one of the preceding claims, characterized in that the fluorescent protein is a protein of the family of the wild fluorescent protein GFP or one of its mutants.

7. Method according to one of claims 1 to 5, characterized in that the fluorescent protein is chosen from the proteins EGFP, EBFP and

EYFP.

8. Method according to one of claims 1 to 4, characterized in that the receptor is labeled using an epitopic marker capable of being detected by immunohistochemistry.

9 Method according to claim 8 characterized in that the epitopic marker is chosen from F hemaglutinin, polyhistidine, myc and flag proteins and viral epitopes.

10. Method according to one of claims 1 to 7 characterized in that the internalization is detected by optical and / or confocal microscopy.

11. Method according to one of the preceding claims, characterized in that the labeled receptor belongs to the family of receptors with 7 transmembrane domains coupled to G proteins, GPCRs, to the family of tyrosine kinase receptors with a single transmembrane domain or to the family of cytokine receptors.

12. Method according to one of the preceding claims, characterized in that one expresses on the surface of a cell, two or more distinct receptors marked respectively by different markers.

13. Use of a method according to one of the preceding claims for detecting and / or identifying the endogenous ligand (s) of an orphan receptor.

14. Use of a method according to one of claims 1 to 13 to identify new agonists or antagonists for a receptor of which the endogenous ligand is known.

15. Ligand of a receptor of interest, identified by the method according to one of claims 1 to 12.