The invention relates to a method of selecting a cellular population comprising preadipocytes expressing the Pref-1 factor and use thereof for preparing a cellular model containing adipocytes, in particular a model of adipose tissue.

PRIOR ART

Adipose tissue, the last tissue to develop in an adult, is among the most abundant in the human body. White adipose tissue is the body's principal site for energy storage in the form of triglycerides and its mass can vary considerably depending on the energy status of the body and in pathological situations such as obesity or cachexia or as part of the ageing process. This variation is due to the increase in size (hyper-hypotrophy) and/or number of adipocytes (hyper- hypoplasia). Lipolysis and lipogenesis are the two mechanisms involved respectively in these processes of release or accumulation of triglycerides by adipocytes and are influenced by numerous hormonal, environmental and metabolic factors.

Over the course of the last decade adipocytes gained the status of secreting cells, synthesizing a large number of components of the extracellular matrix as well as molecules of a peptide nature. Among the latter, we may mention adiponectin, which is a protein of 244 amino acids that is only secreted by differentiated adipocytes and which thus constitutes one of the markers of maturation of the adipocytes indicating their functionality. In the skeletal muscles, the essential action of adiponectin seems to take place at the muscular level and to lead to a decrease in the plasmatic levels of free fatty acids and triglycerides (J Clin Endocrinol Metab, 2003. 88(5): 2014-18) . At the hepatic level , adiponectin causes a decrease in fatty acid transporting proteins as well as a decrease in the hepatic content of triglycerides. Adiponectin is also a hormone that is important because of the negative regulation that it maintains at the level of haematopoiesis and the immune system (Blood, 2000. 96(5): 1723-32). It is thought to inhibit the multiplication of myelomonocyte cell lines by induction of the phenomenon of apoptosis. Thus, it might notably have an anti-inflammatory action by modulating the functions of mature macrophagic cells, for example by moderating their phagocytic action and their production of tumour necrosis factor alpha TNFa. Adiponectin is regulated by a pathway that is dependent on the PPAR Y receptors (Peroxisome proliferator-activated receptors γ. Diabetes, 2001 . 50(9): 2094-09).

Study of its synthesis by adipocytes is therefore of high pharmaceutical interest.

The preadipocytes, adipocyte precursors, are cells of mesenchymal origin that surround the adipocytes. They are contained in the "vascular stromal fraction". Culture of the vascular stromal fraction was first used for studying the differentiation of preadipocytes into mature adipocytes. Several studies indicate that the vascular stromal fraction is a source of multipotent stem cells with abilities for differentiating into mature adipocytes, chondrocytes, myoblasts, osteoblasts and endothelial cells in suitable conditions and a suitable environment (Sang Thrombose Vaisseaux, 2005. 17 (2): 101 -08).

The differentiation of adipocytes from mesenchymal precursors is a multistage process, firstly involving adipoblasts. In vitro, after an exponential growth phase, the adipoblast has a fibroblastic morphotype. At confluence, cessation of cell growth seems to be required for involvement of the cells in the differentiation into adipocytes. The adipoblasts then become preadipocytes.

The process of differentiation can be divided schematically between early and late events. However, at present, little information is yet available concerning control of the early stage of transition between adipoblast and preadipocyte.

Several transcriptional factors playing a sequential and cooperative role in the process of adipocyte differentiation have been identified. They belong, among others, to the family of PPARs (peroxisome proliferator-activated receptors), to the C/EBP (CAAT/ Enhancer Binding proteins) and to the family of proteins with the HLH (helix-loop-helix) motif.

Other factors are involved in the regulation of adipocyte differentiation, notably the preadipocyte factor 1 (PREF-1 ) . Also called DLK1 , PREF-1 is described as an inhibitor of adipocyte differentiation. It is a transmembrane protein possessing six "EGF-like" motifs in the extracellular part, a transmembrane domain and a cytoplasmic tail. There are four different mRNAs called PreMA, Pref-1 B, PreM C and Pref-1 D, which are the result of alternative splicing during transcription. These four forms of Pref-1 are then translated and post-translational modifications then occur to give rise to four soluble isoforms of different molecular weights. The most important is that of 41 kDa.

Pref-1 is only expressed in the preadipocytes and its expression decreases in the course of differentiation, in contrast to the transcription factors C/EBPS and PPARs, which are only detected when the process of differentiation is underway. Pref-1 is therefore a marker of preadipocytes, completely absent in mature adipocytes. It was demonstrated that Pref-1 inhibits adipocyte differentiation (Molecular mechanisms of adipocyte differentiation and inhibitory action of Pref-1. Smas CM, Sul HS. Crit Rev Eukaryot Gene Expr. 1997; 7(4):281 -98, Pref-1 , a protein containing EGF-like repeats, inhibits adipocyte differentiation. Smas CM, Sul HS. Cell. 1993 May 21 ; 73(4):725-34, Pref-1 (preadipocyte factor 1 ) activates the MEK/extracellular signal-regulated kinase pathway to inhibit adipocyte differentiation. Kim KA, Kim JH, Wang Y, Sul HS. Mol Cell Biol. 2007 Mar; 27(6):2294-308).

Deregulations of the expression of Pref-1 in humans have been studied. Thus, patent application (WO2004063222) describes ORFs (Open Reading Frames) of the human genome coding for polypeptides of factor 1 type of preadipocytes, and reagents associated with the latter, as well as ligands and antagonists directed against these polypeptides. This application also describes methods for identifying and producing these molecules, for preparing pharmaceutical compositions containing them, and for using them for the diagnosis, prevention and treatment of diseases involving a deregulation of Pref-1.

Ad ipocyte differentiation is moreover influenced by a large number of hormones, transduction signals, components of the extracellular matrix, and the cellular environment.

Physiologically, the various stages leading to the increase of adipose tissue, in particular subcutaneous, are proliferation of preadipocytes, their differentiation into adipocytes and the maturation of adipocytes, reflected in increased storage of lipids. However, when these adipocytes have reached a high level of filling with lipids, they reach a threshold and cannot store more. They then induce proliferation of new preadipocytes to continue storage of lipids. This phenomenon explains why after dieting which enables the mature adipocytes to destock lipids, a new weight gain and storage of lipids may be higher than before, and this occurs after each dieting, as the number of mature adipocytes does not decrease and the capacities for storage after dieting are systematically increased.

It is therefore important to have model for investigating cutaneous adipose tissue, in particular in mammals, and in particular in humans.

Some knowledge about adipose tissue has been put to good use in the field of skin engineering and various models including adipocyte precursors have been developed in order to study and modulate in particular adipocyte differentiation.

These models are for the most part based on stem cells derived from adipose tissues. This has the drawback of requiring a long time for adipocyte maturation.

The models of adipose tissue engineering described in the prior art are mostly based on the vascular stromal fraction in particular of mesenchymal stem cells derived from adipose tissue obtained by lipectomy or lipoaspiration, after removal of the adipocytes. Modulation between proliferation and adipocyte differentiation is then achieved by a simultaneous or sequential combination of growth factors (FCS, EGF, FGF, PDGF, NGF, SCF etc.) or of adipogenic factors (insulin, transferrin, triiodothyronine (T3), dexamethasone, isobutylmethylxanthine (IBMX) and PPARy agonist during culture (WO2007041869 - US6777231 - FR2843123).

Thus, patents FR2843123 - US6777231 described the selection of mesenchymal stem cells HLA Class I negative (measured by flow cytometry) according to a complex method based on their capacities for adhesion (before or after 12h), the fraction that adheres before 12 hours then being enriched until a quiescent population is obtained that expresses the transcription factor Oct4 expressed by mouse and human embryonic stem cells and that can auto-renew during 200 population doublings. This population, lacking preadipocytes, adipocytes, fibroblasts, endothelial cells, pericytes, mastocytes and smooth muscle cells, can then differentiate into adipocytes, but also into osteoblasts, myocytes, chondrocytes or nerve cells depending on the combination of proliferation/differentiation medium used. Several cell types can be obtained at the same time. The cells can be used in monolayers for evaluating the potential of actives as modulator of adipogenesis, lipolytic or anti-lipolytic substance and can be used for cosmetic or reconstructive surgery, synthesis of recombinant proteins or gene therapy. The screening process for identifying agents that can display lipolytic activity is characterized by 1 - seeding of stem cells in conditions for obtaining adipocytes, 2- contacting with the candidate agent, 3- evaluation of lipolytic activity. When anti-lipolytic activity is sought, the same protocol is applied, but it is anti-lipolytic activity that is evaluated for the candidate agent (FR2843123).

In skin engineering, current research aims to add a hypoderm under the dermis of the reconstructed skin to obtain adipose reconstructed skin or to optimize the models for culture of adipocyte precursors and of hypoderm.

One of the models described (WO2007/041869, J Biomaterial, 2007. 28(18):2850-2860) uses the model without biological substrate. Its morphological appearance is similar to subcutaneous adipose tissue. Moreover, it reproduces the main functions of white adipose tissue: secretion of adipokines, biosynthesis of triglycerides and lipolysis induced by the beta adrenergic receptor. This hypoderm is proposed as a model for metabolic tests in vitro or for reconstruction of autologous soft tissue in reconstructive and cosmetic surgery. It is carried out on a plastic substrate from stromal cells derived from adipose tissues from lipoaspiration. Culture is carried out in two different culture media, the first containing ascorbic acid between 20 and 200 μg/ml and the second containing an ad ipogenic stimulus such as insulin, T3 , dexamethasone, I BMX, and a PPARgamma agonist (rosiglitazone or pioglitazone). These two media can be combined or used sequentially in any order. The authors obtain a layer derived from adipose tissues of 20 to 60 μηη in a minimum of 21 -35 days before it can be manipulated and carefully stacked manually to obtain an area of 3.5 cm ² . Stacking of 3 layers gives a thickness of up to 153.5 μηη after some additional days of culture in the presence of vitamin C to obtain strong cohesion between the layers; complete culture therefore took 30 days in this example, but it may take up to 63 days. Moreover, it is mentioned that a layer derived from adipose tissue is obtained and contains a mixture of preadipocytes and adipocytes with a content of adipocytes between 20% and 90% and therefore of preadipocytes from 80% to 10%. Nevertheless, no method for detecting preadipocytes is described and only the presence of adipocytes is mentioned by staining with Oil Red O including extraction in nonidet/isopropanol and quantification at 520 nm or synthesis of adiponectin and leptin. As adiponectin only appears at 7 d ays without adipogenic induction, it is assumed that the content of preadipocytes is calculated by subtraction but this method does not allow the relative content of preadipocytes to be determined. A model of reconstructed skin is made in 43 days by seeding keratinocytes on the layer obtained either by stromal cells uniquely differentiated or not by the adipogenic cocktail, or by superposition of 2 dermal layers on 2 stromal layers. The mixture of stromal and dermal cells gives a thicker subepithelial portion. Only induction by the adipogenic cocktail provides visualization of the lipid vacuoles. The models of adipose tissues of the prior art additionally require the preparation of several layers that are to be stacked to reconstitute an adipose tissue, which has many drawbacks, including the need for multiple seedings and culture, manipulations of stacks, increased risk of contamination, lack of reproducibility, and the difficulty of industrialization or automation of the production of these models.

The drawbacks of the models of the prior art are therefore the time required for their preparation, and the complexity and multiplicity of stages required, which affects the quality of the reconstructed tissues thus obtained. DESCRIPTION OF THE INVENTION

The invention therefore has the aim of solving the technical problems mentioned above. In particular, the present invention has the aim of shortening the time for making models of adipose tissues, of simplifying their preparation, of improving their quality and reproducibility, and of permitting industrialization.

The present invention therefore has the aim of supplying a population of preadipocytes having capacity for differentiation greater than that of the prior art.

The present invention therefore has the aim of supplying a method of preparing these tissue models more quickly than those of the prior art and/or for improving the quality of this cellular population.

The inventors discovered, particularly surprisingly and unexpectedly, that selection of a population of preadipocytes expressing Pref-1 , despite inhibiting adipocyte differentiation, made it possible to obtain a population of mature adipocytes more quickly, thus reducing the duration of the differentiation phase and at the same time increasing the quality of the tissues thus obtained. This observation is all the more surprising because Pref-1 is known for its opposite action, inhibiting adipocyte differentiation.

The present invention thus relates to a method of cell culture for preparing a cellular population comprising preadipocytes, said method comprising:

- isolation of a cellular population containing preadipocytes, and preferably obtained from adipose tissue of a human being;

- selection of a population of preadipocytes enriched with preadipocytes expressing Pref- 1 ;

- proliferation of the population of preadipocytes;

- optionally differentiation of at least a proportion of the preadipocytes into adipocytes. The present invention also relates to the population of preadipocytes thus obtained by this method which has a greater capacity for differentiation than those available on the market, notably the stem cells derived from adipose tissue according to the techniques previously described in the prior art (Adipose-derived Stem Cells called ADSC), as is demonstrated in example 3.

This method according to the invention makes it possible to supply a new model of adipose tissue, and notably to make a model of reconstructed hypoderm called "hypoderm" according to the invention or a model of reconstructed adipose tissue called "adipose tissue" according to the invention more quickly and therefore also of higher quality. This method in fact makes it possible to prepare models that do not have the drawbacks of the prior art, notably not necessarily requiring the stacking of several layers, which reduces the risk of contamination, increases reproducibility, and allows for the possibility of industrialization or automation for making said models.

The present invention thus relates to these cellular and/or tissue models.

These models are particularly useful as a model tissue for testing in the cosmetic, dermatological and/or pharmaceutical industry, notably for screening active principles and in particular for detecting their properties on adipose tissues. These models can also constitute a support for tissue engineering, surgical repair, reconstructive and cosmetic surgery, cosmetic dermatology, notably for the filling of soft tissues, for example for the filling of wrinkles, loss of substance and scars. Furthermore, these models constitute alternatives models to testing on animals within the scope of tests of pharmacotoxicology and/or activity of active ingredients, notably in cosmetics and pharmaceutics. Finally these models constitute models that are particularly advantageous for investigating the mechanisms involved in the phenomena of lipogenesis and which may be deregulated in certain pathologies such as obesity.

The invention also relates to a cellular population essentially of preadipocytes, enriched with preadipocytes expressing the marker Pref-1 .

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a method of cell culture for preparing a cellular population comprising preadipocytes, said method comprising: - isolation of a cellular population containing preadipocytes, and preferably obtained from adipose tissue of a human being;

- selection of a population of preadipocytes enriched with preadipocytes expressing Pref- 1 ;

- proliferation of the population of preadipocytes;

- optionally differentiation of at least a proportion of the preadipocytes into adipocytes. "Cellular population containing preadipocytes" means a population of various types of cells, at least a proportion of which is constituted of preadipocytes.

"Population of preadipocytes enriched with preadipocytes expressing Pref-1 " means a cellular population containing preadipocytes enriched with preadipocytes expressing Pref-1.

"Enriched with preadipocytes expressing Pref-1 " means that the cellular population is specially selected for increasing the number of preadipocytes expressing the preadipocyte factor-1 cellular marker (Pref-1 ). The method thus comprises selection of at least one population of preadipocytes expressing Pref-1 .

"Preadipocytes" according to the preferred embodiment of the invention means cells extracted from adipose tissue and expressing the preadipocyte factor-1 cellular marker (Pref-1 ).

According to a preferred embodiment of the present invention, the stage of isolation of a cellular population containing preadipocytes comprises cellular extraction from adipose tissue, and preferably by lipectomy of human adipose tissues or lipoaspiration of human adipose tissues preferably the vascular stromal fraction, then tissue separation in order to obtain stromal cells, notably by fractional enzymatic digestion, preferably with a collagenase. The adipocytes and erythrocytes are preferably withdrawn.

According to the invention, a population of preadipocytes enriched with preadipocytes expressing Pref-1 is selected. Various conventional techniques of cellular separation known by a person skilled in the art can be used for this selection stage. We may mention, as particularly suitable examples, the techniques of cell sorting by flow cytometry called CM F or FACS, magnetic beads as well as techniques of solid phase separation, notably MACS.

Particularly advantageously, the inventors also d iscovered that a population of preadipocytes expressing Pref-1 is a population of preadipocytes adhering spontaneously to the cell culture support, preferably plastic plate, dishes and/or flashes. As demonstrated in the examples, expression of Pref-1 is then maintained in the population of preadipocytes thus selected.

Thus, according to a particularly advantageous embodiment, the stage of selection of the preadipocytes comprises contacting the preadipocytes with a cell culture support, preferably plastic plate, dishes and/or flashes, said selection stage comprising selection of preadipocytes adhering to the cell culture support, preferably after 30 minutes, and preferably after 45 minutes, and more preferably after or at about 1 hour of culture in contact with the cell culture support, preferably plastic plate, dishes and/or flashes. Preferably, this contacting is not more than 6 hours, preferably not more than 2 hours.

The population of preadipocytes can be characterized by its expression of the marker Pref-

1 by means of conventional techniques for measuring gene and/or proteome expression.

Preferably, the population of preadipocytes enriched with preadipocytes expressing Pref-1 contains at least 50% of preadipocytes expressing Pref-1 , more preferably at least 70% and more preferably at least 90% of preadipocytes expressing Pref-1. According to a particularly advantageous embodiment, the population of preadipocytes is constituted essentially of preadipocytes expressing Pref-1. According to a particularly advantageous embodiment, the population of preadipocytes is constituted exclusively of preadipocytes expressing Pref-1 .

The invention thus relates, according to another aspect, to a cellular population essentially comprising preadipocytes, characterized in that it is enriched with preadipocytes expressing the marker Pref-1 .

According to the invention, the stage of proliferation of the preadipocytes comprises very advantageously the seeding and culture of the preadipocytes in the presence of a proliferation medium comprising Fibroblast Growth Factor (FGF) or Epidermal Growth Factor (EGF), preferably at a concentration of about 10 ng/ml, and preferably with FGF, and more preferably with beta-FGF. The proliferation medium is also called amplification medium.

Accord ing to an advantageous embodi ment of the invention, the population of preadipocytes enriched with preadipocytes expressing Pref-1 is amplified in the proliferation medium from 1 to 5 times usually designated as passage 0 to passage 5.

Culture in the proliferation medium is advantageously carried out for 12 to 1 8 days, preferably for 14 and 16 days and optimally for 14 days, advantageously in the proliferation medium according to the invention described below. It was in fact discovered that at 14 days of culture after seeding in this medium, the preadipocytes have optimum capacity for differentiation. Surprisingly, a longer culture time does not improve this capacity.

Preferably, the cellular population obtained at the end of the proliferation stage contains predominantly preadipocytes expressing Pref-1 ; preferably at least 90 % of the cell ula r population expresses Pref-1.

The method of cell culture according to the invention is useful for preparing a cellular population containing preadipocytes expressing Pref-1 havi ng opti m um capacities for differentiation. This population is called undifferentiated.

This population can be amplified when renewing culture in the proliferation medium with larger cell culture support, for instance double surface to obtain higher number of preadipocytes expressing Pref-1 in the cellular population containing preadipocytes. As demonstrated in examples, even after 5 and 6 passages, Pref-1 expression is stable and thus amplified cells keep optimum ability to differentiate into adipocytes.

According to an advantageous embodiment of the invention, the method comprises a stage of differentiation of at least a proportion of the population of preadipocytes into adipocytes. This stage advantageously comprises culture of the preadipocytes in the presence of a differentiation medium comprising at least one of the following substances: a calf serum, preferably fetal (FCS), hydrocortisone, dexamethasone, and an activator of cAMP for example 3- isobutyl-1 -methylxanthine (IBMX), and preferably a mixture comprising a calf serum, preferably fetal (FCS), hydrocortisone, and dexamethasone and optionally I BMX. The cells are preferably seeded and cultivated in a differentiation medium according to the invention described below. Culture in the differentiation medium is advantageously carried out for 3 to 20 days, preferably for 4 to 10 days and more preferably for 7 days. Preferably, the method comprises culture of the population of preadipocytes in the proliferation medium carried out for 12 to 18 days, then in the differentiation medium for 3 to 20 days, preferably for 4 to 10 days and more preferably for 7 days.

According to this embodiment, the method according to the invention gives a cellular population containing preadipocytes and adipocytes at a controlled content. According to an advantageous embodiment, the mixed population of preadipocytes and adipocytes contains predominantly mature adipocytes, preferably at least 70% of mature adipocytes.

According to a particular embodiment of the invention, the method can be implemented without matrix substrate such as a gel, a sponge notably based on collagen or collagen, chitosan, glycosaminoglycans (GAG) or any three-dimensional network for supporting the cell culture, and does not require intermediate combining of different sub-parts or layers of the model in order to obtain the desired state of differentiation and thickness for preparing a tissue model according to the invention. This greatly simplifies preparation of the models and means that the method can be carried out reproducibly and can be automated.

The models according to the invention include in particular a monolayer of cells, a tissue model, in particular a model of hypoderm or a model of adipose tissue.

According to the invention, "hypoderm" means a tissue obtained by proliferation and differentiation of at least a proportion of the population of preadipocytes according to the method of the invention, optionally in the presence of other cell types, preferably fibroblasts and/or keratinocytes and/or melanocytes and/or endothelial cells and/or interstitial dendritic cells (IDC) and/or Langerhans' cells (LC) optionally on an inert substrate. When the hypoderm contains fibroblasts, it is commonly called "adipose dermis".

According to the invention, "model of adipose tissue" means a reconstructed 3D tissue, being in layers or in a substrate of the sponge type, notably based on collagen, or based on collagen, chitosan, glycosaminoglycans (GAG) at least one of which comprises adipocytes obtained after proliferation and differentiation of preadipocytes according to the method of the invention. Preferably the model of adipose tissue contains or is a hypoderm according to the invention. The model of adipose tissue is commonly called "model of reconstructed adipose skin" when it contains fibroblasts and keratinocytes. A model of reconstructed adipose skin containing melanocytes is commonly called "model of pigmented reconstructed adipose skin".

The invention thus relates to a hypoderm comprising a population of preadipocytes and adipocytes obtained according to the method of the invention. Advantageously, the adipocytes are mature and differentiated.

According to an advantageous embodiment, the hypoderm contains predominantly mature adipocytes, preferably at least 70% of mature adipocytes.

The method according to the invention makes it possible to obtain, surprisingly, hypoderm having a greater thickness than the models of the prior art. The hypoderm obtained according to the invention has a thickness of at least 70 μηη (micrometers), preferably more than 100 μηη (micrometers), and more preferably more than 130 μηη (micrometers), notably obtained in less than 28 days from a population of preadipocytes according to the invention. This very significant advantage relative to the model of the prior art makes it possible to avoid superposition of layers for preparing adipose tissue, which is a drawback of the models of the prior art. This method of culture, which is easy and can be automated, makes it possible to obtain an adipose tissue or reconstructed hypoderm after a phase of proliferation of the population of preadipocytes in a first medium called proliferation medium permitting the formation of an adipose tissue or three-dimensional undifferentiated hypoderm then in a so-called differentiation medium permitting transformation of the preadipocytes into adipocytes.

According to a particularly advantageous embodiment, the model of hypoderm can contain fibroblasts. The method according to the invention in fact makes possible the co-seeding of fibroblasts and its coculture with the population of preadipocytes according to the invention. According to an alternative embodiment, the seeding of fibroblasts can be carried out on the hypoderm after it is obtained. Generally, seeding of the fibroblasts can therefore be done before, at the same time or after that of the preadipocytes.

According to a particularly advantageous embodiment, the model of hypoderm can also contain endothelial cells, interstitial dendritic cells (IDC) and/or Langerhans' cells (LC), preferably a population of mixed IDC/LC cells such as described in patent application WO03/050271 . The method according to the invention in fact makes possible the co-seeding of endothelial cells and its coculture with the population of preadipocytes according to the invention.

This embodiment has the advantage of supplying a cellular support that is particularly favourable for culture of the population of preadipocytes owing to the capacity of the fibroblasts to secrete and organize the components of the extracellular matrix.

The model of hypoderm or the model of fibroblast hypoderm can be epidermized by keratinocytes cultivated in immersed phase for a sufficient time to permit covering of the hypoderm or of the fibroblast hypoderm then in emerged phase for a sufficient time to permit epidermal differentiation. During culture of the stromal fraction, the adipodermal proliferation medium is used first, then that for adipodermal differentiation. During seeding of keratinocytes, a medium of the Green's medium type is used, which is then depleted during the phase of raising to the air-liquid interface.

Advantageously, the invention is implemented in the cell culture device described in patent (FR2881434) comprising at least one cell culture well intended to receive cells with their culture medium containing at least one layer of porous cell substrate material, said layer being obtained by dehydration of an aqueous gel poured directly into the bottom of said well, or into the bottom of a carriage or insert of suitable size so that it can be inserted in the well. This method of preparation of the cell culture device makes possible the reproducible, safe and reliable high- throughput screening of active principles by a platform that can be automated. The culture substrate corresponds to a gel comprising a mixture of collagen, at least one polysaccharide and chitosan, optionally modified, for example having a degree of acylation, preferably of acetylation, adjusted depending on the application envisaged, different degrees of acetylation being well known by a person skilled in the art and in particular described in the European document (EP 0296 078).

Advantageously the models of hypoderm and of fibroblast hypoderm , optionally epidermalized, can be made in a sponge notably based on collagen or based on collagen, chitosan, glycosaminoglycans (GAG) in particular comprising by weight of the aqueous composition:

between 60 and 90% of collagen between 10 and 30 wt.% of chitosan

- between 0 and 15% of GAG

A second alternative described in patent US6777231 consists of using biocompatible matrices composed of monomers of glycolic and lactic acids, propyl fumarate, caprolactone, glycosaminoglycans, proteins, polysaccharides, polydihydroxy acids, polyorthoesters, polyanhydrides, polyphaophazenes, and other biocompatible synthetic polymers. These agents can be used mixed either to form a crosslinked hydrogel or a sponge notably based on collagen or on collagen, chitosan, glycosaminoglycans (GAG) or a network with porosity between 100 and 300 μηη. These models can be used in particular for making filling kits for surgery, in particular human surgery, such as the filling of wrinkles and of scars as well as for the grafting of adipose tissue in cases of deficiency pathologies.

The functionality of the hypoderm was demonstrated in the examples notably by the expression of markers of maturation of the adipocytes, for example adiponectin, and in the case of a fibroblast hypoderm, from the synthesis of the constituents of the extracellular matrix. Moreover, the functionality of the hypoderm was demonstrated by its capacity for response to conventional slimming agents, notably caffeine, in particular from measurements of changes in thickness and secretion of adiponectin in response to the agents.

The invention thus makes it possible to produce a reliable, economical, and reproducible model very quickly, and production can be automated.

Automated application of the hypoderm according to the invention is possible, in particular according to one or more of the following embodiments:

- no matrix substrate such as a gel, a sponge notably based on collagen or on collagen, chitosan, glycosaminoglycans (GAG) or any three-dimensional network is used for supporting the cell culture, which can be performed in multi-well culture plates with from 6 to 96 wells in immersed phase

- preadipocytes previously selected and optionally characterized can be seeded by an automated station or robot

- the culture phase can be performed using only 3 different media sequentially, renewal of which can be controlled by an automated station or robot

- the full thickness is obtained without any other stage such as a stage of superposition of layers.

The invention also relates to a model of adipose tissue comprising a hypoderm defined previously.

Advantageously, the hypoderm, optionally fibroblastic and/or epidermalized has a sufficient thickness to be used alone for making a model of adipose tissue without the need for self- assembly of 3 layers, which was the major drawback of the models of the prior art. In fact, self- assembly of different layers greatly complicates the method of production of models of adipose tissue or even makes them impossible and does not allow certain pharmaceutical applications to be envisaged, in particular injection. In contrast, the model of adipose tissue according to the invention makes its use possible by injection, for filling in surgery.

The models according to the invention find numerous uses.

The invention also relates to a method for screening active principles, characterized in that it employs a model according to the invention, preferably a hypoderm or a model of adipose tissue. The method according to the invention allows a model to be made that is particularly suitable for the screening of active principles, and notably make it possible to take into account the response of the two populations of preadipocytes and adipocytes. It thus makes it possible to supply active principles acting on the preadipocytes and/or adipocytes part of skin in order to modulate the degree of proliferation and differentiation of preadipocytes, as well as the state of maturity of the adipocytes and hence select slimming active principles and/or those for the prevention and/or treatment of obesity.

Advantageously, the present invention permits identification of substances having the capacity to reduce the proliferation and differentiation of preadipocytes and/or only the level of differentiation/maturity of adipocytes in subjects notably having a skin whose hypoderm is excessive, in particular persons who are overweight or obese or suffer from lipodystrophies, whether or not they have hormonal disorders or fluctuations, and whether or not they have problems of drainage, in particular lymphatic.

The models according to the invention can be used for screening active principles and identifying those that act on the preadipocytes and/or adipocytes part in order to modulate the degree of proliferation and/or differentiation of the preadipocytes and/or on the state of maturity of the adipocytes with the aim of increasing or reducing an individual's adipose skin mass.

The growth of subcutaneous adipose tissue is a function both of the proliferation of the preadipocytes and of their potential for maturation. It is therefore advantageous to evaluate both the effect of a substance on cellular viability, cellular proliferation and differentiation of the preadipocytes in order to evaluate the overall effect of a substance. The stages of measurement can be carried out during the method of culture according to the invention or on the models obtained following application of the method according to the invention.

According to a preferred embodiment, the method of cell culture comprises one or more stages of measurement of the gene and/or protein expression of Pref-1 and/or can be coupled to other criteria for investigation, notably:

- investigation of cellular viability

- investigation of cellular proliferation and notably thickness

- investigation of cellular differentiation and notably measurement of thickness, size of vacuoles and/or secretion of adiponectin

- investigation of expression of the components of the cutaneous extracellular matrix such as collagens, in particular type I, III, V, VI, and/or

- investigation of all factors involved in lipolysis and/or lipogenesis.

Advantageously, the method of evaluation of gene expression is a method of RT-PCR, preferably quantitative.

Advantageously, the method of evaluation of protein expression is a method of Western blot, flow cytometry or immuno-localization on cells or on histological sections.

Advantageously, the quality of the reconstructed hypoderm is evaluated after staining with haematoxylin-phloxine-saffron or Masson trichrome or haematoxyl i n-eosin and permits measurement of thickness.

Advantageously, evaluation of the lipid vacuoles is carried out after staining the lipids for example with Oil Red O. Advantageously, lipolysis is evaluated from secretion of glycerol after stimulation of the reconstructed hypoderm with an adrenergic agonist.

Advantageously, lipogenesis is evaluated in particular by measurement of adiponectin secretion by ELISA.

Investigations of cellular viability and of cellular proliferation can be performed on the preadipocytes after their characterization from the expression of the PREF-1 markers using methods well known by a person skilled in the art for example manual or automatic cell counting, t e s t i n g w i t h M T T ( 3-(4-5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium) or PNPP (paranitrophenyl phosphate).

The investigations of proliferation are performed on preadipocytes in a medium containing optimum concentrations or depleted of proliferation factors but not containing adipogenic factors. The preadipocytes are in the proliferative phase and cannot differentiate owing to lack of adipogenic factors; for example the substance screened is thus considered to be an active principle:

- not modulating the proliferation of preadipocytes if the number of cells does not change or changes little (±15%) from the number of cells not treated with the substance screened after the same incubation time;

- increasing the proliferation of preadipocytes if the number of cells is 15%, preferably 30%, greater than the number of cells not treated with the substance screened after the same incubation time;

- decreasing the proliferation of preadipocytes if the number of cells is 15%, preferably 30%, less than the number of cells not treated with the substance screened after the same incubation time; in this last case, the substance screened can also be toxic to the preadipocytes. Toxicity and decrease in potential for proliferation can be differentiated by the observation of phenomena of necrosis or apoptosis.

The following studies of gene and protein expression are to be understood as at a constant number of cells.

Increase in gene expression means, according to the invention, an increase greater than or equal to 20% (or 1 .2 times), preferably 50% , preferably 1 00% (or 2 times) the level of expression of the cells in culture without the screened molecule, at the same incubation time.

Decrease in gene expression means, according to the invention, a decrease greater than or equal to 20% (or 0.8 times), preferably 50% (or 0.5 times) the level of expression of cells in culture without the screened molecule, at the same incubation time.

Increase in protein expression means, according to the invention, an increase greater than or equal to 10% (or 1 .1 times), preferably 30% (or 1 .3 times) the level of expression of the cells in culture without the screened molecule, at the same incubation time.

Decrease in gene expression means, according to the invention, a decrease greater than or equal to 10% (or 0.9 times), preferably 30% (or 0.7 times) the level of expression of the cells in culture without the screened molecule, at the same incubation time.

The investigations of differentiation can be performed on preadipocytes in a medium containing optimum concentrations or depleted of proliferation factors with or without adipogenic factors. The preadipocytes are in the proliferative phase and can differentiate depending on the presence of adipogenic factors. For example, the substance screened is thus considered as being an active principle:

- lowering the potential for maturation of the preadipocytes if gene and/or protein expression of Pref-1 is increased;

- triggering maturation of the preadipocytes if gene and/or protein expression of Pref-1 is decreased.

Although the release in lipids stored by mature adipocytes can easily be evaluated from the release of labelled glycerol, the possibility of adipocyte dedifferentiation has never been clearly demonstrated since it is necessary to start from a population no longer containing preadipocytes or mesenchymal precursors. It would optionally be possible to study this phenomenon on mature adipocytes no longer expressing Pref-1 but expressing at least one marker such as adiponectin, in differentiation media preferably without proliferative agent and adipogenic factors. In this medium, the adipocytes may possibly dedifferentiate, and for example the screened substance is considered as being an active principle:

- not modulating the maturation of the adipocytes if the difference in gene and/or protein expression of Pref-1 does not change or changes little (±10%) from the gene expression of cells not treated with the screened substance during the same incubation time;

- reversing the maturation of the adipocytes if the gene and/or protein expression of Pref- 1 is increased by more than 10%, preferably 30%, relative to the gene expression of cells not treated with the screened substance during the same incubation time.

The substances thus selected for their capacities for stimulating the proliferation of the preadipocytes, their differentiation into adipocytes and the maturation of the adipocytes can be used as active principle for preparing a composition for modulating the reactivity of an individual's cutaneous adipose tissue, in particular for increasing the proliferation and differentiation of the preadipocytes and/or only the level of differentiation/maturity of the adipocytes in subjects notably having skin whose hypoderm is not sufficiently thick or firm, such as elderly human beings or those suffering from cachexia or lipodystrophies, whether or not they have hormonal disorders or fluctuations.

The invention further relates to the use of a model according to the invention, in particular a hypoderm or a model of adipose tissue according to the invention, as a support for tissue engineering, surgical repair, reconstructive surgery, cosmetic surgery, cosmetic dermatology, notably for the filling of soft tissues, for example for the filling of wrinkles, loss of substance and scars.

The present invention also relates to a pharmaceutical composition comprising a population of preadipocytes according to the invention. Said composition advantageously comprises a pharmaceutically acceptable excipient, and optionally other pharmaceutically active substances.

The present invention also relates to an injectable pharmaceutical product containing a population of pread ipocytes accord i ng to the invention in suspension in a suitable pharmaceutical support such as an injectable gel or liquid or other injectable pharmaceutical vehicle. The present invention also relates to the use of said product or composition for filling wrinkles and loss of substance, prevention and/or repair of scars, and/or of lipodystrophies by subcutaneous or intramuscular injection.

The invention also relates to the use of a model according to the invention, in particular a hypoderm or a model of adipose tissue according to the invention, as a model for performing tests for pharmacotoxicology and/or activity of active ingredients, notably in cosmetics and pharmaceutics.

The invention also relates to a medium for cellular proliferation of the population of preadipocytes enriched with preadipocytes expressing Pref-1 according to the invention, characterized in that it comprises preadipocytes and FGF or EGF, preferably at a concentration between 5 and 20 ng/ml, more preferably between 8 and 12 ng/ml. The inventors have in fact demonstrated that this medium is particularly advantageous for maintaining expression of Pref-1 and constitutes the preferred proliferation medium (experiment 1 B/D).

According to one embodiment, the population of cells isolated and/or selected according to the method of the invention can be frozen in a suitable medium. A medium that is particularly suitable for freezing preadipocytes and preserving their undifferentiated character has been identified, and comprises fetal calf serum to more than 50%, preferably to more than 75%, and preferably to more than 80% , and preferably to about 90% , by volume relative to the total volume of the medium. The inventors have in fact demonstrated that said freezing medium permits expression of Pref-1 to be maintained (experiment 2).

The invention thus relates to a population of preadipocytes, optionally enriched with preadipocytes expressing Pref-1 in a freezing medium according to the invention.

The invention also relates to a med ium for cellular differentiation of preadipocytes, characterized in that it comprises a population of preadipocytes, preferably expressing Pref-1 , and fetal calf serum, preferably at a concentration between 5 and 20% by volume relative to the total volume of the medium, hydrocortisone, dexamethasone, and an activator of cAMP for example 3-isobutyl-1 -methylxanthine (I BMX), and preferably a mixture comprising calf serum, preferably fetal (FCS), hydrocortisone, and dexamethasone, and optionally IBMX. The inventors have in fact demonstrated that said differentiation medium permits expression of Pref-1 to be maintained (experiment 3).

Other aims, characteristics and advantages of the invention will become clear to a person skilled in the art after reading the explanatory description, which makes reference to examples, which are only given as illustration and are not intended to limit the scope of the invention in any way.

The examples form an integral part of the present invention and any characteristic appearing novel relative to any prior art from the description taken in its entirety, including the examples, forms an integral part of the invention in its function and in its generality.

Thus, each example is of general scope.

Moreover, in the examples, all the percentages are percentages by weight, unless stated otherwise, the temperature is expressed in degrees Celsius unless stated otherwise, and the pressure is atmospheric pressure, unless stated otherwise. The percentages of serum are expressed by volume relative to the total volume of the medium in question. EXAMPLES

Example 1 : Protocol for optimization of extraction and proliferation of preadipocytes

A- Extraction

The subcutaneous adipose tissue (AT) is collected from the operating theatre during cosmetic or reconstructive surgery, lipectomy, preferably lipoaspiration. The preadipocytes are extracted for example according to Bjorntorp's method or by a gentler method by syringe:

- The classical method of lipoaspiration, used by many surgeons. Removal is via a cannula introduced under the skin through small incisions. It is connected to a powerful suction system. The standard suction pressure is 700 mmHg. Through movements of the cannula to and fro, guided by the surgeon, the excess fat is broken up and aspirated.

- The adipose tissue can also be removed by a technique similar to the lipostructure of Coleman et al. The basic principle is the atraumatic character of manipulation of the adipose tissue. Extraction must be done using a special suction cannula with a diameter of 3 mm, a length of 15 cm, having a tip made of foam and a double orifice wide enough for the adipocytes to pass through. The vacuum in the syringe is created manually and progressively, so that the adipocytes are not subjected to excessive vacuum. Making multiple tunnels during extraction can reduce the risk of haemorrhage and trauma.

Tissue separation to obtain the stromal cells is carried out in a stage of fractional digestion. Briefly, the adipose tissue is digested in a solution of collagenase (EC 3.4.24.3 Roche 0.18-0.22 U/ml) at varying concentration from 0.2 to 1 mg/ml with constant agitation for 15 to 60 minutes in order to select the optimum conditions for obtaining the best rate of growth and extraction yield of cells expressing Pref-1 .

The action of the enzyme is then stopped by adding a solution of DMEM supplemented with 10% by volume of FCS. The floating adipocytes are withdrawn.

After centrifugation, the cell pellet is taken up in PBS to remove all traces of serum and then it is filtered at 70 μηη. The remaining erythrocytes are lysed by adding a lysis buffer (NH4CI at 8.7 g/L) for 10 minutes at 37°C. Repeat centrifugation removes the residues of erythrocytes from the cells of the vascular stromal fraction. B- Optimization of a proliferation medium for preserving the protein expression of the marker Pref-1

An experimental matrix was carried out to evaluate the effect of three different growth factors selected according to the literature: EGF, FGF and VEGF, at two concentrations 0 and 10 ng/ml in a base medium containing: a DMEM/Ham's F12 mixture, supplemented with 10% of fetal calf serum, 0.4 μg/ml of hydrocortisone, 0.12 I U/ml of insulin, 24.3 μg/ml (or 18.5 nM) of adenine, 2 nM of triiodothyronine, isoprenaline (Isuprel at 0.4 μg/ml) or cholera toxin 10 nM and of a mixture of antibiotics (100 lU/ml of penicillin, 20 μg/ml of gentamicin and 1 μg/ml of Fungizone).

Briefly, the preadipocytes are seeded in plates at a density of about 7000 cells per cm ² . The medium is changed every three days until confluence of the cells. At confluence, the cells are detached with trypsin 0.0125%/EDTA 0.005%, centrifuged at 1000 rpm, and taken up in 10 ml of medium. The cells are counted on Malassez cells. The population doubling time (PD) and growth rate (GR) are calculated. The population doubling time (PD) corresponds to the number of multiplications that occurred during the culture time and the growth rate (GR) corresponds to the number of population doublings per day.

PD = [In (number of cells obtained / number of cells seeded)]/ln2

GR = PD/(number of days of culture)

The cells are seeded in the appropriate culture medium and cultivated at 37°C and 5% CO2. Culture is performed in triplicate for 6 passages (from P0 to P5) on precursors obtained from three different donors. C- Characterization of the preadipocytes by Western blot of the marker Pref-1

Selection of the proliferation medium was therefore carried out taking into account expression of Pref-1 in the preadipocytes in this medium. The selection criteria were to maintain protein expression of Pref-1 obtained on the day of extraction throughout the passages in the different proliferation media tested.

Briefly, the cells of non-adherent fraction, adherent for 1 hour and of the total fraction were washed once with phosphate buffer (PBS), and the proteins were extracted for 30 min at 4°C in the lysis buffer (TBS - Tris 50 mM, NaCI 250 mM), pH 7.5, 0.1 % sodium dodecyl sulphate (SDS), 2 mM dithiothreitol (DTT), in the presence of protease inhibitors. The lysates were centrifuged for 15 min at 13 000 g at 4°C. The supernatants are diluted with Laemmli buffer in the presence of beta mercaptoethanol before electrophoresis. The same amount of sample (equivalent content in total proteins) is deposited for each condition. For immunodetection, the proteins are separated by electrophoresis on SDS-polyacrylamide 4-12% gel. The proteins were transferred onto a nitrocellulose membrane. The membranes are then saturated for one hour at room temperature in TBS buffer in the presence of 3% of bovine serum albumin (BSA). The proteins are finally immunodetected after incubation of the goat or mouse anti-human PREF-1 primary antibody, diluted in a 1 % solution of TBS/BSA at 4°C overnight. The primary antibody was then detected with an anti-goat or monkey or-mouse secondary antibody coupled to peroxidase using a chemiluminescent substrate or detected with a secondary antibody coupled to Alexa 488 revealed in fluorescence. The intensity of the bands can be evaluated by image analysis and referred to the content of total proteins in the sample. An anti-actin Western blot is used for normalizing the results.

D- Results

The results demonstrate that the best extraction in a reasonable time is concentration performed for 30 min. The optimum collagenase concentration is selected at 0.5 mg/ml since the cumulative results for times of 15 and 30 min are the most advantageous at this concentration.

The results of Western blot using the Pref-1 marker are presented in Fig.1 : result 1 being obtained with total fraction (adherent +not adherent), result 2 being obtained with not adherent fraction, result 3 being obtained with adherent fraction. These results show that the adherent fraction expresses Pref-1 . Moreover, these results made it possible to confirm the "preadipocyte" phenotype of the cells cultivated in the medium containing for example FGF at 10 ng/ml. The results clearly demonstrate that the fraction that underwent a stage of adhesion for 1 hour can give a population enriched with early preadipocytes strongly expressing Pref-1. This protocol, which includes a stage of adhesion, makes it possible to select preadipocyte cells expressing Pref-1 from the stromal cells extracted. Adhesion for one hour is sufficient to enrich the population in preadipocytes expressing Pref-1 , but a longer time does not give a significant difference.

E- Results for extraction medium and proliferation medium

Mathematical analysis was able to show that:

Taken in isolation, beta-FGF and EGF at 10 ng/ml exert a strong positive influence, whereas vEGF does not lead to a response, contrary to what was expected according to the prior art.

The most striking interaction is the negative effect that EGF exerts on beta-FGF. It might be thought that combining several growth factors having a positive influence would potentiate the action instead of reducing it, which is not so.

Overall, FGF gives the best proliferation results.

Concerning expression of the marker Pref-1 (Fig. 2), it was demonstrated that the medium containing EGF or FGF makes it possible to maintain expression of Pref-1 . In fact, the preadipocyte phenotype is maintained in media containing EGF or FGF at 10 ng/ml at least after 5 passages. The Pref-1 antibody does not recognize any of the proteins extracted from the fibroblasts (negative control), thus demonstrating its specificity.

Example 2: Selection of a freezing medium for preadipocytes for preserving their undifferentiated character

At preconfluence, the preadipocytes are trypsinized and the cellular suspension is taken up in 1 ml of freezing medium. 3 freezing media were tested in order to select the one giving the best cellular uptake after thawing.

Table 3: Composition of the 3 freezing media as percentage by volume

On thawing, the preadipocytes obtained from two different donors are resuspended in the proliferation medium, and the live and dead cells are counted to evaluate the percentage cellular viability as well as the yield from 2 passages. The cells are seeded in 3 plates of 25 cm ² at a rate of about 10 000 cells/cm ² in the proliferation medium. At confluence, the cells were trypsinized, and the population doubling time (PD) and growth rate (GR) were calculated according to the formula given in example 1 . The preadipocytes were also seeded in an additional plate to verify the preadipocyte phenotype of the cells thawed at the 3rd passage in comparison with preadipocytes cultivated up to P3 but without freezing according to the protocol described in example 1 .

Analysis of cell morphology by light microscopy demonstrates that regardless of the freezing medium used, cell morphology and size do not change: they all have a fibroblastic appearance of undifferentiated preadipocytic cells.

Table 4: Yield and percentage viability of cells on thawing (n=4) Regardless of the medium, the results show that the three freezing media tested permit satisfactory survival of the cells on thawing since the percentage viability is roughly equal to or greater than 70%. However, the best percentage cellular viability is obtained with the medium that contains 90% by volume of serum. The standard deviation of the percentage viability with this medium is the lowest. Therefore it gives the most constant results. Moreover, regardless of the strain, the passage or the medium, the cells are capable of adhering and resuming their proliferation. The results for the growth rate of the cells after thawing are not significantly different from one medium to another (Student test; p<0.05). These results show that the surviving cells are capable of proliferating with a growth rate identical to that before thawing.

Regardless of the freezing medium, the characteristic band of Pref-1 is present as in the cells in the 3rd passage that did not undergo freezing (Fig. 3). All the results show that the medium containing 90% by volume of serum is optimum for preserving the preadipocytes and maintaining their undifferentiated character after thawing.

Example 3: Selection of a medium for differentiation of preadipocytes into mature adipocytes

On the basis of bibliographical data, an optimum experimental strategy was used, making it possible to quantify the influence of the various adipogenic elements added to the culture medium. Three successive experimental matrices made it possible to select the minimum medium giving good differentiation of the preadipocytes into mature adipocytes (results of the second experimental matrix), the last experiment having made it possible to select an optimum medium.

In each case, the preadipocytes in passage 3 were seeded at a density of about 40 000 cells/cm ² in plates of 25 cm ² in various media. The results were interpreted according to 3 criteria: the number and size of the mature adipocytes, and the secretion of adiponectin. Mathematical analysis made it possible to study the factors separately and in interaction two by two. The adipogenic factors were selected from the factors known in the literature for modulating adipogenic differentiation: calf serum, insulin, hydrocortisone, dexamethasone, cAMP activators (I BMX, lsuprel=isoprenaline). No PPARy agonist was added to the medium. The calf serum used is not supplemented with ingredients known to have a PPARy agonist effect (Hyclone supplier certificate).

The base medium is a medium derived from the composition of Green's medium used for the culture of reconstructed skin, the formula of which was adapted on the basis of three successive experimental matrices firstly for inducing adipogenic differentiation.

Table 5: Green's medium and experimental matrix for the differentiation medium

A- First experimental matrix

The first experimental matrix comprises 1 1 media, which were compared with the proliferation medium, the following four ingredients of which were evaluated at variable concentrations relative to Green's medium.

- Fetal calf serum (FCS): 0 or 10% by volume,

- Dexamethasone: 25 nM or 1 μΜ,

- Hydrocortisone: 0 or 10 nM,

- cAMP activators: I BMX at 0.5 mM or isoprenaline (Isuprel) at 0.4 μg/ml.

I BMX is added for three days only because of its recognized cytotoxicity, whereas isoprenaline can be left throughout the culture in the differentiation medium. FCS

Experiment No. (% by Dexamethasone Hydrocortisone CAMP

volume)

A 1 0% 25 nM 0 IBMX (0.5 mM)

A 2 0% 1 μΜ 10 nM Isuprel (0.4 μςΛηΙ)

A 3 10% 1 μΜ 0 Isuprel (0.4 μςΛηΙ)

A 4 10% 25 nM 10 nM Isuprel (0.4 μςΛηΙ)

A 5 10% 1 μΜ 10 nM IBMX (0.5 mM)

A 6 10% 25 nM 10 nM IBMX (0.5 mM)

A 7 10% 1 μΜ 0 IBMX (0.5 mM)

A 8 10% 25 nM 0 Isuprel (0.4 μςΛηΙ)

A 9 0% 1 μΜ 10 nM IBMX (0.5 mM)

A 10 0% 25 nM 10 nM Isuprel (0.4 μςΛηΙ)

A 1 1 0% 1 μΜ 0 Isuprel (0.4 μςΛηΙ)

Table 6: Composition of ingredients evaluated in the differentiation media of the first experimental matrix Observation in optical microscopy (Fig. 4)

First line results obtained with: 2 weeks proliferation, second line results obtained with 3 weeks proliferation, third line results obtained with 4 weeks proliferation.

First column results obtained with no maturation, second column results obtained with 5 days maturation, third column results obtained with 19 days maturation.

In the proliferation medium (control medium), the cells organize spontaneously into thick layers, which are easily removed. No mature adipocyte within this layer is observed on the surface. The only difference between the proliferation medium and the other media, apart from the variables, is the presence of adenine and EGF. The preadipocytes cultivated in a medium without serum have difficulty adhering. The adipocytes are smaller, with small lipid lobules. In the media with serum, the cells adhere and differentiate. The number of mature adipocytes is counted and they are distributed for each medium into medium or large adipocytes.

Determination of adiponectin by radioimmunoassay

The assay is performed using a commercial kit (Linco research) for ELISA with fixed concentration of antigen labelled with iodine 125. The control is a Green's medium.

Exp. No. Mean from 2 strains

A 1 2.15

A 2 1 .15

A 3 4.15

A 4 3.55

A 5 5.95

A 6 7.35 A 7 5.7

A 8 0

A 9 2.9

A 10 0.9

A 1 1 2.65

Control 3.15

Table 7: Secretion of adiponectin as a function of the differentiation medium (ng/ml/48h) Conclusion:

Based on all the criteria, the medium must contain serum (indispensable for all the evaluation criteria), IBMX (indispensable to obtain medium and large adipocytes), hydrocortisone and dexamethasone at 25 nM (indispensable for maximum secretion of adiponectin). This medium corresponds to medium 6.

B- Second experimental matrix

The first experimental matrix was able to show the strong influence of fetal calf serum at 10% by volume and of IBMX at 0.5 mM on all the criteria investigated. However, the experiment shows that there are differences according to the type of serum, for example between newborn calf serum or fetal calf serum . Moreover, the optimum dose has yet to be refined for dexamethasone and hydrocortisone.

The second experimental matrix comprises 7 media, which were compared with the proliferation medium, the following three ingredients of which were evaluated at variable concentrations or grades relative to Green's medium containing IBMX at 0.5 mM.

Newborn calf serum (NCS) or fetal calf serum (FCS) at 10%

Hydrocortisone: 5 or 10 nM

Dexamethasone: 0 or 1 μΜ

Table 8: Composition of ingredients evaluated in the differentiation media of the second experimental matrix Observation in optical microscopy

The cells cultivated in medi um with NCS (medi um 1 , 2 , 5, 6) have a fi broblastic appearance and are organized in a thick layer. In the media with FCS (3, 4, 7), they have the appearance of mature adipocytes with lipid lobules and remain as a monolayer without forming a layer.

Table 9: Secretion of adiponectin on a strain depending on the differentiation medium (ng/ml/48h)

Conclusions:

Factors taken in isolation: No factor taken in isolation appears to be determining for the secretion of adiponectin.

Interactions between variables: detailed statistical analysis makes it possible to show a positive trend of the com bi nation of fetal calf seru m with hyd rocortisone and with dexamethasone for increasing the response. Newborn calf serum has no effect. To summarize, fetal calf serum, dexamethasone and hydrocortisone at 10 nM are therefore indispensable for maximum differentiation of preadipocytes into mature adipocytes.

C-Third experimental matrix

In this last experiment, fetal calf serum was selected at 10% by volume, hydrocortisone at 10 nM, IBMX at 0.5 mM and T3 at 2 nM. Dexamethasone was re-tested between 25 and 1000 nM and insulin between 0.150 and 1.5 Ill/ml.

The fol lowi ng med i um was the medium adopted for optimum differentiation of preadipocytes into adipocytes, i.e. a medium containing FCS, hydrocortisone, dexamethasone, and IBMX.

Table 10: Composition of the optimized differentiation medium Example 4: Comparison of adiponectin synthesis during differentiation between a cell population derived from ADSC stem cells and a cell population derived from preadipocytes selected according to the method of the invention

Protocol:

The population of preadipocytes (PAA07085) is isolated and selected according to example 1.

The population of ADSC stem cells is a classical population that is commercially available (hADSC, Invitrogen®).

The cells are cultured in the same conditions. The proliferation medium is that described in example 1 (optimized proliferation medium) for 15 days before induction of differentiation (DO) and the differentiation medium is the optimized medium described in example 3.

Adiponectin synthesis is followed by ELISA colorimetric assay (R&D System) and the presence of adipocytes is determined with Oil Red O.

Results:

Table 1 1

^* DO being the moment of induction of differentiation

These results are presented in Fig. 5.

Conclusion: The population obtained from preadipocytes expressing Pref-1 according to the invention (PAA07085) is capable of synthesizing adiponectin more quickly than one and the same population obtained from stem cells (ADSC). Moreover, the cells obtained according to the method of the invention have, at D14, larger vacuoles than the cells obtained from ADSC.

This signifies that selection of preadipocytes expressing Pref-1 makes it possible to shorten the duration of the differentiation phase and more quickly generate a population of adipocytes that are functional and therefore of better quality.

It was also observed that the rate of proliferation is greater for the population obtained according to the method of the invention than the population derived from ADSC that has a lower growth rate. It was in fact noted that at the same seeding rate, the population derived from ADSC reaches confluence on D4 vs D3 for the population derived from preadipocytes according to the invention.

Example 5: Development of a hypoderm according to the invention

In the preceding examples of development of differentiation medium, it was observed that the preadipocytes could organize into layers when they were cultivated in certain media. However, the conditions of preparation of the hypoderm were optimized so as to obtain a cellular population in which preadipocytes and mature adipocytes coexist in a controlled manner.

The protocol followed uses preadipocytes seeded for example in passage 3 at a density of about 40 000 cells/cm ² in plates of 75 cm ² .

The cells are cultivated for 2, 3 or 4 weeks in the proliferation medium supplemented with vitamin C (50 g/ml) as well as for 19 days in a differentiation medium composed for example of 10% by volume of fetal calf serum (FCS), 2 nM of triiodothyronine, 0.15 lU/ml of insulin, 1 μΜ of dexamethasone, 10 nM of hydrocortisone and 0.5 mM of I BMX and without vitamin C. After 3 days of differentiation the IBMX is withdrawn from the medium. The medium is changed every 3 days.

The cells are observed in optical microscopy at 5 days in differentiation medium.

Histological staining and immunohistological labelling HPS staining

The samples of layers were detached and then fixed in 10% formol solution. After dehydration, they are transferred to liquid paraffin at 60°C and embedded in blocks of paraffin. 5 μηη sections are deparaffined and stained with haematoxylin-phloxine-saffron (HPS).

Staining with Oil Red O

7 μηη sections were fixed in isopropanol and then stained with a solution of Oil Red O. The cryosection was then counter-stained with Harris haematoxylin.

Immunohistological analyses

The layers were fixed in 10% formol solution and embedded in paraffin. 5 μηη sections were deparaffined and the antigenic sites were unmasked with an enzyme specific to each type of antibody. The sections were then incubated in a solution of PBS/NGS 3% for one hour. The primary antibody (rabbit polyclonal anti-human collagen I and V antibodies, mouse monoclonal anti-human collagen III and VI) was deposited on slides overnight at room temperature. Rabbit and mouse anti-lgGs coupled to peroxidase were used for detecting the immunocomplexes, using diaminobenzidine as substrate. Counter-staining was carried out with Harris haematoxylin. For the negative controls, the primary antibody was omitted.

Secretion of adiponectin according to the protocol described in example 4.

Western blot of Pref-1 according to the protocol described in example 1. Results:

The results show that regardless of the culture time in the proliferation medium alone, the cells have a fibroblastic appearance. Starting from 1 week of culture, the cells self-assemble into a layer but at this culture time the layer is too thin to be removed easily.

It is observed that culture for 14 days in the proliferation medium is optimum, in that a longer culture time in the proliferation medium does not give better results for thickness and differentiation of preadipocytes into mature adipocytes. Combined with culture for 19 days in the differentiation medium, i.e. 33 days total culture time, it is the condition that makes it possible to obtain the largest amount of mature adipocytes.

Moreover, the mature adipocytes appear very early, since at 3 days after the change to the differentiation medium, the cells have small lipid vacuoles, which increase in size throughout the culture.

At 28 days the total thickness of the hypoderm is about 70 μηη.

Moreover, the presence of the components of the extracellular matrix ECM, in particular the gradual appearance of different types of collagen was observed, was demonstrated within the reconstructed hypoderm thus proving its quality and functionality.

- Secretion of adiponectin by the layers according to the culture time in the proliferation medium and in the differentiation medium (ng/ml/48h)

Table 12

Medium Culture time before measAdiponectin in

urement (in days) ng/ml/48h

Experiment 1 Proliferation 14 0

Differentiation 7 1 .9

14 8

Experiment 2 Proliferation 21 1 .9

Differentiation 7 2.5

14 2.7

Experiment 3 Proliferation 28 1 .9

Differentiation 7 1 .9

14 2.3 The assays of adiponectin show that the cells' capacity to differentiate is optimum after a culture time in the proliferation medium of 14 days, since a longer culture time in the proliferation medium did not give greater secretion of adiponectin during differentiation.

This therefore confirms that the optimum duration of the proliferation phase in the proliferation medium is 14 days for obtaining good differentiation.

Analysis of Pref-1 by Western blot (Fig. 6) shows that regardless of the layer, the specific band of Pref-1 is also present in the initial population of preadipocytes.

The results from Western blot confirmed the presence of preadipocytes, characterized by the presence of Pref-1 within all the layers and demonstrates that:

- expression of the protein Pref-1 increases with the proliferation time of the cells (Fig. 6A - obtained at 2 weeks, 3 weeks, 4 weeks)

- expression of the protein Pref-1 decreases after 16 days of differentiation (Fig. 6B - obtained at 4 weeks, 4 weeks+ 16 days). Example 6: Investigation of the effect of active ingredients on cellular proliferation and clonogenicity

Cells derived from 3 strains from different donors are seeded for example at 7000 cells per cm ² in a 6-well plate.

The active substance tested is a hydrolysate of kappa-carrageenan corresponding to the product marketed under the name Slim-excess™ by BASF BEAUTY CARE SOLUTIONS FRANCE and described in patent application WO2009000935.

The active substance is added starting from the first change of culture medium, which corresponds to the second day of culture. The preadipocytes are cultivated up to confluence, i.e. 7 days of culture, and are detached with trypsin/EDTA. After centhfugation, the cells are counted on Malassez cells and population doubling (PD) and the growth rate (GR) are calculated (formula described in example 1 ).

Conditions Standard

Mean

deviation

Untreated control 2.70 0.33

Active substance 0.2% 1.71 0.27

Active substance 1 % 0.167 0.07

Table 13: Number of millions of cells after 7 days of culture of preadipocytes in the proliferation medium

Standard

Conditions Mean

deviation

Untreated control 0.876 0.03

Active substance 0.2% 0.766 0.04

Active substance 1 % 0.190 0.10

Table 14: Evaluation of the growth rate (GR) of the preadipocytes after 7 days of proliferation of the preadipocytes and evaluation of the growth rate The results show that the active substance evaluated (at 0.2% and at 1 %) decreases the proliferation of the preadipocytes significantly relative to the negative control (Student test p<0.05).

The active substance therefore limits the proliferation of the preadipocytes without cytotoxicity.

Example 7: Investigation of effects of active substances on a hypoderm

The hypoderm is prepared according to the protocol described in example 5 (optimized protocol).

The cells were cultivated for 14 days in the previous proliferation medium described in example 5 then for 14 days in the differentiation medium previously described in example 5, the 2 media being supplemented with vitamin C at 50 μg/ml.

The ingredients tested, namely the active substance and caffeine, were added at various times (2, 14 and 21 days) and until the end of the experiment. The active substance was tested at two concentrations (0.2% and 1 %) versus the positive control (caffeine at 1 mM) and the negative control (neither active substance nor caffeine) on a population according to the invention.

For each condition (the active substance or caffeine), the model developed allows the ingredients to be added at three different times:

- After 2 days of culture in the proliferation medium and until the end of the experiment to evaluate any effect on the proliferation of preadipocytes.

- After 14 days of culture, i.e. when differentiation of the preadipocytes begins and up to the end of the experiment, in order to evaluate any preventive effect on the d ifferentiation of preadipocytes.

- After 21 days when the adipocytes have partly matured and up to the end of the experiment, in order to demonstrate any preventive effect potentially associated with any healing action since the model contains both preadipocytes and mature adipocytes.

At the end of the experiments, the hypoderms were investigated histologically, by staining with HPS and Oil Red O and with immunolabelling of collagen I, III, V according to the protocols described in example 5.

Histological examination (Fig. 7) permitted measurement of the thickness of each of the hypoderms and the immunolabelling for characterization of the presence of the collagen extracellular matrix for the conditions "untreated", "caffeine" and "active substance" at concentration of 0.2% on several cellular strains. In figure 7, first line results are control, second line results are obtained with caffeine and third line with active at 0.2%. 1 ^st column results obtained when ingredient was added after 2 days and 2 ^nd column when ingredient was added after 14 days. Staining H PS was performed after 28 days total culture and scale is 50 micrometers.

It is clear that the model has a thickness of more than 150 μηη after 28 days of culture

(untreated control conditions). It also appears that the model makes it possible to demonstrate a decrease in thickness of the layers treated with the active substance or caffeine regardless of the time of addition of the ingredients.

In these experiments, the model used was able to demonstrate that the two ingredients tested: the active substance and caffeine, have an effect on adipocyte proliferation and differentiation since:

when the ingredients are contacted in the proliferation medium they inhibit only the proliferation of the preadipocytes here;

when the ingredients are added after the proliferation phase, the active substances induce a decrease in differentiation since the final thickness of the hypoderms is less than that of the untreated hypoderms.

This result is confirmed by staining with Oil Red O, which reveals the presence of smaller adipocytes in the cases of treatment with the active substance or caffeine than in the cases of the control hypoderm.

A decrease in thickness of the hypoderm by more than 50% was observed, regardless of the time of addition of the caffeine and the active substance.

Moreover, it is observed that the components of the extracellular matrix ECM are maintained in the case of the active substance, since the labelling of collagen I is of the same intensity as the negative control.

Example 8: Selection of modulators of adipocyte differentiation by evaluation of Pref-1 in quantitative RT-PCR

Expression of Pref-1 and of actin was analysed by real-time RT-PCR (quantitative reverse transcriptase polymerase chain reaction).

Briefly, the preadipocytes are cultivated for 24 hours in the proliferation medium and their differentiation is induced or not for 24h at least in the differentiation medium. The media are removed, the cells lawns are rinsed and the total RNAs are purified with the "SV 96Total RNA Isolation System" kit (Promega, Charbonnieres, France). The purified RNAs are eluted in 100 μΙ_ of water (Promega, Charbonnieres, France), assayed and distributed in plates (96-well, 10 ml. of total RNAs at 5 ng/mL). The primers selected for performing this work are as follows:

Table 11 : Primers of the marker Pref-1 Real-time RT-PCR is performed with the "Quanti Techt SYBR Green RT-PCR" kit (Qiagen, France) on the wells containing mRNA, in an OPTICON thermocycler, which performs the proliferation cycles. Reverse transcription (RT) is carried out for 30 minutes at 50°C, followed by 15 minutes at 95°C to inhibit the reverse transcriptase, activate the polymerase and denature the complementary DNA (cDNA) obtained. 50 cycles of polymerase chain reaction (PCR) are carried out (15 seconds at 95°C, 30 seconds at 60.9°C, 30 seconds at 72°C). At the end of each cycle, the fluorescence, which is proportional to the number of amplified fragments, is read. The level of expression is defined as the ratio of expression of each gene relative to actin.

I n th is model , it was demonstrated that Pref-1 is weakly expressed i n the preadipocytes in a monolayer but disappears after differentiation. The active principles are selected when the degree of expression is increased or decreased by 20%, i.e. a factor of 1.2 or 0.8 times the degree of expression in the case of the untreated control after normalization with actin.

The active principles stimulating adipogenesis are selected when the degree of expression of Pref-1 is decreased by 20%, i.e. a factor of 0.8 times the degree of expression in the case of the untreated control after normalization with actin.

The active principles that cause adipogenesis to decrease are selected when the degree of expression of Pref-1 is stable or is increased by 20%, i.e. a factor of 0.8 times the degree of expression in the case of the untreated control after normalization with actin.