PROCESS FOR PREPARING BIOCOMPATIBLE AND BIODEGRADABLE POROUS THREE-DIMENSIONAL POLYMER MATRICES AND USES THEREOF

Abstract

The present invention relates to a process for preparing a biocompatible and biodegradable, porous three-dimensional polymer matrix, to the porous polymer matrix obtained by means of such a process, and also to the uses thereof, in particular as a support and for cell culture or in regenerative medicine, and in particular for cell therapy, in particular cardiac cell therapy.

Claims

1. Process for preparing a biocompatible and biodegradable polymer matrix comprising a network of open and interconnected pores, said process comprising at least the following steps: 1) preparing an aqueous solution comprising at least one biocompatible anionic polysaccharide and at least one biocompatible cationic polymer, 2) mechanically stirring said solution obtained above in the preceding step, in the presence of a foaming agent or of a pressurized gas, so as to form a foam, 3) freezing the foam obtained above in the preceding step, so as to obtain a frozen foam, 4) gelling the frozen foam obtained above in the preceding step, by adding, to said foam, at least one gelling agent in solution in a solvent, so as to obtain a gelled foam, 5) dehydrating the gelled foam obtained above in the preceding step, so as to obtain a dehydrated gelled foam, then 6) drying the dehydrated gelled foam obtained above in the preceding step, by treatment with supercritical CO.sub.2, so as to obtain said polymer matrix.

2. Process according to claim 1, wherein the biocompatible anionic polysaccharide(s) have an average molecular weight (MwA) of greater than or equal to 75 000 Daltons.

3. Process according to claim 1, wherein the biocompatible anionic polysaccharide(s) are chosen from alginates and modified alginates.

4. Process according to claim 1, wherein the amount of anionic polysaccharides present in the aqueous solution of step 1) ranges from 0.5% to 8% by weight relative to the total weight of said aqueous solution.

5. Process according to claim 1, wherein the biocompatible cationic polymer(s) have an average molecular weight (Mw.sub.C) of greater than or equal to 100 000 Daltons.

6. Process according to claim 5, wherein the cationic polymers are chosen from chitosan, particular saline forms of chitosan and chitosan derivatives.

7. Process according to claim 1, wherein the amount of cationic polymers present in the aqueous solution of step 1) ranges from 0.5% to 15% by weight relative to the total weight of said aqueous solution.

8. Process according to claim, the weight ratio of anionic polysaccharides (W.sub.AP)/cationic polymers (W.sub.CP) present in the aqueous solution of step 1) ranges from 20/80 to 80/20.

9. Process according to claim 1, wherein the aqueous solution prepared during step 1) also comprises at least one hydrophilic surfactant.

10. Process according to claim 1, wherein the hydrophilic surfactant(s) represent from 0.01% to 5% by weight relative to the total weight of the aqueous solution of step 1).

11. Process according to claim 1, wherein the foam is formed in the presence of a foaming agent which is therefore added to the solution prepared in step 1) just before carrying out step 2).

12. Process according to claim 1, wherein, during step 2), the foam is formed by introducing a pressurized gas into the aqueous solution prepared in step 1).

13. Process according to claim 1, wherein the freezing step 3) is followed by a step 3a) of lyophilizing the foam obtained at the end of step 3).

14. Process according to claim 1, wherein the gelling agent used during step 4) of gelling the frozen foam is a solution of at least one salt of a divalent or trivalent cation in a solvent.

15. Process according to claim 1, wherein step 6) of drying, with supercritical CO.sub.2, the dehydrated foam obtained at the end of step 5) is carried out at a temperature ranging from 35 to 50 C., and at a pressure ranging from 45 to 95 bar.

16. Biocompatible and biodegradable polymer matrix obtained by carrying out the process as defined in claim 1, wherein said matrix is in the form of a honeycombed material constituted of a porous polymer matrix resulting from the gelling of a foam of at least one biocompatible anionic polysaccharide and of at least one biocompatible cationic polymer, and in that said matrix: comprises open and interconnected pores having an average dimension d.sub.A ranging from 0.2 m to 400 m; has a pore volume ranging from 60% to 98% of the total volume of the matrix; has an elastic modulus at 50% deformation (E.sub.50%) ranging from 1 to 100 kPa.

17. Matrix according to claim 16, wherein said matrix has in the rehydrated state a tensile Young's modulus ranging from 0.3 to 20 kPa.

18. A biocompatible and biodegradable polymer matrix as defined in claim 16, wherein said matrix is configured as a support for animal or human cells and/or for the culture of animal or human cells in vitro.

19. The matrix according to claim 18, wherein said cells are undifferentiated mammalian cells.

20. Cell support comprising a porous polymer matrix containing animal cells, wherein the porous polymer matrix as defined in claim 16, and in that said cells are predominantly present in the pores of said matrix.

21. Cell support according to claim 20, wherein said cell support is configured for use in regenerative medicine.

22. Cell support according to claim 20, wherein said cell support is configured for use in cell therapy in particular in cardiac cell therapy.

Description

EXAMPLES

[0098] The starting materials used in the examples which follow are listed below:

[0099] Sodium alginate from brown algae, of average molecular weight between 80 000 and 120 000 Da, and of viscosity2000 mPa-s (at 2% by weight in water, and at 25 C.), sold under the trade name Alginic acid sodium salt from brown algaeMedium viscosity by the company Sigma-Aldrich;

[0100] Deacetylated chitosan at 80%, of average molecular weight between 190 000 and 300 000 Da, of viscosity between 200 and 800 mPa-s (at 1% by weight in 1% acetic acid and at 25 C.), sold under the trade name Chitosan medium molecular weight by the company Aldrich;

[0101] NaCl and calcium carbonate (the company BDH Prolabo);

[0102] pure acetic acid (the company Fisher Chemical);

[0103] absolute ethanol, HEPES buffer (the company Sigma-Aldrich).

[0104] Physicochemical Characterizations

[0105] Scanning Electron Microscopy (SEM):

[0106] The polymer matrices prepared in the examples which follow were metallized with silver by argon sputtering using a machine sold under the trade name Sputter Coater S 150B by the company Edwards, then observed using a JSM-6400 scanning electron microscope from the company Jeol under a voltage of 10 kV. For each sample, 10 measurements of the pore diameter were carried out at the surface and in cross section and then the mean was calculated.

[0107] Environmental Scanning Electron Microscopy:

[0108] Cubic samples of approximate dimensions 552.5 mm were cut out from the polymer matrices prepared in the examples which follow and were introduced in the dry state into the chamber of a Quanta 250 FEG ESEM microscope from the company FEI. The samples were gradually hydrated therein by controlling the level of humidity in the chamber, which gradually increases from 85% to 99%, by adjusting the water vapour pressure. The acceleration voltage was adjusted to 15 kV and the temperature to 2 C.

[0109] Mechanical Strength (Uniaxial Compression Test):

[0110] Uniaxial compression tests were carried out on samples of the polymer matrices prepared in the examples which follow, after hydration for 24 hours in a cell culture medium. Each sample was subjected to 3 uniaxial compression tests and the measurements were carried out in triplicate. The measurements were carried out using a texturometer sold under the trade name TA-XT2 Texture Analyser by the company Stable Micro Systems, with a cylindrical aluminium piston which has a diameter of 20 mm and a compression speed of 2 mm/s for measuring the force required to compress the samples to 50% of their initial height. The elastic modulus at 50% deformation was then calculated by applying the following formula:

[00002]$E_{50.Math.\%}=\frac{\left(F_{50.Math.\%}/S\right)}{0.5}1000$

[0111] in which E.sub.50% and F.sub.50% are respectively the elastic modulus (in kPa) and the force (in N) required to obtain 50% deformation, and S is the surface area of the sample (in mm.sup.2) in contact with the piston.

[0112] Mechanical Strength (Uniaxial Tensile Test):

[0113] In the examples which follow, tensile tests were carried out on the dumbbell-shaped samples of the polymer matrices (according to standard ASTM D638-10 (Type I)), after rehydration for 24 hours in the cell culture medium. The Young's modulus of the hydrated samples was determined in uniaxial tensile testing using a texturometer sold under the trade name TA-XT2 Texture Analyser by the company Stable Micro Systems, with a constant speed of 0.5 mm/s for measuring the force required until break. The curve of the strain (equal to the force in Newtons related to the surface area in mm.sup.2) as a function of the strain (as %) is plotted and the Young's modulus is then calculated as being the slope at the origin in the linear part of this curve. The measurements are carried out on 3 to 5 samples for each alginate/chitosan ratio. Such a method is described for example in the article by Andersen et al., 2012, Biomacromolecules.

[0114] Stem Cell Cultures:

[0115] Cultures of rat bone marrow mesenchymal stem cells (rMSCs) were carried out in the following way:

[0116] A cell culture medium, complete alpha Minimum Essential Medium (MEM) (complete MEM), was prepared by mixing 450 ml of GlutaMAX a MEM medium sold under the reference 32561 by Gibco, 5 ml of a penicillin/streptomycin mixture (10 000 U/ml) sold under the reference 15140 by Gibco and 50 ml of foetal calf serum sold under the reference A15-043 by PAA.

[0117] The rMSCs were thawed in 15 ml of complete MEM culture medium preheated to 37 C. After centrifugation for 5 min at 1200 rpm, the supernatant was suctioned off and then the cell pellet was taken up in 25 of complete MEM culture medium. The cells were then seeded in a culture flask at a density of 10 000 cells/cm.sup.2. The complete MEM culture medium was changed every 2 to 3 days. The cells were passaged at confluence and re-seeded at a density of 10 000 cells/cm.sup.2.

[0118] Evaluation of the Biocompatibility In Vitro:

[0119] The rMSCs are washed twice with 1PBS buffer, then detached with trypsin and counted. They were then centrifuged for 5 min at 1200 rpm, and then the cell pellet was taken up in complete MEM culture medium. 15 l of cell suspension containing 100 000 rMSCs were deposited on the samples of polymer matrices in the dry state in a 48-well plate, centrifuged for 1 min at 400 g and at 25 C. in order to obtain uniform seeding in terms of depth and, finally, hydrated in complete MEM culture medium. The plates were maintained under conditions of culture at 37 C. in a 5% CO.sub.2 atmosphere.

[0120] Evaluation of the Cell Viability in the Polymer Matrices:

[0121] Physiological saline: solution of NaCl at 0.9% by weight in deionized water,

[0122] Fluorescent labels: viability and cytotoxicity assay kit using calcein AM and Ethidium-III, sold under the name Viability/Cytotoxicity Assay Kit for Live & Dead Cells, reference FP-BF4710, by the company Interchim Fluo Probes, France.

[0123] A cell-labelling solution was prepared just before use by diluting the labels to 1/10 in a 1/1 (v/v) physiological saline/ MEM culture medium mixture so as to obtain a solution of labels containing 2 M of Ethidium-III and 1 M of calcein AM. The labelling solution was kept in the dark until use.

[0124] The rMSCs were washed once with a 1/1 (v/v) physiological saline/ MEM culture medium mixture and then incubated with the labelling solution for 30 min at 37 C. in the dark. After incubation, the cells were washed with physiological saline and stored in physiological saline at 37 C. until observation.

[0125] The observation of the labelling was carried out using a Zeiss 780 confocal microscope: calcein AM excitation wavelength: 495 nm; calcein AM emission wavelength: 515 nm; Ethidium-III excitation wavelength: 495 nm, Ethidium-III emission wavelength: 635 nm. Successive deep images of the matrices were acquired, then a 3-dimensional reconstruction was obtained using the software associated with the microscope.

[0126] Evaluation of the Biocompatibility In Vivo:

[0127] The in vivo biocompatibility of the polymer matrices was evaluated on 3 female rats of the Lewis strain having an average weight of 200 g. For the anaesthesia, the animal was placed in a gas induction box and received a gas mixture of O.sub.2+isoflurane at 4%. After total loss of the peripheral reflexes, the animal was placed on its back. The gas anaesthesia was maintained at 2% or 3% of isoflurane. A subcutaneous injection of buprenorphine (100 g/kg) was given. The abdomen was extensively shaved, and then disinfected with 70 alcohol. After having verified the depth of the anaesthesia and the total loss of peripheral reflexes, a cutaneous incision was made in order to expose the pectoral muscles. A flexible spreader was inserted. The polymer matrix was slipped between 2 muscle planes. The muscle pocket was then closed with a suture using a Prolene 7/0 single thread. The spreader was removed and peritoneal cleaning was carried out. The cutaneous plane was closed again using the Ethilon 5/0 skin thread. The entire procedure was carried out by a surgeon under an operating microscope (Zeiss OPM1 FC).

[0128] Evaluation of the Angiogenic Effects In Vivo:

[0129] The in vivo angiogenic effects of the polymer matrices were evaluated on rats after intramuscular implantation of the matrices. The implantation of the matrices was carried out according to the same protocol as that used above for the evaluation of the biocompatibility in vivo. The formation of capillary vessels and of more mature vessels (arterioles) was studied after 28 days of implantation, by immunofluorescence using antibodies directed against Von Willebrand factor (VWF) making it possible to detect the endothelial cells and against smooth muscle alpha actin (-SMA) making it possible to detect the muscles of the arterioles according to the protocols below:

[0130] HistologyImmunolabellings:

[0131] Twenty-eight days after their implantation, the matrices were removed, rinsed with physiological saline and immediately fixed with 4% paraformaldehyde (in 1PBS, pH 7.4) for 48 h, then transferred into 70% ethanol After embedding in paraffin, histological sections 4 to 6 m thick were cut on a microtome. Hematoxylin-eosin histological staining and anti--SMA (smooth muscle actin) and anti-VWF (Von Willebrand Factor) immunofluorescent labellings were carried out. For the immunofluorescence labellings, the sections on slides were first deparaffinized in xylene (3 baths of 5 min), then rehydrated in successive baths of ethanol (each 5 min) and finally in water (5 min) The antigenic sites were unmasked in a (10 mM) Tris-(1 mM) EDTA buffer containing 0.05% of Tween 20 at 121 for 3 min. The samples were then permeabilized with Triton (0.5%) and the unreacted aldehyde functions of the fixer were neutralized in a 0.1 M glycine solution (2 baths each of 10 min) The non-specific antigenic sites were saturated with a PBS buffer solution containing 2% of goat serum, 1% of bovine serum albumin and 0.2% of Triton (30 min). The slides were then labelled with an anti--SMA antibody (mouse anti-alpha-SMA monoclonal, A2547, Sigma, 1/1000.sup.th dilution) and an anti-VWF antibody (rabbit anti-human VW Factor polyclonal, A0082, Dako, 1/200.sup.th dilution) diluted in the saturation buffer solution. After three washes (PBS-0.2% Tween 20, each for 10 min), secondary antibodies were added: Alexa Fluor 568 goat anti-mouse (A11019, Life Technologies) and AlexaFluor 488 goat anti-rabbit (A11008, Life Technologies) for 30 min in the dark. After 3 washes, the nuclei were stained with DAPI (D9542, Sigma, dilution 0.05 g/ml in PBS, 10 min) Finally, the slides were washed and mounted with a coverslip using a mounting solution for fluorescence (F4680, Sigma). All the steps were carried out is at ambient temperature.

[0132] Observation of the Immunolabellings by Confocal Microscopy

[0133] The immunolabellings were observed using a Zeiss LSM 780 confocal microscope (Carl Zeiss Microscopy) at the 63 magnification. For each animal, the number of vessels positive for -SMA of which the lumen is closed and of which the diameter is greater than or equal to 5 m was counted in the area of the implant on at least 5 non-overlapping photos. Knowing the total surface area of the optical field (in mm.sup.2), the density of arterioles was calculated as being equal to the number of vessels/mm.sup.2

[0134] Statistics Regarding the Immunolabellings

[0135] For the comparison of the number of vessels per unit of surface area between the implanted groups (L+ or reference matrices, acellular or containing MSCs), the two-sided Student's t test was used. The statistical analysis was carried out with the software sold under the trade name GraphPadPrism version 4 (PrismGraphPad, San Diego, Calif.). The Gaussian distribution of the data was tested with a normality test and the results were expressed by their meanstandard error of the mean. A test is considered to be significant if the p-value is less than 0.05.

Example 1

[0136] Preparation of Porous Polymer Matrices in Accordance with the Present Invention and of Comparative Porous Polymer Matrices which are Not Part of the InventionCharacterizations

[0137] In this example, porous polymer matrices in accordance with the invention based on alginate as anionic polysaccharide and on chitosan as cationic biocompatible polymer were prepared using various alginate/chitosan weight ratios according to Table 1 below:

TABLE-US-00001 TABLE 1 Matrix Matrix Matrix Matrix Matrix M100/0.sup.(*.sup.) M60/40 50/50 40/60 0/100.sup.(*.sup.) Alginate/chitosan 100/0 60/40 5050 40/60 0/100 weight ratio .sup.(*.sup.)Comparative matrices not part of the invention

[0138] 1) Preparation of the Polymer Matrices

[0139] The following compositions were prepared:

[0140] Alginate solvent (for 200 g): 1.8 g of NaCl, made up to 200 g with deionized water;

[0141] Chitosan solvent (for 100 g): 0.9 g of NaCl+1.5 g of pure acetic acid (i.e. 0.25 M) and made up to 100 g with deionized water;

[0142] Buffer I (for 1000 g): 9.0026 g of NaCl+3.2540 g of HEPES buffer, made up to 1000 g with Milli-Q water and the pH adjusted to 7.4 with 1 M or 2 M hydrochloric acid (HCl);

[0143] Gelling buffer II (for 500 ml): 5 g of calcium chloride (i.e. 0.1 M)+50 g of pure acetic acid and made up to 500 g with Milli-Q water (Merck) then homogenized;

[0144] Pore-forming agent tested: Sodium bicarbonate:NaHCO.sub.3 (the company Sigma-Aldrich)introduced in step 1;

[0145] Surfactant tested: Polysorbate 20 sold under the trade name Montanox 20 DF (the company SEPPIC), introduced in step 1.

[0146] Solutions A, B, C and D, the specifications of which are given in Table 2 below, were then prepared:

TABLE-US-00002 TABLE 2 Solutions A B C D Alginate solvent (g) 200 Alginate powder (g) 6 Chitosan solvent (g) 100 100 100 Chitosan powder (g) 2 3 4.5

[0147] Solutions A, B, C and D were stirred at between 1600 and 1800 rpm for 60 min.

[0148] The foams having the composition indicated in Table 3 below (weight percentages) were then prepared according to the protocol previously described (steps 1 to 6 of the process in accordance with the invention):

TABLE-US-00003 TABLE 3 Alginate/chitosan Foam Foam Foam Foam Foam weight ratio 100/0.sup.(*.sup.) 60/40 50/50 40/60 0/100.sup.(*.sup.) Solution A (g) 50 Solution B (g) 50 50 Solution C (g) 50 Solution D (g) 50 50 50 50 Alginate solvent 50 Chitosan solvent 50 Final % alginate 1.5 1.5 1.5 1.5 0 Final % chitosan 0 1 1.5 2.25 1.5 Final % foaming agent 0.9 0.9 0.9 0.9 0.9 Final % Montanox 20 1 1 1 1 1 .sup.(*.sup.)Comparative matrices which are not part of the invention

[0149] The various ingredients making up the foams were mixed and the resulting compositions thus obtained were stirred at 1800 rpm for 30 minutes.

[0150] Each of the foams thus obtained was then poured into 48-well plates in a proportion of 500 l per well, and then immediately frozen at 20 C.

[0151] After freezing, a part of the foams (called L.sup.+) was lyophilized at a temperature of 50 C. and a pressure of between 10 and 100 m of mercury under vacuum (according to step 3a of the process in accordance with the invention). Another part of the foams frozen (called L.sup.) was not subjected to this lyophilization step.

[0152] The frozen and lyophilized foams L.sup.+ and the frozen foams L.sup. were then gelled by adding 500 l of gelling buffer II to each of the wells, it being understood that, in order to perform the gelling of the 0/100 foam, 500 l of a 1 M NaOH solution was used in place of the gelling buffer II, since it is known that sodium hydroxide used at this concentration causes chitosan to gel.

[0153] At the end of one hour, the plates were rinsed several times using buffer I so as to completely remove the surfactant (3 washes).

[0154] The gelled foams were then dehydrated by immersing the plates in successive baths of increasing concentration of absolute ethanol: 20%, 40% and 80% at a rate of 3 successive immersions for 10 min in each of the baths, the final dehydration having been carried out in a bath of absolute ethanol at 100% at a rate of 3 successive immersions for 15 minutes.

[0155] The gelled and dehydrated foams were then dried with supercritical CO.sub.2. To do this, the foams were removed from the 48-well plates, and placed in sample racks which are placed in the chamber of an E3000 Series Critical Point Dryer apparatus for drying with supercritical CO.sub.2, from the company Quorum is Technologies. The drying with supercritical CO.sub.2 was carried out at a temperature of 44 C. under a pressure of 85 bar for 25 minutes. The depressurization of the chamber was carried out at a rate of 2 bar/min until atmospheric pressure was reached. The chamber was then opened and the expected matrices were recovered.

[0156] 2) Results of the Characterizations

[0157] The macroscopic appearance of the matrices thus obtained is shown in the appended FIG. 1, which represents the microscopic appearance of the internal structure of the dried or hydrated L.sup.+ matrices. The images of the dried matrices are obtained by SEM. The images of the hydrated matrices are obtained by environmental SEM by gradually increasing the level of humidity from 85% to 99%. The scale bar corresponds to 100 m.

[0158] In this figure, FIG. 1a corresponds to the SEM photos of the dry matrices. The photos of the hydrated matrices were obtained by environmental SEM by gradually increasing the level of humidity from 85% (FIG. 1b) to 99% (FIG. 1d), FIG. 1c corresponding to an intermediate hydration state. In FIG. 1, the scale bar corresponds to 100 m.

[0159] These photos show, in all the matrices presented, the presence of interconnected pores which open up under the effect of the rehydration; the microscopic observation makes it possible to visualize the open porosity towards the exterior of the matrices obtained.

[0160] The quantitative evaluation of the porosity of the matrices thus obtained is given by the appended FIG. 2: quantification of the pore size of the L.sup.+ dry matrices (corresponding to the photos in FIG. 1a). The values displayed are in the form meanstandard error of the mean. The value of p<0.05 is considered to be significant. * p<0.05; ** p<0.01; *** p<0.001.

[0161] The surface porosity (pore diameter measured by SEM at the surface) is reported in FIG. 2a (pore diameter in m as a function of the matrices prepared). The porosity in section (transverse diameter of the pores, measured by SEM) is reported in FIG. 2b (transverse diameter of the pores in m as a function of matrices prepared).

[0162] The results show a similarity in porosity between alginate matrices and alginate/chitosan matrices, which overall show themselves to be superior to that of the matrices of chitosan alone, although they are all in the porosity range recognized as being favourable to cell survival and proliferation. The alginate and alginate/chitosan matrices thus appear to be more suitable for 3D-seeding.

[0163] In the appended FIG. 3, the results of compressive strength are given showing the mechanical compressive properties of the L.sup.+ matrices hydrated in cell culture medium, subjected to three successive compression cycles, and also in the appended FIG. 4, which compares the results obtained with the M40/60 L.sup.+ matrix, having undergone the intermediate lyophilization step 3a, to those obtained with the M40/60 L.sup. matrix not having undergone said step. As regards the results presented in FIG. 3, the statistical comparisons are carried out versus M100/0. * p<0.05; *** p<0.001.

[0164] In FIG. 3, the differential elastic modulus (in kPa) is as a function of the nature of the matrices prepared; the white bars correspond to the first compression cycle, the grey bars to the second compression cycle and the black bars to the third compression cycle.

[0165] In FIG. 4, the force (in Newtons, N) is as a function of the deformation (as %); the continuous-line curves correspond to the first compression cycle, the discontinuous-line curves to the second compression cycle and the dotted-line curves to the third compression cycle.

[0166] The results of FIG. 3 show the synergy associated with the interaction between the alginate and the chitosan, which results in matrices with optimized mechanical properties compared with the matrices based on alginate alone (M100/0) or on chitosan alone (M0/100). The method of drying with supercritical CO.sub.2 according to step 6 of the process in accordance with the invention made it possible to obtain aerogels which retain the porosity of the foam generated in step 2) then during the gelling in step 4), and the mechanical properties associated with the formation of complexes of polyelectrolytes of opposite charges. The matrix in accordance with the invention in which the alginate/chitosan ratio is 50/50 (matrix M50/50) exhibits the best mechanical strength properties. Conversely, the matrices not in accordance with the invention (matrices M0/100 and M100/0) exhibit weaker mechanical properties which may be an impairment to their use for implantation.

[0167] The results presented in FIG. 4 show that the lyophilization step 3a, although it is optional, makes it possible to improve the mechanical properties of the matrices in accordance with the invention, the compression resistance being greater when the matrix has undergone said step.

[0168] The results of the assays for viability of the rMSCs in the matrices M0/100, M40/60 and M100/0 after 7 days of culture are reported in Table 4 below:

TABLE-US-00004 TABLE 4 Alginate/chitosan ratio L.sup. matrices L.sup.+ matrices M100/0.sup.(*.sup.) + + M40/60 ++ ++ M0/100.sup.(*.sup.) ++ + .sup.(*.sup.)Comparative matrices which are not part of the invention

[0169] In Table 4, the + signs relate to the presence of live cells (cells which appear green when observed with a confocal microscope because they are stained with calcein AM). The number of + relates to the number of detectable live cells. All the matrices contain live cells after 7 days of culture, which demonstrates the biocompatibility of the matrices in accordance with the invention.

[0170] Finally, it emerges from the tests for evaluating the biocompatibility after implantation in rats (evaluation 1 week after intra-muscular implantation at the pectoral level) that:

[0171] the matrices obtained by means of the process in accordance with the invention lend themselves well to surgical manipulation and to implantation without this damaging them;

[0172] the implantation of the matrices does not cause the animals to experience any impairment in terms of moving, feeding, etc.;

[0173] the implantation does not cause any massive inflammatory reaction or any other physiological reaction that might endanger the health of the animals having undergone implantation;

[0174] one week after their implantation, the matrices are kept in place (at the site of implantation) and retain their integrity (no matrix debris).

Example 2

[0175] Study of the Behaviour with Respect to Rehydration of Two Matrices in Accordance with the Present Invention

[0176] In this example, the behaviour with respect to rehydration of the M40/60 matrix in accordance with the invention which had undergone an intermediate lyophilization step 3a, as prepared according to the process described above in Example 1 (M40/60 L.sup.+ matrix), was compared with that of the M40/60 matrix in accordance with the invention but which had not undergone this intermediate lyophilization step 3a, as also prepared according to the process described above in Example 1 (M40/60 L.sup. matrix).

[0177] To do this, samples of identical diameter of each of these two matrices were immersed in water for 5 min. Photos of each of the matrices taken before and after immersion are given in the appended FIG. 5.

[0178] It can be observed that the M40/60 L.sup.+ matrix having undergone the intermediate lyophilization step 3a of the process in accordance with the invention swells faster in water and reaches its definitive swelling level (maximum hydration state) in less than 5 min, whereas the M40/60 L.sup. matrix not having undergone said step swells more slowly and does not reach its maximum hydration level within this period. Consequently, these results show that, although it is optional, this intermediate lyophilization step makes it possible to improve the rehydration properties of the matrices in accordance with the invention.

Example 3

[0179] Evaluation of the Angiogenic Effects of a Matrix in Accordance with the Invention Compared with a Matrix Not in Accordance with the Present Invention

[0180] In this example, the angiogenic effects of the M40/60 L.sup.+ matrix in accordance with the invention and as prepared above in Example 1 were evaluated in comparison with those of a matrix not in accordance with the invention, obtained according to a preparation process identical in all respects to that of the M40/60L.sup.+ matrix, except that the last two steps of dehydration and drying with supercritical CO.sub.2 were replaced with a further freezing step and then a further lyophilization step, said steps being carried out under the same conditions as the steps for freezing the foam and for lyophilizing the frozen foam, described above in Example 1. Said matrix, which is therefore doubly lyophilized, is called M40/60 REF.

[0181] Each of these two matrices was tested after implantation without multiplication in rats (M40/60 L.sup.+ matrix in accordance with the invention and M40/60 REF matrix not in accordance with the invention), and also after prior seeding with 500 000 rMSCs (M40/60 L.sup.+-MSC matrix in accordance with the invention and M40/60 REF-MSC matrix not in accordance with the invention). Each matrix was tested on 2 rats.

[0182] The evaluations were carried out after 28 days of implantation.

[0183] The appended FIG. 6 gives the number of -SMA-positive vessels per mm.sup.2 after 28 days of implantation for each of the matrices tested.

[0184] Fluorescence microscope observations (not represented) showed that the M40/60 L.sup.+ matrix implanted is richly vascularized and that the vessels are distributed throughout the whole granular tissue which forms following biodegradation thereof The vessels present in the matrices are functional (because they contain red blood cells). In the presence of these rMSCs seeded in the matrix, the vascularization of the latter is greater (23744 vessels per mm.sup.2 in the M40/60 L.sup.+ group compared with 391127 vessels per mm.sup.2 in the 40/60 L.sup.+-MSC group). In the case of the M40/60 REF matrix not in accordance with the invention, the same trend is observed in the presence of MSCs (16611 vessels per mm.sup.2 compared with 19921 vessels per mm.sup.2) However, it is noted that the vascularization of the M40/60 REF matrix is significantly less abundant than that of the M40/60 L.sup.+ matrix dried with supercritical CO.sub.2 in accordance with the present invention.

[0185] These tests demonstrate that the choice of the final step of drying with supercritical CO.sub.2 is not a simple alternative to a lyophilization step, but, on the contrary, this drying method influences the properties of the resulting matrix, in particular its angiogenic properties after implantation.