TGFß (Transforming Growth Factor- ?) is a cytokine which plays an important role in regulating cell growth and differentiation. Smad proteins translate the signal from TGFß receptors by regulating the transcription of target genes, either directly or in combination with other specific sequence transcription factors.

The intracellular signalling cascade for members of the TGFß superfamily is initiated by binding of the cytokine to a complex of membrane receptors made up of two types of receptor with serine/threonine kinase activity, type I receptors and type II receptors. Binding of the cytokine to the type II receptor induces recruitment and phosphorylation of the type I receptor. Translation of the signal from the membrane to the nucleus then takes place by virtue of a family of proteins, the Smads. The Smad family can be divided up into three subfamilies according to

structural and functional criteria. R-Smads (Receptor- regulated Smads), which are specifically activated by the type I receptors, comprise Smadl, Smad5 and Smad8 which are specific for the BMP pathway, and Smad2 and Smad3 which are specific for the TGFß/activin pathway.

Phosphorylation of the R-Smads by the type I receptors takes place on serine residues located at the C-terminal end of the protein (SSXS motif). Activation of these R-Smads allows the formation of a heterocomplex with another member of the Smad family, common. to the various R-Smads, Smad4. This protein, also referred to as co-Smad (common-partner Smad), lacks the SSXS sequence and allows the translocation into the nucleus of the R-Smad/Smad4 complex, which can then act as a transcription factor. The third subfamily consists of Smad6 and Smad7, referred to as I-Smads (Inhibitory-Smads), which prevent the phosphorylation of R-Smads and thus their association with Smad4 and/or the translocation of the R-Smad/Smad4 complex into the nucleus by interacting directly with Smad4.

TUFA is also involved in other signalling pathways, such as that of MAP/SAP kinase or else the Ras pathway.

At the current time a series of TGFß-related pathological conditions exist for which therapeutic solutions are necessary.

Specifically, TUFA is a key cytokine in the

control of connective tissue homeostasis and plays a major role in the profibrotic activity by increasing the expression of connective tissue-specific genes (Mauviel A and Uitto J. (1993) Wounds 5: 137-52). In addition, TGF ? plays an important role in immunomodulation processes, by acting on lymphocyte differentiation (Kehrl et al, 1986, J Immunol, 137, 3855-3860, Gorelik and Flavell, 2002, Nature, 2, 46-53).

The Applicant has noted, surprisingly, that the use of an PAR receptor antagonist results in potentiation of the action of TGFß. Thus, the present invention proposes a solution for treating pathological conditions associated with a deficiency in the TGFß signal or requiring an amplification of this signal.

This use is in no way evident in view of the prior art since said prior art does not enable those skilled in the art to identify any advantage in using RAR receptor antagonists in the treatment of pathological conditions associated with a deficiency in the TGFß signal or requiring an amplification of this signal.

In fact, the prior art describes that retinoids and TGFß follow complex signalling pathways, which are all the more complex since they differ according to the subtype of receptor activated ; by way of example, for retinoids, the signalling pathway

depends on the RAR and RXR receptors and on the subtype involved (a, a or y). Finally, it should be noted that these signalling pathways are independent. Thus, the possibility of a retinoid having any effect on the TGFP signalling pathway was not envisaged.

Two classes of nuclear receptors, RAR receptors and RXR receptors, are involved in the transcriptional response to retinoic acid (RA).

The classes of retinoic acid receptors (RARs) and of retinoid X receptors (RXRs) are respectively subdivided into a, ß and y subtypes, with more than one isoform for each subtype (Chambon, 1996). RARs, associated with RXRs to form RAR-RXR heterodimers, bind the target DNA sequence, namely the retinoic acid response elements (RARs), consisting of a direct repetition of the consensus sequence 5'-PuG (G/A) (T/A) CA-3', separated by two or five nucleotides. The RAR family recognizes two natural stereoisomers of RA, all-trans-RA and 9-cis-RA, whereas the RXR family recognizes exclusively 9-cis-RA.

Retinoids are a family of compounds exhibiting a wide spectrum of therapeutic activities ranging from an anticancer activity to a skin-repairing activity (Altucci L and Gronemeyer H 2001 Nature Review 181-193, Sun SY and Lotan R (2002) Oncology Hematology 41: 41-55).

Thus, the present invention relates to the

use of at least one RAR receptor antagonist, for preparing a pharmaceutical composition intended for the treatment of pathological conditions associated with a deficiency in the TGFß signal or requiring an amplification of this signal.

The expression"RAR receptor antagonist"is intended to mean a synthetic or natural structural analogue of retinoic acid, capable of specifically binding and activating RAR receptors.

Preferably, the RAR receptor antagonists according to the invention are compounds which are RARy receptor antagonists. The conventional techniques of the state of the art enable those skilled in the art to identify such RARy receptor antagonists without particular difficulty, in particular using the mouse embryonic teratocarcinoma cell (F9) differentiation assay (Cancer Research 43, p. 5268,1983) or the assay for ornithine decarboxylase inhibition after induction with TPA in mice (Cancer Research 38, p. 793-801, 1978).

Synthetic antagonists will be preferred, preferably 4- (5, 5-dimethyl-8-p-tolyl- 5,6-dihydronaphthalen-2-ylethynyl) benzoic acid.

The term"potentiating agent"is intended to mean a compound which potentiates the transcription induced by Smad factors via TGF ? signalling.

Thus, the composition used according to the

present invention will be intended for the treatment of pathological conditions associated with a disorder of the transcription induced by Smad factors via TGFß signalling.

The expression"pathological conditions associated with a deficiency in the TGF ? signal or requiring an amplification of this signal"is intended to mean: - wound-healing disorders, such as ulcers, in particular venous ulcers, diabetic ulcers and pressure sores; - the improvement of normal wound healing, such as post-operative wound healing, in particular in the acceleration of wound healing (Bernstein E, Harisiadis L, Salomon G et al., Transforming Growth Factor-b improves healing of radiation-impaired wounds.

J Invest Dermatol 1991 ; 97: 430-434; Mauviel A and Uitto J. The extracellular Matrix in Wound Healing: Role of the cytokine Network. Wounds 1993; 5 (3): 137-152) ; some cancers, in particular those associated with a loss of control of TGFß on cell proliferation (Engle SJ et al. , 1999 Transforming growth factor betal suppresses nonmetastatic colon cancer at an early stage of tumorigenesis-Cancer Res. 59,3379-3386 ; Tang B et al. , 1998); - transplant rejection, in particular by promoting

immune tolerance (Eikmans M et al., 2002 High transforming growth factor-beta and extracellular matrix mRNA response in renal allografts during early acute rejection is associated with absence of chronic rejection-Transplantation. 73,4, 573-579); - pathological conditions involving collagen type VII (COL7A1), in particular genetic pathological conditions in which a single allele is affected, such as epidermolysis bullosa or. certain cases of wound healing which can be improved by stimulating the synthesis of collagen type VII, promoting epithelialization (Christiano AM, Greenspan DS, Hoffman GG, Zhang X, Tamai Y, Lin AN, Dietz HC, Hovnanian A, Uitto J. A missense mutation in type VII collagen in two affected siblings with recessive dystrophic epidermolysis bullosa. Nat Genet. 1993 May; 4 (1) : 62-6.

In particular, the use according to the present invention is suitable for the treatment of pathological conditions associated with collagen type VII (COL7A1).

Finally, the present invention relates to pharmaceutical compositions comprising at least one RAR receptor antagonist, intended for treatments for pathological conditions associated with a deficiency in the TGF signal or requiring an amplification of this signal.

EXAMPLES Example 1: Measurement of the potentiation of the action of RARs on the transactivation of an Smad3-dependent promoter by RAR antagonists A first approach for examining the potentiation of RAR signalling and of that of TGFß/Smad is to determine the effect of the various RAR agonists and antagonists on TGFß-induced, Smad-dependènt transactivation of the luciferase reporter gene (lux) in WI-26 lung fibroblasts transfected with the construct (CAGA) 9-lux. The latter is made up of nine copies of the CAGA nucleotide motif, known to bind Smad3 and Smad4 with high affinity, cloned upstream of the adenoviral major late promoter (MLP) controlling expression of the lux reporter gene.

A culture of WI-26 fibroblasts, at confluency, was co-transfected with (CAGA) 9-lux and an expression vector encoding RARy, in the absence (-) or in the presence (+) of the expression vector expressing Smad3. Three hours after transfection, four compounds are added: - the RAR antagonist: 4- (5, 5-dimethyl-8-p-tolyl- 5,6-dihydronaphthalen-2-ylethynyl) benzoic acid at 10-6 M (1 pl of a 10-3 M solution in DMSO introduced into 1 ml of 1% medium); - the RAR agonist: 4- [ (5, 5,8, 8-tetramethyl- 5,6, 7,8-tetrahydronaphthalen-2-ylamino) methyl] benzoic

acid at 10-7 M (1 1 of a 10-4 M solution introduced in DMSO into 1 ml of 1% medium); - ATRA (all-trans-retinoic acid) at 10-7 M (1 pl of a 10-4 M solution introduced in DMSO into 1 ml of 1% medium) ; - control (Ctrl): 1 1 of DMSO.

The luciferase activity is determined 24 hours later.

The role of the synthetic RAR agonists and antagonists on the functional interactions between Smad3 and RARy was examined, and the results are given in Figure 1.

4- (5, 5-Dimethyl-8-p-tolyl-5,6-dihydro- naphthalen-2-ylethynyl) benzoic acid (RAR antagonist) potentiates the effect of RARY on Smad3-dependent transcription, whereas 4- [ (5, 5,8, 8-tetramethyl- 5,6, 7,8-tetrahydronaphthalen-2-ylamino) methyl] benzoic acid (RAR agonist) abolishes the effect of RAR.

Example 2: Demonstration of the dose-dependent potentiating effect of the RAR antagonists on the TGFß-induced, Smad-specific transcriptional response A culture of WI-26 fibroblasts, at confluency, is transfected with the (CAGA) 9-lux gene.

After a glycerol shock, DMEM containing 1% of FCS (foetal calf serum) is added. Three hours later, increasing concentrations of the synthetic RAR antagonist, 4- (5, 5-dimethyl-8-p-tolyl- 5,6-dihydronaphthalen-2-ylethynyl) benzoic acid, are added.

These results (represented in Figure 2) show that the antagonist, 4- (5, 5-dimethyl-8-p-tolyl- 5,6-dihydronaphthalen-2-ylethynyl) benzoic acid, potentiates in a dose-dependent manner the TGFß-induced, Smad-dependent transcriptional response, with a maximum effect (+190%) at the concentration of 10-6 M. At a concentration of 10-8 M, the 4- (5, 5-dimethyl-8-p-tolyl-5,6-dihydronaphthalen- 2-ylethynyl) benzoic acid doubles the extent of the TGF-dependent response.

Example 3: Molecular demonstration of the RAR antagonists/Smad interaction via immunoprecipitation (IP) and immunoblotting experiments Cos-7 cells are transfected with the expression vectors for Smad3 labelled with Flag and Gal4BD-RARy-D-E-F, in the presence or absence (see Figure 3, lanes 1-3) of either 4- [ (5,5, 8,8-tetramethyl- 5,6, 7,8-tetrahydronaphthalen-2-ylamino) methyl] benzoic acid (lanes 4-6), or 4- (5, 5-dimethyl-8-p-tolyl- 5,6-dihydronaphthalen-2-ylethynyl) benzoic acid (lanes 7-9).

The empty Gal 4 expression vector (Gal4-e) is used as a control.

Three hours after transfection, 10-6 M of 4- (5, 5-dimethyl-8-p-tolyl-5,6-dihydronaphthalen- 2-ylethynyl) benzoic acid, are added. Forty hours later, the cell lysate is immunoprecipitated with anti-Gal4

antibodies and the co-precipitation of Smad3 is detected by immunoblotting with anti-FlaG antibodies (labelling of Smad) (top panel). The expression of RARy and Smad3 is confirmed by Western blotting using anti-Flag (middle panel) and anti-Gal4 (bottom panel) antibodies.

The migration bands in lanes 2,5 and 8, obtained with the anti-Gal4 antibody correspond to the Gal4BD protein not fused to RARy, expressed by the empty Gal4BD vector.

After immunoprecipitation with an anti-Gal4BD- antibody (top panel), Western blotting with anti-Flag antibodies reveals the co-immunoprecipitation of Smad3 with RARy in lane 3.

The amount of RARy binding Smad3 is significantly increased by the presence of 4- (5, 5-dimethyl-8-p-tolyl-5,6-dihydronaphthalen- 2-ylethynyl) benzoic acid, lane 9. This indicates that the potentiation. by the RAR antagonist, 4- (5, 5-dimethyl-8-p-tolyl-5,6-dihydronaphthalen- 2-ylethynyl) benzoic acid, of the action of TGFß, i. e. the increase in the transcriptional response induced by TGFß, involves an increase in the amount of Smad3-RARy heterocomplex formed.

Example 4: Effects of the RAR antagonists on the TGF-induced activity of the promoter for collagen type VII (COL7A1) The effect of the RAR antagonists on COL7A1 promoters (collagen type VII promoter) is quantified via the transactivation of the luciferase reporter gene (lux).

In the presence of the RAR antagonist, 4- (5, 5-dimethyl-8-p-tolyl-5,6-dihydronaphthalen- 2-ylethynyl) benzoic acid, and of TGFß, a dose-dependent increase in the TGFß-induced transcription of the reporter gene under the control of the promoter for collagen type VII (COL7A1) is observed (Figure 4).

Example 5: Effect of the RAR antagonists and agonists on the modulation of the response of endogenous COL7A1 to TGFP Furthermore, Figure 5 represents a Northern blotting gel on which it is observed, without any transfection, that the response of endogenous COL7A1 to TGFß is potentiated by the RAR antagonist. This assay also makes it possible to confirm all the transfection experiments described above.

This assay was carried out according to the following protocol: confluent HaCaT cultures were treated with the agonist, 4- [ (5, 5,8, 8-tetramethyl- 5,6, 7, 8-tetrahydronaphthalen-2-ylamino) methyl] benzoic acid, at 10-7 M, and the antagonist, 4- (5, 5-dimethyl-

8-p-tolyl-5,6-dihydronaphthalen-2-ylethynyl) benzoic acid, at 10-6 M, as described in Example 1, in the presence or absence of TGF ? (at 10 ng/ml) for 48 hours.

After extraction of the total RNA, the collagen type VII mRNA level was quantified by Northern blotting using a 32P-labelled specific cDNA probe. A probe characteristic of the mRNA of the GADPDH protein was used as a control for loading.

The genes which are targets for the action of the antagonist, 4- (5, 5-dimethyl-8-p-tolyl- 5,6-dihydronaphthalen-2-ylethynyl) benzoic acid, are the genes involved in particular pathological conditions, involving the promoter for collagen type VII (COL7A1), such as: certain genetic pathological conditions, for instance epidermolysis bullosa, and some cases of wound healing which can be improved by stimulating the synthesis of collagen type VII, promoting epithelialization.

Thus, the potentiation of its activity with these antagonists may prevent or treat such pathological conditions.