BIOMATERIALS COMPRISING A SCAFFOLD CONTAINING A MINERAL COMPOUND, AND USES THEREOF AS BONE SUBSTITUTES

Abstract

The present invention concerns a biomaterial comprising a scaffold containing a mineral component, wherein said mineral component comprises at least one calcium phosphate compound, and wherein said scaffold has a surface coated with an interrupted coating made of multilayered droplets, said multilayered droplets being droplets composed of at least one layer pair consisting of a layer of polyanions and a layer of polycations.

Claims

1. A biomaterial comprising a scaffold containing a mineral component, wherein said mineral component comprises at least one calcium phosphate compound, wherein said scaffold has a surface coated with an interrupted coating made of multilayered droplets, said multilayered droplets being droplets composed of at least one layer pair consisting of a layer of polyanions and a layer of polycations.

2. The biomaterial of claim 1, wherein the scaffold further contains a polymeric component.

3. The biomaterial of claim 1, further comprising a therapeutic molecule within at least one multilayered droplet or forming at least one multilayered droplet when said therapeutic molecule is charged.

4. The biomaterial of claim 3, wherein the therapeutic molecule is a growth factor selected from the group consisting of: a vascular endothelial growth factor (VEGF), a bone morphogenetic protein (BMP), a transforming growth factor (TGF), a fibroblast growth factor (FGF), a nucleic acid coding therefor, and mixtures thereof.

5. The biomaterial of claim 1, wherein the calcium phosphate compound is selected from the group consisting of: hydroxyapatite (HA), amorphous calcium phosphate (ACP), monocalcium phosphate anhydrous (MCPA), monocalcium phosphate monohydrate (MCPM), dicalcium phosphate dihydrate (DCPD), dicalcium phosphate anhydrous (DCPA), precipitated or calcium-deficient apatite (CDA), β-tricalcium phosphate (β-TCP), tetracalcium phosphate (TTCP), and mixtures thereof.

6. The biomaterial of claim 2, wherein the polymeric component is made of a polymer chosen from the group consisting of: poly(ε-caprolactone), collagen, fibrin, poly(lactic acid), poly(glycolic acid), poly(ethylene glycol)-terephtalate, poly(butylenes terephtalate), or co-polymers thereof, and mixtures thereof.

7. The biomaterial of claim 1, wherein the polycations are chosen from the group consisting of: poly(lysine) polypeptides (PLL), covalently-coupled cyclodextrin-poly(lysine) (PLL-CDs), poly(arginine) polypeptides, poly(histidine) polypeptides, poly(ornithine) polypeptides, Dendri-Graft Poly-lysines (e.g. Dendri-Graft Poly-L-lysines), chitosan, and mixtures thereof.

8. The biomaterial of claim 1, wherein the polyanions are chosen from the group consisting of: poly(glutamic acid) polypeptides (PGA), poly(aspartic acid) polypeptides, and mixtures thereof.

9. The biomaterial of claim 1, further comprising living cells.

10. A method for preparing the biomaterial of claim 1, said method comprising a step of coating a scaffold containing a mineral component and an optional polymeric component with at least one layer pair consisting of a layer of polyanions and a layer of polycations.

11. The method of claim 10, wherein the step of coating with at least one layer pair comprises the following steps: i immersing the scaffold in a solution comprising the polycations; ii. rinsing the scaffold obtained at the end of step (i); iii. immersing the scaffold obtained at the end of step (ii) in a solution comprising the polyanions; iv. rinsing the scaffold obtained at the end of step (iii); and, optionally; v. repeating step (i) to (iv) for at least a second time; and, optionally; vi. sterilizing the scaffold obtained at the end of step (iv) or (v).

12. A method of using the biomaterial of claim 1 as a bone substitute.

13. The biomaterial according to claim 1 configured for use as a bone and/or cartilage defect filling material, or for use in bone and/or cartilage regeneration.

14. The biomaterial according to claim 1 configured for use in the treatment of a bone and/or cartilage defect.

Description

FIGURES

[0138] FIG. 1. Third-generation bone substitutes nano-functionalized with angiogenic molecules.

[0139] (A) The combination of angiogenic nanoreservoirs, human mesenchymal stem cells and biphasic bone substitutes used in the study.

[0140] (B-D) SEM pictures of VEGF (B), HEP (C) and HEP/VEGF complex (D) after deposition on glass. Scale bar: 500 nm.

[0141] (E) The HEP/VEGF complex shown in (C), representative of the complexes screened (n=6), was also acquired using atomic force microscopy (AFM). False colours indicate depth, as for the side color bar.

[0142] (F) Molecular modelling of the HEP/VEGF complex. VEGF is displayed as homology model (green); HEP is displayed as stick model; yellow indicates sulphur is moieties.

[0143] FIG. 2. Nano-functionalization of the Anatartik® sponge with HEP/VEGF-NRs and its effect on endothelial cells.

[0144] (A-D) SEM pictures of the Antartik® sponge deposited with empty (NF-) (A,B) or HEP/VEGF-(C,D) NRs. The NRs were found on both the mineral (A,C) and the protein (B,D) constituents of the sponge. Scale bar: 6 μm.

[0145] (E) Fluorescence micrographies of GFP-HUVECs on the Antartik® sponge deposited with NF-(top micrographies), HEP/BSA-(middle micrographies) and HEP/VEGF-NRs (lower micrographies) after 21 days of culture (E). Nuclei counterstained with DAPI. Scale bar: 150 μm.

[0146] (F) Number of GFP-HUVECs found organized in vessel-like structures (red asterisk in E) on NF-(right), HEP/BSA-(middle) and HEP/VEGF-NRs (left) bone substitutes. Bars represent mean±SEM (n=20 per condition); ***: p<0.001.

[0147] (G) Reduced Alamar Blue@ was assessed (in %) for the cells seeded on bone substitutes either deposited with NF-(grey bars) or HEP/VEGF-NRs (black bars), after 3, 14 and 21 days of culture. Values expressed as mean±SEM (n=4). *: p<0.1; ***: p<0.01.

[0148] FIG. 3. Host vascular infiltration in bone substitutes subcutaneously implanted in nude mice.

[0149] (A-B) H&E staining of Antartik® biphasic sponge (delimited by the yellow frames), either deposited with empty (NF-) (A) or with HEP/VEGF-(B) NRs, at 12 dpi. White arrowheads indicate blood vessels within the bone substitute.

[0150] (C) Quantitative analysis of the blood vessels found within the implanted bone substitute. The average number (left) and the average diameter (middle) of the blood vessels, together with the average surface covered by the blood vessels (right) found in the explanted bone substitutes are given at 4 (left bars) and 12 dpi (right bars). All values are expressed as mean±SEM of at least 5 images/section, 4 sections/sample; ***: p≤0.01. Scale bars in A, B: 200 μm.

[0151] FIG. 4. Quantitative analyses of the host vasculature infiltration in the bone substitutes implanted in critical size calvarial bone defect mouse models.

[0152] (A) Micro-CT scans of two representative bone substitutes, either deposited with NF-NRs (2NF) or with HEP/VEGF-NRs (2F). All vessels are shown; the vessels within the bone substitutes are colored in red. Scale bar: 1 mm.

[0153] (B) Tridimensional reconstruction of the vascular network found in 4 representative bone substitutes, either deposited with empty NF-NRs (1NF, 2NF) or with HEP/VEGF-NRs (1F, 2F). Skeletons of the segmented images are represented. Lighter colours display vessels of larger size.

[0154] (C) Relative density of the host vascular network within the bone substitutes shown in B, represented as the average distance to the closest neighbor vessel.

[0155] FIG. 5. Human MSCs contributed to the vascularization of the bone substitutes subcutaneously implanted in nude mice.

[0156] (A,B) Ultrastructural view of a blood vessel found within the HEP/VEGF-NRs bone substitute, as transverse section (A). The presence of a mural cell (a pericyte, red asterisks) enveloping the blood vessel suggests the good functionality of the blood vessels that infiltrated the active bone substitute (yellow insert in A; B). No blood cells could be seen as the Microfil® MV-122 contrast agent was perfused, which can be observed in the lumen of the vessel (blue asterisks). E: endothelial cell; eN: nucleus of the endothelial cell; P: pericyte; pN: nucleus of the pericyte. Scale bars: 5 μm in A, 2 μm in B.

[0157] (C) The presence of cells of human origin within the HEP/VEGF-NRs bone substitutes implanted subcutaneously was revealed by immunohistochemistry using an antibody specific for human PECAM1 (C, top panels). The same specimen was also subject to immunohistochemistry using an antibody with cross-reactivity for both human and mouse PECAM1 (C, mid panel). In the lower panel, the anti-human PECAM1 antibody was used on a control mouse bone (C, lower panel). Scale bar: 200 μm.

EXAMPLES

Materials and Methods

[0158] Deposition of the VEGF/HEP Complex

[0159] Drops of HEP (500 μg ml.sup.−1 in 20 mM/0.15 mM Tris/NaCl, pH 6.8), VEGF (200 μg ml.sup.−1 in 20 mM/0.15 mM Tris/NaCl, pH 6.8) or HEP/VEGF (500 μg ml.sup.−1/200 μg ml.sup.−1) solutions (Sigma-Aldrich, Saint-Quentin-Fallavier, France) were laid on cover glass and dried. Salt crystals from buffer were solubilised in deionized water, by 2 rinsing steps of 5 min each. This deposition procedure was repeated 6 times to increase the quantity of material.

[0160] Scanning Electron Microscopy (SEM) and Atomic Force Microscopy (AFM) Study

[0161] To analyze the formation of HEP/VEGF-NRs on the bone substitute, samples were fixed with 4% paraformaldehyde (PFA) for 10 min at 4° C. After dehydration, the specimens were observed by mean of SEM (either with Hitachi TM1000 or FEG Sirion XL; FEI) in conventional high vacuum mode with a secondary electron detector. A commercial stand-alone AFM microscope Solver Pro (Nt-Mdt Inc., Moscow, Russia) was used to acquire AFM images. Tapping imaging mode was used, with NSG 10 cantilever with a typical resonance frequency of 105 kHz and a spring constant of 2 N m.sup.−1. The image resolution was set to 512×512, with a scanning rate of 1 Hz. Images were analyzed using the open source software Gwyddion 2.24 5 (Nec̆as D, Klapetek P. Gwyddion: an open-source software for SPM data analysis. In: Open Physics (ed{circumflex over ( )}(eds) (2012)).

[0162] Nanoreservoirs Deposition on Bone Substitute

[0163] Antartik® sponges (10% collagen I and III, 90% ceramic; Medical Biomat, Vaulx-en-Velin, France) were cut in 5 mm wide fragments and placed in a 96-well plate. They were sterilized with UV light (254 nm, 30 W, distance 20 cm, 30 min exposure). Chitosan-HEP-VEGF-NRs were applied via layer-by-layer deposition, as previously described..sup.[29] Briefly, bone substitutes were alternately dipped in 500 μg ml.sup.−1 chitosan solution (Protasan UP CL 113, Novamatrix, Sandvika, Norway) and 500 μg ml.sup.−1/200 ng ml.sup.−1 HEP/VEGF complex solution in 20 mM/0.15 mM Tris/NaCl, pH 6.8. After each bath, bone substitutes were rinsed three times for 5 min in Tris/NaCl buffer. Before use, bone substitutes were equilibrated in serum-free Dulbecco's modified Eagle's medium (D-MEM).

[0164] Molecular Modelling of HEP-VEGF Complexes

[0165] To prepare the starting structure for simulation, the HEP Sodium salt (CAS 9041-08-1) and VEGF.sub.165 heparin binding domain coordinates (pdb id: 2VGH, 55amino acids) were extracted from the PDB structure files (Fairbrother W J, Champe M A, Christinger H W, Keyt B A, Starovasnik M A. Solution structure of the heparin-binding domain of vascular endothelial growth factor. Structure 6, 637-648 (1998)). Partial atomic charges for heparin sodium salt molecule was assigned based on the AM1-BCC method using the antechamber program of AmberTools. The van der Waals and bonded parameters for the heparin were taken from the general amber force field (GAFF)(Wang J, Wang W, Kollman P A, Case D A. Automatic atom type and bond type perception in molecular mechanical calculations. J Mol Graph Model 25, 247-260 (2006)). The AMBER format files of these molecules were then converted to the GROMACS format using the acpype python script. Using Cluspro and Patchdock docking programs, complexes of heparin and VEGF were generated. The coordinate of 6 best complexes was the selected to generate the topology file for each complex to run molecular dynamics (MD) simulations.

[0166] MD simulations were performed using GROMACS-4.6 for a period of 30 nsec using explicit water model. The complex was placed in the centre of a cubic periodic box and solvated by the addition of SPC water molecules. The net charge on the system was then neutralized by adding counter ions as required. The energy minimization was done using the steepest descent algorithm. The temperature and pressure were maintained at 300 K and 1 atm using the v-rescale temperature and Parrinello-Rahman pressure coupling method. Production simulations were performed for 30 nsec with a 2 fsec time step. In order to calculate the interaction free energies of the heparin and VEGF complexes, we used the MM/PBSA protocol. The calculations were performed using the g_mmpbsa tool82 which implements the MM-PBSA approach using the GROMACS software packages. The MM/PBSA energies were obtained from samples of 100 snapshots (for each complex) that were extracted from the MD trajectories. All the calculations were done using computational time on BWUniCluster Karlsruhe.

[0167] Cell Culture

[0168] Green fluorescent protein-expressing HUVECs (EssenBiotech, France) and hMSCs from the bone marrow (PromoCell, Heidelberg, Germany) were cultured in the respective complete media (endothelial growth medium; mesenchymal stem cells growth medium, PromoCell) at 37° C. in a humidified atmosphere with 5% CO.sub.2. 1.5×10.sup.4 hMSCs were seeded on each Antartik® sponge fragment deposited or not with HEP/VEGF-NRs, and cultured for 7 days in mesenchymal stem cells growth medium. After 7 days, 5.5×10.sup.4 GFP-HUVECs were seeded on the same samples and cultured in a medium consisting of half hMSCs growth medium, and half endothelial growth medium. Bone substitute fragments seeded with cells were cultured for a total of 21 days, and then processed according to downstream applications.

[0169] Cell Biocompatibility Analysis

[0170] Alamar Blue® (Thermo Fischer Scientific, Waltham, Mass., USA) was used to assess cell metabolic activity over time (n=4). At 3, 14 and 21 days after GFP-HUVECs seeding, cells were incubated in 10% Alamar Blue® in D-MEM without phenol red (Lonza, Levallois-Perret, France) at 37° C. and 5% CO.sub.2. After 4 hours, the supernatant was transferred to 96-well plates and the absorbances at 570 and 595 nm were measured in a Multiskan FC plate reader (ThermoFischer Scientific, Illkirch-Graffenstaden, France) to determine the percentage of reduced Alamar Blue®.

[0171] Immunofluorescence

[0172] Bone substitutes seeded with cells were fixed with 4% PFA for 10 min at 4° C. and rinsed with PBS, then demineralized and embedded in paraffin for 4 μm serial sections. Rabbit anti-human PECAM1/CD31 (Microm, Brignais, France) and non species-specific anti-PECAM1/CD31 (Abcam, Paris, France), then secondary antibody raised against rabbit antibodies and coupled with Alexa 594 fluorochrome (Molecular Probes, Life Technologies, ThermoFisher Scientific), diluted 1:200 in PBS 1% BSA, were used. After multiple rinses in PBS, samples were incubated 10 min in 200 nM 4′,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich), for nuclear counterstaining. Specimens were observed with a LEICA DM4000B epifluorescence microscope (Leica Microsystems, Nanterre, France).

[0173] Implant of Bone Substitutes in Nude Mice

[0174] All procedures regarding animals and tissues were designed in agreement with the recommendations of the European Union (2010/63/EU), and were performed by investigators authorized, by the Préfecture du Bas-Rhin. All experiments were performed in the Animalerie Centrale de la Faculté de Médecine de Strasbourg with approval number C-67-482-35 from the Veterinary Public Health Service of the Préfecture du Bas-Rhin, representing the French Ministry of Agriculture, Department of Veterinary Science. Nude male mice (Crl: NIH-Foxn1.sup.nu; Charles River, L'arbresle, France), 6 weeks old, were anesthetized with an intra-peritoneal injection of 100 mg kg.sup.−1 of ketamine (VIRBAC Santé Animale; Centravet, Nancy, France) mixed with 10 mg kg.sup.−1 of Xylazine (Rompun® 2%, VIRBAC Santé Animale; Centravet, Nancy, France) and placed on a heating plate kept at 37° C. Five 5 days before implant, each bone substitute was seeded with 1.0×10.sup.5 hMSCs. For subcutaneous implant, dorsal skin incision was performed and bone substitute (diameter=5 mm) was placed between the skin and the muscle below. Incisions were sutured with resorbable material and mice were kept under observation for the whole experimentation time. After either 4 (n=6) or 12 (n=12) dpi, implanted mice were sacrificed with an intra-peritoneal injection of a lethal dose of ketamine. For implant in bone critical size defect, skin incision was performed, calvaria (parietal zone of the skull) was drilled using a sterile round burr (500 μm deep and 5 mm in diameter) and bone substitute (diameter=5 mm) was placed in the defect. Incisions were sutured with resorbable material and mice were kept under observation for the whole experimentation time. Mice (n=12) were sacrificed 12 dpi and explants were subject to histological, TEM and/or micro-CT assessments.

[0175] Hematoxylin-Eosin (HE) Histological Staining

[0176] For HE staining, subcutaneous explants were fixed with 4% PFA, demineralized and embedded in paraffin for 7 μm serial sectioning. Samples were then subject to HE staining and observed on a Leica DM4000B microscope (Leica Microsystems). For quantitative analysis of the vascularization, imageJ software was used. Number, mean size and total area of the blood vessels found in the explants were analysed. At least, 5 images per section, 4 sections per sample were analyzed.

[0177] Microangiography and Quantitative Analysis of the Vasculature

[0178] Twelve days after subcutaneous implant, mice were subject to deep general anaesthesia (sodium Pentobarbital 120 mg Kg.sup.−1). After opening the thoracic cage, an infusion needle was placed in the left ventricle. Mice were in turn perfused (rate of 2 ml min.sup.−1) with heparin (50 U ml.sup.−1; to purge the cardiovascular system), 4% PFA, PBS and radiopaque silicone rubber (Microfil® MV-122, Flow Tech Inc.). After perfusion, the heart was clamped to avoid leaks of the contrast agent; mice were then placed at 4° C., overnight to allow the polymerization of the contrast agent. Explants were then post-fixed in 4% PFA for 48 hours and then demineralized with Ethylenediaminetetraacetic acid (EDTA, 15%, PH=7,4) for 1 week at 37° C. under constant slow agitation. The explants were mounted in 1% agar, in order to avoid any movement of the sample during micro-CT acquisition. The tomography experiments were carried out using the micro-CT X-ray system nanotom® m (GE Sensing & Inspection Technologies GmbH, Wunstorf, Germany) equipped with a 180 kV—15 W high-power nanofocus tube with a tungsten transmission target. The X-ray micro-CT was so performed with an isotropic pixel size of 5 mm.sup.2 and a field of view of about 15.4×12.0 mm.sup.2. For each measurement, the sample was irradiated by X-rays of 60 kV acceleration voltage and 310 mA beam current. At each rotation angle position six images were acquired and averaged to a projection. 1700 projections over 360° resulted in a total scan duration of about 100 min.

[0179] The data acquisition and reconstruction were performed with the phoenix datos|x 2.0 software (phoenix|x-ray, GE Sensing & Inspection Technologies GmbH, Wunstorf, Germany).

[0180] Grey level images obtained from micro-CT scan were segmented into two classes to distinguish the voxels within the vessels from those outside. For each image, manually tuned threshold was applied to intensity levels. A structuring element 10 μm wide was applied to each image in order to de-noise the segmentation. Then, the connected sets of voxels smaller than 20 μm.sup.3 were erased. After segmentation, the resulting vessels were sketched as skeletons, where a skeleton is defined as the ordered set of the points that defines both the center line and the local radius of a vessel-like shape, as for the algorithm previously described.

[0181] Transmission Electron Microscopy (TEM)

[0182] After the micro-CT scan, explants at 12 dpi were fixed in 2.5% glutaraldehyde and 2.5% PFA in 0.1 M cacodylate buffer, pH 7.4. The samples were post-fixed in 1% osmium tetroxide, dehydrated, conditioned in propylene oxide and embedded in Epon 812, Spurr's epoxy resin (Electron Microscopy Sciences, Ft. Washington, Pa.). Semithin sections (2 μm) and ultrathin sections (70 nm) were prepared on an Leica Ultracut UCT ultra microtome (Leica Microsystems), contrasted with uranyl acetate (Laurylab, Brindas, France) and lead citrate (Euromedex EMS, Souffelweyersheim, France) and examined at 70 kv with a Morgagni 268D electron microscope (FEI-Phillips, ThermoFisher Scientific). Digital images were captured using a Mega View III camera (Soft Imaging System, Munster, Germany).

[0183] Statistical Analysis

[0184] Statistical analysis was carried out with BioStatGV (Sentiweb, France) and Prism5 (GraphPad, La Jolla, Calif., USA). Alamar Blue metabolic assay, histological data and microangiography data were analyzed with unpaired, two-tailed Student's t-test. When variance was found different between sets of data, then the Welch correction was applied.

Example 1

Design and Modelling of a Bone Substitute Nano-Functionalized with Angiogenic Complexed Molecules

[0185] In order to promote the vascularization of a transplanted bone substitute, the inventors functionalized the FDA-approved Antartik® sponge with the NR technology (FIG. 1A)(Mendoza-Palomares C, et al. Smart hybrid materials equipped by nanoreservoirs of therapeutics. ACS Nano 6, 483-490 (2012); Eap S, et al. Collagen implants equipped with ‘fish scale’-like nanoreservoirs of growth factors for bone regeneration. Nanomedicine (Lond) 9, 1253-1261 (2014); Schiavi J, et al. Active implant combining human stem cell microtissues and growth factors for bone-regenerative nanomedicine. Nanomedicine (Lond) 10, 753-763 (2015); and Eap S, et al. A living thick nanofibrous implant bifunctionalized with active growth factor and stem cells for bone regeneration. Int J Nanomedicine 10, 1061-1075 (2015)). Heparin is critical for the angiogenic activity of VEGF, binding the growth factor through the heparin-binding domain located in its carboxy-terminal domain (Fairbrother W J, Champe M A, Christinger H W, Keyt B A, Starovasnik M A. Solution structure of the heparin-binding domain of vascular endothelial growth factor. Structure 6, 637-648 (1998); Krilleke D, Ng Y-Shan E, Shima David T. The heparin-binding domain confers diverse functions of VEGF-A in development and disease: a structure-function study. Biochemical Society Transactions 37, 1201-1206 (2009)). To maximize the angiogenic effects through NR-mediated cell contact-dependant delivery, the inventors first investigated the suitability of the HEP/VEGF.sub.165 complex to functionalize the bone substitute. Heparin, VEGF or a complex thereof was deposited drop-wise on a glass coverslip, and visualized by scanning electron microscopy (FIGS. 1B-D). Complexed HEP/VEGF.sub.165 formed large agglomerates (60.0±15.4 nm), as confirmed by AFM imaging (FIG. 1E).

[0186] To understand why complexed HEP/VEGF aggregates had such a large size, the inventors modelled the molecular interaction between HEP and VEGF.sub.165 using molecular dynamics simulations. Heparin is a linear sulphated polysaccharide, with high negative charge, while the surface electrostatic potential of the HEP-binding domain of VEGF.sub.165 is positively charged (Ono K, Hattori H, Takeshita S, Kurita A,

[0187] Ishihara M. Structural features in heparin that interact with VEGF165 and modulate its biological activity. Glycobiology 9, 705-711 (1999)). Docking simulations of the HEP model to the NMR model of VEGF.sub.165 were therefore performed using the sulphate anions of HEP and the arginine (Arg) residues of VEGF as a guide to is model the molecular interaction. Most of the positive Arg residues are either clustered in the central domain of VEGF.sub.165 (Arg35, Arg39, Arg46, Arg49) or within a loop in its N-terminal domain (Arg13, Arg14). These regions, according to the calculated electrostatic potential of the solvent accessible surface, represent the binding site for a high negatively charged molecule, such as HEP. The HEP/VEGF complex models generated using molecular docking were subjected to steepest decent energy minimization, followed by equilibration and a 30 nsec molecular dynamics production run. The complex models were then ranked according to their energy, by means of the Molecular Mechanics Poisson-Boltzmann Surface Area (MM/PBSA) method. The resulting HEP/VEGF complex structure with the lowest energy (FIG. 1F) clearly indicates that the net charge plays a significant role in the affinity of HEP (stick model in FIG. 1F) to VEGF.sub.165 (homology model in FIG. 1F). The sulphate groups of HEP (yellow in FIG. 1F) strongly interact with the side chains of both the Arg residues mentioned above and the leucine 17 and threonine 47 residues of VEGF.sub.165, stabilizing the HEP/VEGF complex and likely promoting the formation of large agglomerates.

Example 2

Bone Substitutes Nano-Functionalized with HEP/VEGF-NRs Improve the Organization of Endothelial Cells In Vitro

[0188] Heparin/VEGF complexes were integrated into chitosan NRs and deposited on the Antartik® sponge. The functionalized bone substitutes were observed at the SEM and compared to chitosan only-NRs (non-functionalized NRs: NF-NRs). After 6 cycles of deposition, we observed a homogeneous distribution of either HEP/VEGF- or NF-NRs, on both the mineral (FIGS. 2A,C) and the collagen (FIGS. 2B,D) portions of the bone substitute. To investigate the pro-angiogenic effects of the nano-functionalized Antartik® bone substitute, either HEP/VEGF-NRs, HEP/BSA-NRs or NF-NRs were deposited on fragments of bone substitutes that were in turn seeded with hMSCs and green fluorescent protein (GFP)-expressing Human Umbilical Vein Endothelial Cells (GFP-HUVECs). The organization of the endothelial cells was monitored after 21 days of culture. In the presence of HEP/VEGF-NRs, GFP-HUVECs organized in stretches of cells, resembling vessel-like structures (red asterisks in FIG. 2E), which in many cases showed a lumen. On the contrary, in the presence of either NF-NRs or HEP/BSA-NRs (FIG. 2E), the endothelial cells remained mostly distributed as single cells on the graft. The vessel-like structures were assessed by means of number of GFP-HUVECs aligned in a continuous stretch. On nanoactive scaffolds, these were on average 3.4±0.5 cells long, while on non functionalized scaffolds or on scaffolds functionalized with non-angiogenic HEP/BSA, these were 1.2±0.1 and 1.4±0.2 cells long, respectively (p≤0.001) (FIG. 2F). Since the presence of HEP/BSA-NRs produced results similar to those found for NF-NRs, we concluded that the capacity of the nanoactive bone substitute to promote the organization of endothelial cells in vitro depended on the presence and cellular availability of the HEP/VEGF complex. We also assessed the metabolic activity of the cells seeded on non-functionalized scaffolds, and compared it with that of cells seeded in the presence of HEP/VEGF-NRs, by means of alamar blue assay. In both conditions, the reduction of the alamar blue increased from day 0 to day 21 (FIG. 2G), suggesting that no cytotoxic cues affected the cells. No significant differences were observed between NF- and HEP/VEGF-NRs along the culture period considered. However, while the cells cultured with NF-NRs showed an abrupt metabolic increase between day 3 and day 14 (p≤0.1), the cells cultured in the presence of HEP/VEGF-NRs showed a more constant metabolic rate (p≤0.05 and p≤0.01, for day 3-14 and day 14-21, respectively). These data suggest that on non-functionalized scaffolds, the cells may proliferate more than in the presence of HEP/VEGF-NRs.

Example 3

Nanoactive Bone Substitutes Implanted Subcutaneously Recruited Host Vasculature

[0189] Bone substitutes functionalized with HEP/VEGF-NRs improved the organization of endothelial cells in vitro, eliciting the formation of vessel-like structures. Therefore, the inventors assessed if and how the angiogenic nanoactive bone substitute could effectively trigger vasculoneogenesis in vivo. The inventors subcutaneously implanted either HEP/VEGF-NRs or NF-NRs bone substitute in the dorsal of nude mice. Prior to implant, all bone substitutes were seeded with hMSCs and cultured for 7 days. Bone substitutes were explanted from mice after 4 or 12 days post-implant (dpi), and their degree of vascularization was evaluated at the histological levels. In general, explanted nanoactive bone substitutes showed a better recruitment of host blood vessels compared to NF ones (white arrowheads in FIGS. 3A,B). The quantification of the blood vessels found in the core of the bone substitutes showed a significant increase only in the presence of HEP/VEGF-NRs (14.8 ±1.7 mm.sup.−1 and 61.0±12.9 mm.sup.−1 at 4 and 21 dpi, respectively; p=0.002), compared to NF bone substitutes (13.2±2.74 mm.sup.−1 and 22.6±8.0 mm.sup.−1 at 4 and 12 dpi, respectively) (FIG. 3C). No significant differences were observed at 4 dpi in the diameter of the blood vessels recruited (17±1 μm vs. 18±2 μm, in HEP/VEGF-NRs and NF-NRs bone substitutes, respectively) (FIG. 3E) and in the relative vessel area (below 1% in either condition) (FIG. 3E). The picture changed quite dramatically at 12 dpi, when both the size and the relative area of the host blood vessels found in the HEP/VEGF-NRs bone substitutes increased of 1.5-(p=0.002) and 8-fold (p=0.0003), respectively, while no differences were observed in the NF-NRs bone substitutes (FIG. 3E).

[0190] These results show how the presence and prolonged availability of the HEP/VEGF complex thanks to the NR technology increased the number and the size of the blood vessels recruited from the host in the core of the bone substitute, suggesting that a better functionality of the graft could also be expected.

Example 4

Nanoactive Bone Substitutes Promoted Vasculoneogenesis in a Critical Size Calvarial Bone Defect Mouse Model

[0191] Eventually, we tested the effect of the nanoactive HEP/VEGF bone substitute in a mouse model of critical size bone defect. A portion of the skull roof was drilled out in mice (induced calvarial bone defect), and filled with either the HEP/VEGF-NRs or the NF-NRs bone substitute. At 12 dpi, implanted animals were perfused with a radiodense rubber contrast agent and bone substitutes were then explanted and subject to micro-CT scan. 3D micro-angiographies of the transplanted mice (3D rendering of micro-CT scan images) were prepared in order to show the influence elicited by the presence of the nanoactive HEP/VEGF complexes on the recruitment of host vasculature, compared to empty NRs (FIG. 4A). The newly formed vessels penetrated intimately the HEP/VEGF-NRs bone substitutes, covering virtually the entire volume of the implant (FIG. 4B, left panels). On the contrary, the host vasculature colonized the NF-NRs bone substitutes sparsely (FIG. 4B, right panels). Remarkably, when using Standard Euler distances to generate the distance io maps to the closest neighbouring vessel in the segmented images, it was seen that both the trend of the curves (FIG. 4C) and the mode of the distributions (0.32, 0.25 and 0.05 mm for 1NF, 2NF and 3NF, respectively; 0.06, 0.02 and 0.05 mm for 1F, 2F and 3F, respectively) clearly indicated that the average distance of any point to the nearest neighbouring vessel is smaller in the nanoactive bone substitutes compared to the NF one (p=0.0423). The analyses also revealed that the vascular density in NF bone substitutes was 0.75±0.99%, where in HEP/VEGF-NRs bone substitutes was 2.8±0.76% (p=0.0167), which is almost 4 times higher.

[0192] In summary, these data indicated that the nano-functionalization of the bone substitutes resulted in a denser cloud of new vessels formed inside the treated calvarial bone defect, which is an essential precondition for the bone regeneration.

[0193] In presence of a mineral block, the blood vessels are either stopped or cut in pieces, resulting in numerous very small vessels, as shown in FIG. 4. It must be noted that, in an effort to render the vascular network visible, the block was demineralized beforehand, meaning that its mineral elements were eliminated.

[0194] Using the innovative technology of the invention, as illustrated in FIG. 4, the vessels no longer encounter obstacles and are thus able to pursue their paths.

Example 5

Human MSCs Seeded on the Bone Substitutes Contributed to Vasculoneogenesis

[0195] The vessels observed within the bone substitutes in the micro-angiographies look well connected to the surrounding vessels. Also, the implanted animals did not suffer bleeding, and when perfused with the contrast agent, they did not show any leakage of the agent in the surrounding bone, which could point towards a sub-functional vasculature. In order to have a closer look at the morphology of the newly formed vessels, we analyzed them at the ultrastructural level. The vessels found within the bone substitutes were characterized by the presence of tightly connected endothelial cells of normal morphology, surrounded by mural cells that provide structural stability to the vessel (FIGS. 5A and 5B). The functionality of the blood vessels that are found within the bone substitute after implant, together with the speed at which they form are both of crucial importance to avoid the necrosis of the cellular component of the transplanted bone substitutes. Both functionalized and non-functionalized bone substitutes were seeded with hMSCs before implant, in order to regrow the missing bone. Mesenchymal stem cells are multipotent stem cells known to be able to differentiate in many cell types, like osteoblasts, adipocytes, chondrocytes and even neurons. Studies also suggested that MSCs could efficiently differentiate in endothelial cells, at least in vitro. Therefore, the inventors investigated whether the hMSCs seeded on the bone substitute contributed to the formation of the blood vessels found within the implant. By means of immunohistochemistry, the inventors detected several endothelial cells positive to anti-human-PECAM1 antibody (FIG. 5C, top panels), which were part of the endothelium of the newly formed blood vessels within the bone substitute (white asterisks in FIG. 5C). Only a portion of these cells originated from the grafted hMSCs, as shown by the immunohistochemistry using an antibody that cross-reacted with both human and mouse PECAM1 (FIG. 5C, mid panels). As expected, no signal was detected using the anti-human-PECAM1 antibody on a control mouse bone (FIG. 5C, lower panels). These results indicate that the formation of new blood vessels within the implanted bone substitutes depend on the simultaneous presence of both the NRs and the hMSCs. It was therefore concluded that the cell contact-dependent release of angiogenic growth factors over a prolonged time combined to the active differentiation of hMSCs in cells of the endothelium synergistically contribute to the functional vascularization that infiltrate the nano-functionalized bone substitutes.

[0196] The present invention is based on the nano-functionalization of the bone substitute material, which allows cell contact-dependent release of VEGF. In order to maximise the angiogenic effect, VEGF.sub.165 was complexed with HEP. The collected evidences showed that the HEP/VEGF complex formed larger aggregates than VEGF alone (FIG. 1B-E), as a result of the chemical-physical interaction with HEP, as anticipated by computer modelling (FIG. 1F). Nanoreservoir-mediated functionalization of the bone substitute is extremely advantageous, as it effectively overcomes the side effects associated with a high local dose of VEGF. Moreover, since VEGF.sub.165 has a short half-life of approximately 90 min, when exposed to the extracellular environment, within the chitosan NRs it does not undergo degradation, and is therefore available to the cells for a prolonged time. In this work, we showed how HEP/VEGF-NRs induced the organization of endothelial cells in vessel-like structures in vitro (FIGS. 2E,F) and could improve the vascularization of a clinically used biphasic bone substitute, implanted either subcutaneously (FIGS. 3A-C) or in a critical size calvarial bone defect (FIGS. 4A-C). Although the spatial distribution of the host blood vessels in one NF-NRs bone substitutes implanted in the calvarial defect was found similar to that found for HEP/VEGF-NRs bone substitutes, the vascular density was lower (1.9%) compared to nanoactive bone substitutes (2.8%). These results probably owe to the average diameter of the blood vessels and could be explained by a difference in the maturity of the vessels, as observed in some histology specimens.

[0197] Together with HEP/VEGF-NRs, the Antartik® biphasic sponge was seeded with hMSCs, prior implant. Recently, with the emergence of the third generation biomaterials, the use of adult mesenchymal stem cells as therapeutic agents has struck a great interest for clinical applications. These cells are not only able to differentiate into mature tissues, but also modulate immune reactions, prevent apoptosis and promote angiogenesis, thanks to the secretion of trophic factors. Moreover, mesenchymal stem cells were shown to stabilize newly formed blood vessels, by differentiating in both mural and endothelial cells. The combined use of hMSCs, endothelial cells, and 3D biomaterials, was shown to especially increase the formation of both a vascular network and new bone tissue.

[0198] Therefore, the inventors combined the presence of angiogenic NRs with hMSCs onto a clinically used biphasic sponge and observed how these two players could synergistically promote vasculoneogenesis in the Antartik® bone substitute in vivo. We showed that, besides an improved recruitment of the endothelial cells from the host, the hMSCs actively contributed themselves to the formation of blood vessels, as several endothelial cells of human origin were found in the vasculature that colonized the bone substitute (FIG. 5C).

[0199] In summary, thanks to the simultaneous presence of cell contact-dependent sustained release of angiogenic factors and of hMSCs on the Antartik® bone substitutes, the following observations were made: 1) the formation of a functional vasculature, proved by the presence of mural cells also at the core of the large graft (FIGS. 5A,B); 2) a better vascularization of the bone substitute subcutaneously implanted, in terms of number, size and relative area of the blood vessels found in the presence of VEGF/HEP-NRs and hMSCs (FIGS. 3A-C); 3) a denser network of blood vessels in the bone substitutes implanted in a calvarial large bone defect (FIGS. 4A-C); and 4) an active contribution of the hMSCs to vasculoneogenesis (FIG. 5C).

[0200] As of today, autologous bone grafting (auto-transplant) is the gold standard in the treatment of large bone defects. With 2 million procedures per year worldwide, it is the second most common tissue transplantation after blood transfusion. However, it has downsides. Bone autografts present more complications than the use of synthetic bone substitutes in terms of infections and they have a higher cost. Allogenic bone grafts are even more expansive than autograft, as they have to be properly treated prior to the clinical use. However, autografts necessarily introduce a second operative site, a longer operating room time and a longer post-operative is chronic pain, which contribute to increase the overall costs of this technique. Therefore, advanced functionalized materials are needed to overcome the current limitations (the lack of blood vessels recruitment from host tissues and the necrosis induced at the core of the implanted bone substitute) and to induce the vascularization of the implanted bone substitutes, which is the cornerstone for the regeneration of functional bone tissue.

[0201] The present invention thus relates to a nano-functionalized bone substitute based on the combined presence of angiogenic nanoreservoirs and mesenchymal stem cells that aims to improve vasculoneogenesis in critical size bone defect. The results presented here have great relevance from both biomedical and public health perspectives. They showed that this innovative strategy, applied to a bone substitute already used in the clinic, was able 1) to organize endothelial cells in vessel-like structures in vitro, 2) to contribute to new vessel formation in subcutaneous implants and 3) to improve the vasculoneogenesis in a critical bone defect mouse model. The nano-functionalized bone substitute of the invention could now replace auto- and allografts in the treatment of large bone defects and, with the due modifications in the composition of the active players (cells and growth factors), could also be used for the regeneration of other tissues.