Patent Application
20090269586
Collagen is a scleroprotein with mechanical functions, that is mostly present in the animal and human connective tissue. In human, collagen makes about 30% of total proteins and is the primary component of cartilage, ligaments and tendons. Several types of collagen exist that are mainly differentiated in relation to the function of the tissue from which they originate. It is well known that the properties of collagen depend not only on the structure of protein monomers composing the triple helix of the collagen fiber, but mainly on the complex macromolecular organization that distinguishes this and other structural proteins with a mechanical function (e.g. myosin and tropomyosin in muscle fibers, etc) and which involves an organization in fibrils, their supercoiling, their head-to-tail assembly and so on. The molecular organization of native collagen fibers is partially lost during collagen extraction procedures. It can be partly restored artificially, for instance by chemical cross-linking. This treatment restores, even though only partially, the mechanical resistance of collagen fibrils and also makes the final product more resistant to proteolytic digestion.
So far, several collagen-based preparations have been introduced in medical practice, including preparations in which collagen is present together with other natural components, either in the form of gel and felt, membrane or patch, as stimulating agents for wound cicatrization, or as vehicle or device for the slow release of other drugs or as matrices suitable for interactions with cells. In fact, the importance of collagen as substrate that favours cellular interactions and growth is well known. It is also well known that cell lines of different origin exhibit preferences for different types of collagen: for instance skin fibroblasts and epithelial cells (Murray J. C. et al. Cancer Res. 40, 347, 1980; Wicha M. S. et al. Exp. Cell Res. 124, 81, 1979) grow better on type IV collagen (the collagen present in the basal membrane) whereas chondrocytes prefer type II collagen (that is present in hyaline cartilage) (Hewit A. T. et al. Proc. Natl. Acad. Sci. USA 77, 385, 1980) and adapt well also to grow on type I and II collagens.
The presence of fibronectin, endogenously produced by cells (Grinnel F. et al. Proc. Natl. Acad. Sci. USA 75, 4408 (1978) has been proven to be important in this type of application mediating fibroblast adhesion to the collagen substrate (Pearlstein E., Int. J. Cancer 22, 32, 1978), and of chondronectin, mediating the interaction of chondrocytes with collagen (Hewit et al. Proc. Natl. Acad. Sci. USA 77, 385, 1980).
Methods for extraction and preparation of collagen membranes are described in U.S. Pat. No. 5,785,983 claiming a process for preparation of collagen membranes composed by chemically unmodified type I collagen, that is obtained from collagen gels, placed in non-adherent Teflon trays, by slow drying, under suitable conditions of temperature, vacuum and nitrogen flow. The preparation of collagen gels from bovine tendon is also disclosed in WO98/44809.
Methods for preparation of felts or collagen membranes by freezing and quick drying of collagen solutions, in the form of gels, have also been described. The structure of said felts is not organized, as such procedures fix as in an instantaneous photograph the disorder of collagen fibers in solution. The resistance of these membranes is limited because the collagen, that is denatured in the extraction process, is not sufficiently reorganized.
Patent WO95/18638 claims reabsorbable collagen membranes as a guide in tissue repair, characterized by two opposite sides, of which one is fibrous and suitable for cell growth and the other is smooth and suitable to prevent adherences. Said membranes are obtained from peritoneal or placental membranes of several mammals, including calf, prepared by a gross process of cleaning, dehydration and degreasing with acetone and N-hexane. Due to the preparation method employed, membranes have the same size as the organ from which they originate, and also their peculiar features, such as porosity and type of collagen, reflect those of the tissue of origin.
Patent WO 96/25961 claims reabsorbable matrices of type II collagen (from cartilage) suitable for the reconstruction of natural cartilagineous tissues. Said matrices are prepared by freezing and lyophilization and can be integrated with glycosaminoglycans.
Patent WO 99/19005 claims a multilayered membrane comprising of a matrix primarily made of type II collagen, composed of a spongious side and a relatively impermeable side. Said membrane is prepared by sequential stratification of collagen matrices (previously degreased with acetone); during subsequent, discontinuous operations, portions of collagen gel are then lyophilized on such membranes.
Thus, up to now, collagen membrane preparation procedures involve either the extraction of collagen in a solid, semi-purified form or its solubilization in the form of gel that is then rapidly frozen and desiccated. After desiccation, collagen is generally also chemically modified by cross-linking that occurs randomly and prevents the macromolecular reorganization typical of the native fibril. Alternatively, membranes can be prepared by slow evaporation of a collagen gel or they can be prepared directly from naturally pre-existing collagen membranes that and only partially purified.
Patent EP1307247 B1 describes the preparation of collagen membranes from collagen, by exsiccation, in a magnetic field. During exsiccation, membranes acquire a macromolecular organization that can be shown by scanning electron microscopy on one side of the membrane, a feature that makes them especially suitable as coadjuvant support for cell growth.
The process described in EP1307247 B1 is further optimized to generate membranes endowed with complex molecular microstructures with distinct and original characteristics compared to those already described.
The present invention relates to multimicrolamellar collagen membranes of several types and structures (e.g. type I, type II, type III, etc.), of only one type or two types in association, but more particularly of type I and type II and/or of their mixture.
The multimolecular arrangement is obtained by building the membrane in several steps involving the sequential addition of collagen gel.
Membranes look thin, with a variably rough surface depending on the type of collagen used but always with a parallel horizontal lamellar microstructure so that they can be soaked with different fluids, act as three-dimensional scaffold for cultured cells are apt to be penetrated and colonized by cells in vivo, therefore functioning as a support suitable for direct, guided tissue regeneration. Furthermore, the present invention relates to the preparation of said membranes from collagen gels and to optimization of the preparation of the collagen gel from tissues.
a) 100% type I collagen multimicrolamellar membrane. Scanning electron microscopy (SEM) photograph. 2500× magnification.
b) 100% type II collagen multimicrolamellar membrane. Scanning electron microscopy (SEM) photograph. 5520× magnification.
c) Mixed type I and II collagen (60%-40%) multimicrolamellar membrane. Scanning electron microscopy (SEM) photograph. 5030× magnification.
The present invention relates to multilayered membranes made of collagen from various sources, in particular from equines or other mammals, particularly of type type I and II collagen and/or of their mixture and arranged as overlapping, parallel microlamellae.
Due to the purification process applied to collagen, said membranes are biocompatible, non-cytotoxic, non-sensitising and non-immunogenic; furthermore, due to the production process used, they are reabsorbable, very hygroscopic and their thickness and collagen content can be modulated.
The arrangement as “overlapping and parallel microlamellae”, hereafter also referred to as multilamellar, a definition that will be better explained below, is clearly distinct, by electron scanning microscopy (SEM), from that defined “macromolecular” arrangement, described in EP 1307247. Moreover, the multimicrolamellar structure is present throughout the membrane's depth, while the macromolecular structure is observed primarily at the surface level. Membranes obtained by the procedure involving the sequential stratification of collagen gels, even if of different composition as further detailed below, have a collagen content that can be modulated depending on the number and composition of the gel layers poured. Generally this content is comprised from 6 to 60 mg/cm2, preferably from 9 to 30 mg/cm2 and preferably from 15 to 25 mg/cm2. Within these limits, the yielding point value of the membranes is from −0.005 to 50 Mpa (Megapascal). More particularly, the yielding point value is comprised from 0.01 to 0.1 Mpa, for membranes comprising exclusively or in part the type II collagen, and from 0.5 to 50 Mpa, or preferably from 2.5 to 15 Mpa, for membranes primarily composed of type I collagen.
The Young modulus (an index of mechanical resistance but also of rigidity of membranes) is from 0.001 to 100 Mpa. Young module and other properties of collagen membranes, such as deformability and yielding point, are measured by methods well known in the art, for instance as described in Falini G. et al. J Mater Chem, 2004, 14:2297-2302.
According to a preferred aspect, the membrane Young modulus is comprised from 0.01 to 0.1 Mpa, for membranes composed exclusively or in part by type II collagen, and preferably from 0.5 to 60 Mpa, and more preferably from 5 to 40 Mpa, for membranes essentially composed of type I collagen.
According to a preferred embodiment, the membranes of the invention are made of different types of collagen, preferably type I and II collagen, arranged either in different layers or in a way so that each layer contains different types of collagen. According to the latter embodiment, at least one layer of the membranes of the invention is composed by a mixture of type I and II collagen. Even more preferably said mixed composition a percentage of type II collagen that is from 10 to 70% (w/w) of the total collagen. Even more preferably, the percentage of type II collagen is comprised from 20 to 60% (w/w) or even more preferably the percentage is from 30 to 50% (w/w).
A further aspect of the present invention is a procedure for preparing collagen membranes, wherein collagen gel is poured into electrostatically charged trays or plates according to the known art and is then partially exsiccated and then under the same conditions, additional layers of collagen gel are poured on top of the previous layer.
The first layer is allowed to exsiccate until the residual water content is comprised between 30 and 70% or, according to a preferred embodiment, until the volume of the gel layer has been reduced of at least 30-70%. Said procedure is characterized in that the next layers of collagen gel are poured after partial exsiccation of the previous layer and the gel layer that was just poured is subjected to a vertical electrostatic field. The partial exsiccation of the first layer and of subsequent layers takes place preferably at a temperature comprised from 10 to 40° C., preferably from 20 to 30° C., even more preferably from 22 to 28° C., or even more preferably from 24 to 26° C., in an aerated incubator without turbulence, such as an incubator with diaphragmed air flow intake.
After partial exsiccation of the first layer, a volume of collagen gel is poured, whose composition may differ from the composition of the previous layer. After pouring, this gel layer, having a thickness from 1.5 to 20 mm, is optionally flattened with a spatula and then exsiccated under the same conditions described for the first layer.
Also the gel layer that was subsequently poured is exsiccated until the residual water content in the first or in the previous layer is from 30 to 70%, or, as seen above, until the original gel volume is reduced of at least 30-70%.
This procedure can be repeated until reaching the useful amount of collagen/cm2 or the desired thickness.
The trays for the collagen gels are exposed to a dispersion surface, such as water, in order to completely discharge their electrostatic charge, they are filled with gel, and then they are electrostatically charged by placing them for at least 2 minutes, and preferably 5 minutes, between two flat and parallel copper foils attached to a high voltage generator. The static electrical field with vertical orientation that develops between the two foils is at least 100 kv/m, more preferably comprised between 100-220 kv/m, between 150 and 200 kv/m, o even more preferably between 160 and 180 kv/m. Trays are then placed in the desiccator. At the moment when a second or a further load of collagen gel is added to the tray, the trays are transferred onto a dispersion surface, loaded with gel, placed in the vertical electrical field, then placed again in the exsiccator. Such a procedure is also applied to any subsequent load of collagen gel.
Therefore, schematically the procedure for preparation of membranes from collagen gels involves the following steps:
After pouring the last gel layer, the membrane is left under the same temperature conditions, but without ventilation, until complete or ≧80% exsiccation, resulting in the spontaneous detachment from the bottom of the template.
The membranes thus obtained can be sterilized, for instance by physical methods, such as γ-ray irradiation with a dose comprised between 2.5 and 25 Kgy.
Thus, in the process of the invention, the concomitant induction of an electrical field and of at least one additional pouring of collagen gel onto a first, partially exsiccated collagen layer, is associated with the observation, by electron microscopy, of a multimicrolamellar structure, wherein fibrils are arranged in an organized multimicrolamellar structure.
The gel used in the procedure according to the invention is made of collagen, in aqueous solution, at a concentration from 0.1 to 5% preferably from 0.4 3%, even more preferably from 0.8 to 1.5% (weight/volume), where the weight is expressed as dry collagen residue and the aqueous solution preferably consists of a weak carboxylic acid or water.
The composition of collagen gels preferably consists of type I, type II, type III collagen or other types of collagen, or is a mixed composition. Type I and/or type II collagen or mixtures of type I and II collagen in different ratios are preferably used, preferably in a ratio 10% to 70% of type II to type I collagen gel. Particularly preferred is the solution comprising a mixture of collagen gels of different types, in which the ratios between the two gel solutions are such that the amount of collagen II is from 10 to 70% (v/v), or better 20-60% (v/v), 30-50% (v/v) and the remainder is type I collagen gel.
Preferably the collagen gel solutions used for preparation of the collagen gel mixture have a collagen concentration from 0.5 to 2%, preferably between 0.8-1.2% (w/v). Thus, in the final preferred mixture of type I and II collagen, the percentage of type II collagen is from 0.1 to 0.7%, even better 0.2-0.6%, even more preferably 0.3-0.5% (w/v), while the percentage of type I collagen is from 0.5 to 0.7%, 0.4-0.8% or 0.3-0.9% (w/v).
The essential steps, according to the prior art, of the process for collagen gel preparation, including optimization of the alkaline treatment based on the type of collagen used, and of the alkaline treatment are described next:
i) mechanical mincing of a collagen-rich tissue and its homogenization to a collagen suspension,
ii) treatment of the collagen suspension with proteolytic enzymes in acidic pH aqueous solution, optionally followed by filtering,
iii) alkaline treatment of the suspension until a pH preferably higher than 8, and its neutralization with acids to reach a pH preferably comprised between 5 and 6,
iv) collagen precipitation by salting, preferably followed by centrifugation and formation of the collagen gel upon resuspension of the precipitate at a collagen concentration comprised from 0.1 to 5%, preferably from 0.4 to 3%, even more preferably 0.8-1.5% (weight/volume), where the weight is expressed as dry collagen residue in a diluted aqueous solution of a weak carboxylic acid, at a concentration from 0.1 to 1%, preferably from 0.2 to 0.5%, or in water.
Therefore, the starting tissue is preferably a tissue rich in collagen. The tendon is preferably used as the source of type I collagen, even more preferably from horse. Tracheal or joint cartilage is preferably used as source of type II collagen, preferably equine, bovine or swine. Isolated fibroblasts in culture can also be used as source of type I collagen, while cultured chondrocytes can be used as source of type II collagen.
Type III collagen can be isolated from swine skin or from equine, bovine or swine blood vessels. Alternative collagen sources to mammalian connective tissues, for instance fish cartilaginous or bone-cartilage tissue, are well known and usable. Several types of collagens are in fact prepared starting from specific organs or tissues of vertebrates or invertebrate animals, from whole invertebrate organisms, from colonies or agglomerates of animal cells.
According to the present invention, the NaOH treatment, which removes the glycosidic component and possible transmissible contaminating agents and which is mandatory according to the most recent Communitarian Directives (e.g. 75/318/EEC December 1991, EMEA 410/01 rev. 2—October 2003), is performed according to appropriate weight ratios relative to the starting tissue. For tissues rich in type I collagen, the ratio between grams of starting tissue and grams of NaOH is from 0.05 to 0.5, preferably from 0.1 to 0.3, even more preferably from 0.12 to 0.18 or is 0.15; for tissues that are rich in type II collagen, said ratio should not exceed 3.5, and is preferably from 1.8 to 2.8, or even more preferably from 2.1 to 2.5.
Said ratios allow the optimal removal of the immunogenic fraction composed of the glycidic components of the molecule, in particular hexoses, pentoses and hexosamines, while keeping an optimal resistance of the collagen fibers, measured as resistance to enzymatic digestion with collagenases.
After addition of sodium hydroxide in the preferred ratios, NaOH concentration is adjusted to 1 N by addition of an aqueous solvent, preferably water, and the alkaline solution is incubated for at least 1 hour.
The above indicated preferred ratios are especially useful for treatment of type II collagen fibers, which show the highest extent of glycosylation among the collagens that are most commonly used for industrial purposes (I, II, III). It should also be underlined that treatment according to said optimal ratios, defined by the inventors, makes possible to treat with a 1 N sodium hydroxide solution even type II collagen fibers, which show reduced and apparently very weak fibers.
After alkaline incubation, the pH of the collagen suspension is neutralized with a solution of a strong acid: to a pH from 5 to 6, preferably from 5.3 to 5.7 more preferably 5.4-5.6 for type I collagen. Alkalinization and subsequent neutralization, (steps (iii and iv) of the procedure), are very gradual in order to avoid massive precipitation of collagen fibers.
Alternately, when a strong alkalinization of the starting material is either not required, or in other circumstances that are well known to the person skilled in the art, the pH can be adjusted to values comprised between 5 and 6 right after the enzymatic digestion by using a strong base. At this pH (comprised between 5 and 6) type I collagen precipitates upon salting as a whitish mass that can be collected and separated.
The type I collagen precipitate is then washed and resuspended at the most suitable concentration for use (expressed as weight of the dry collagen residue/volume), preferably after removing the precipitated salts by washing with purified sterile water, in a diluted solution of a weak carboxylic acid. Said solution is preferably an acetic acid solution at a concentration from 0.1 to 1% (w/v), preferably 0.3%. Other weak acids, as for instance citric acid or tartaric acid, but not ascorbic acid, can be used as an alternative to acetic acid, and their concentration is comprised between 0.5-2.5%, otherwise water can also be used. Type II collagen fibers are preferably precipitated after alkaline treatment, by adjusting the solution to pH 3.5, with a strong acid, and then adding NaCl at a concentration from 0.65 to 0.75 M, preferably 0.7M.
The precipitated type II collagen fibers can be further concentrated by filtration and/or centrifugation at 4500 rpm for 20 minutes, are subsequently dialyzed to remove salts and then dissolved in a diluted solution of weak carboxylic acid. This acid is even more preferably acetic acid at a concentration from 0.1 to 1% (w/v), preferably 0.3%. Alternatively, water can also be used.
The precipitation of collagen fibers can be optionally repeated under the same conditions if a cleaner product is desired.
The collagen gel, obtained as described above, can be then homogenized to facilitate resuspension of the precipitate.
Prior to pouring, the gel is preferably warmed up at 24-30° C., and then degassed under vacuum. Other substances can be added to the collagen gel prior to pouring, as for instance biological components of specific body parts or tissues. An example of biological components is represented by polysaccharides of animal or vegetal origin, or matrix proteins such as fibronectin and laminin, or else hormones and growth factors, etc.
The composition of layered collagen gels and the number of layers determine the properties of the membranes: in fact, membranes with different multimicrolamellar structure, biphasic, triphasic and so on, and with different characteristics of resistance and hygroscopy, are obtained by stratification of different types of collagen gels, where for instance the first layer is made of type I collagen and the second of type II collagen or vice versa. In particular, type I collagen multimicrolamellar membranes are much more hygroscopic than type I collagen membranes with the same thickness and prepared according to the prior art by a single pouring and drying (see table 5). The former can absorb a noticeable amount of water, thus swelling until their volume increases 4-5 fold compared to the original volume.
At the same time, such membranes are also more resistant to traction as they have a Young modulus (an index of resistance but also of rigidity of the membrane) that is 7-fold higher compared to membranes known in the art with the same thickness, and a 8-fold higher yielding point.
Type II collagen multimicrolamellar membranes are more hygroscopic than membranes consisting exclusively of type I collagen: they can swell up to a volume that is 7-fold higher that their original volume.
1) the maximum deformability upon traction, that is the maximum degree of stretching reached, is 13 fold less for a membrane composed exclusively by type II collagen compared to a membrane composed exclusively by type I collagen. Increasing amounts of type I collagen result in a deformability that is slightly higher and non-proportional to the amount of type I collagen present in the membranes. For a membrane composed exclusively by type II collagen, the Young's module, measured according to methods known in the art, is about 3000-fold lower compared to a membrane composed exclusively by type I collagen. For membranes composed by a mixture of the two types of collagen, and only those containing at least 50% type I collagen, the Young's module reaches a maximum value that is 3-fold higher than the value of type II collagen membranes;
2) the yielding point, that is the maximum load which can be applied, appears to increase with increasing proportions of type I collagen, but it is anyway always lower than 70% in all the membranes that are different from those composed exclusively by type I collagen.
3) the differences among the various types of membrane are less pronounced for the breaking stress, a parameter measuring the load that causes rupture.
4) breaking strain follows a profile that is proportional to the content in type I collagen but is dramatically higher for a membrane composed exclusively by type I collagen than for the others (about 20-100 fold, depending on the ratio between the two types of collagen).
Multiphasic membranes, composed by multiple layers each consisting of a single type of collagen, keep the respective characteristics of type I or type II collagen multimicrolamellar membrane. The biphasic membrane, also termed C1+C2, becomes slightly sticky on the surface of type II collagen once it is dipped in water, and is more hygroscopic than a membrane composed exclusively by type II collagen, although it does not reach the level of hygroscopicity of a membrane composed by a 50% mixture of type I and II collagens (C2A-50%).
The membranes, whose preparation is described herein, have different characteristics depending on the type of collagen used and the number and type of stratification performed. However, to our best knowledge, the multimicrolamellar arrangement is novel and original compared to all other membranes that have been obtained according to procedures known in the art, and this arrangement is common to all membranes obtained from collagen gels according to the procedure herein described.
Said structure can be well evidentiated by analysis with the scanning electron microscope (SEM), that reveals a structure of the membranes that is characterized by parallel overlapping microlamellae. Lamellae are formed regardless of the number of sequential additions of collagen gel to the tray: two sequential additions of collagen gel, under the conditions described, are sufficient to obtain a membrane composed of microlamellae. A higher number of additions results in thicker membranes and in a higher content of collagen/cm2, thus these membranes are also composed of a higher number of microlamellae.
Under the scanning electron microscope, the multimicrolamellar organization of transversely cut type I collagen membranes appears as parallel, overlapping leaflets, which are wide and rather regular but are placed at a close distance that is not constant (see
The multimicrolamellar organization of transversely cut type II collagen membranes appears to consist of small, roundish, irregular, frayed lamellae that are arranged as fish scales (see
Surprisingly, the mixture of type I and II collagen gels results in membranes composed of lamellae that are more homogeneous and regular than those described for type I or II only (see
As can be seen also in the SEM images, microlamellae adopt a more ordered structure when the ratio between the two types of collagen is comprised between 20-60% of type II collagen and 80-40% of type I collagen, and membranes with this type of structure are preferred.
Membranes obtained with type I and II collagen mixtures are less resistant to traction than membranes composed exclusively of type I collagen and, surprisingly, are just a little more resistant than membranes composed exclusively of type II collagen, thus they do not show a reliable correlation between proportion of type I collagen and resistance. However they are highly hygroscopic due to the presence of type II collagen: in fact, the incorporation of type II collagen in the organization of type I collagen molecules evidently leads to a different structure of the membrane, making it more easily penetrable by fluids and cells.
In vivo experiments show that multimicrolamellar membranes, preferably the membranes with mixed composition of type I and II collagen, even more preferably those containing a 20-60% proportion of type II collagen, particularly between 40 and 50%, are especially suitable for regeneration of the cartilage tissue.
In fact, in vivo repair experiments conducted on animal models reveal that the lesion is populated by cells with characteristics of the cartilaginous lineage already at 3 months after the induction. These cells are apparently viable, they are organized in clusters and are separated by a not really abundant matrix that is intensely basophilic and poorly alcianophilic, indicating the presence of a newly formed matrix whose glycosaminoglycan content is still low.
At 9 months, the lesion is filled with repair tissue that is well anchored to the underlying bone and is composed by cartilaginous cells and a newly formed cartilaginous hyaline matrix, as clearly shown in
Histological evidence of cell-mediated immune reactions was not observed in any of the experiments performed. On the contrary, cells attracted to the lesion site and stimulated to grow and/or to differentiate by the membrane three-dimensional support or scaffold, give rise to de novo formation of cartilage. In this respect, in vitro experiments provide evidence that chondrocytes proliferate and maintain their phenotype for a long time when they are cultured on multimicrolamellar membranes composed of type I and II collagen and that pluripotent mesenchymal stromal cells, preferably isolated, proliferate and differentiate as chondrocytes when they are cultured on the same multimicrolamellar collagen membranes with mixed composition (type I and II collagen). Cellular proliferation can be measured, for instance, by the tetrazolium salt test, while chondrocytic differentiation is measured, for instance, by histochemical or immunohistochemical staining with Safranin-O, which primarily reacts with type I collagen and chondroitin sulphate. Multimicrolamellar membranes composed of type I collagen and of type II/I collagen, in a 40/60 ratio, are particularly preferred for both differentiation and proliferation.
In particular, a higher degree of proliferation of adult chondrocytes is observed on the mixed membrane composed of type II/I collagen.
MSC proliferate on the type I collagen multimicrolamellar membrane and differentiate into cells that stain only weakly with Safranin-O and produce type I collagen and chondroitin sulphate. On the contrary, when using the mixed type II/I collagen multimicrolamellar membrane (containing a collagen gel mixture composed preferably of 40% type II collagen and 60% type I collagen), MSC differentiate into chondroblasts showing a clear immune reaction to the S-100 protein, chondroitin sulphate and type II collagen, which are typical reactions of differentiated cartilaginous cells.
The advantages of the membranes according to the invention can be summarized as follows:
Various experiments, involving the in vivo implant of membranes of the present invention, show that the repair of osteochondral lesions occurs through the following phases:
The above described characteristics make these membranes especially suitable for direct use in guided tissue regeneration, without a step of cell expansion in vitro. They function as support to the growth of various cell types involved in regeneration, in particular of various types of stem cells such as mesenchymal, stromal and various types of differentiated adult cells. In the case of stem cell, the above described characteristics make these membranes especially suitable also for proper differentiation and regeneration of bone, connective and cartilaginous tissues, depending on the collagen/collagens that compose the membrane. Therefore the multimicrolamellar membranes according to the invention, preferably those with mixed composition including type I and II collagen, are suitable for repair of cartilaginous lesions, in animals and humans. Such repair is optimized by a subchondral perforation of the bone at the site of lesion performed before placing the membranes of the invention. As mentioned above, such perforation facilitates the attraction of mesenchymal cells from the bone and colonization of the membrane, followed by their in situ differentiation in chondrocyte-like cells. Multimicrolamellar membranes according to the present invention, prepared with the most suitable type of collagens, are used for repair of cartilaginous and bone tissue and for reconstruction of tendons, ligaments and muscle. They are especially useful for the repair of osteochondral lesions or as support membranes. Therefore the membranes of the invention are especially suitable for reconstruction of osteo-cartilaginous tissue, tendons, bones, joints and as support for non-embryonic stems cells. As described in the experimental part, they function both as physical support and as support capable of inducing differentiation of non-embryonic stem cells in cells of the osteochondral lineage. Therefore, they are also suitable as support for in vitro colonization by non-embryonic stem cells. According to a further aspect, the invention relates to a therapeutic method for repair of cartilaginous lesions, carried out by direct insertion of the membranes in the lesion site, preferably without membrane cellularization (colonization) step and also without cell expansion in vitro.
Said therapeutic method is optimized by performing subchondral perforations of the bone before placing the membrane of the invention in the lesion site.
Method of induction of the electrical field: the trays, completely discharged of electrostatic charge, are placed for 5 minutes between two flat and parallel copper foils connected to a high voltage generator. A static electrical field of about 170 kv/m, with vertical orientation, develops between the laminae.
Method for loading trays for collagen gel stratification: after charging in the electrical field induced as described above and then deposited on an insulating surface the trays are loaded with the first collagen gel layer, transferred again in the vertical electrical field of 170 kv/m for 5 minutes and then placed in an exsiccator. At the time when a second load of collagen gel is added to the tray, the trays are removed from the exsiccator, transferred on a dispersion surface, loaded with gel, placed in the vertical electrical field of 170 kv/m for 5 minutes, then placed again in the exsiccator. Said procedure is also applied to any subsequent collagen gel loads.
Human Dermal Fibroblasts
Human dermal fibroblasts have been used, that were isolated from skin biopsies from healthy adult subjects, cultured in Dulbecco's Modification of Eagle's Medium (DMEM) with high glucose concentration, containing L-glutamine 2 mM, penicillin 100 UI/ml, streptomycin 100 μg/ml, Na pyruvate 1 mM, non essential amino acids (NEAA) 1× and fetal bovine serum (FBS) at 10% final concentration. Fibroblasts were seeded in a two-well “chamber slide” (area=8.26 cm2), 30000-50000 cells/well, final volume 2 ml. The culture medium was changed every 24 hours.
Human Chondrocytes
Human chondrocytes were used from biopsies of healthy adult subjects who suffered traumatic lesions. The cell suspension, obtained by trypsinization of a cartilaginous fragment, was washed several times with complete DMEM and chondrocytes were counted with the vital staining method that makes use of eosin, seeded in culture flasks (Costar®, Cambridge, Mass., USA) in complete DMEM medium, at a density of 20000/cm2, and then placed in a 37° C. incubator in humidified atmosphere with 7% CO2. After adhesion to the support, cells were washed twice with calcium- and magnesium-free phosphate buffer (PBS 1×) and then detached with trypsin/EDTA 1× at 37° C. for 10 minutes. The reaction was then blocked with complete DMEM, cells were collected, transferred to large sterile tubes and centrifuged for 5′ at 1800 rpm. Two-three ml of complete DMEM were added to the pellet, cells were counted after eosin staining and re-seeded in larger flasks. The medium was changed twice a week.
Electrophoresis by the SDS-PAGE PhastSystem Amersham was performed in Phastgel gradient 4-15% slides.
50 g of equine joint cartilage powder were repeatedly washed in water, drained and placed in 3 liters of water that were brought to pH 2.5 with hydrochloric acid solution. Five grams of pepsin were added to the suspension (10% of the starting cartilage) under mechanical stirring.
After about 20 hours at about 27° C., the suspension was again stirred repeatedly, filtered, and the fluid was collected, diluted to 3.75 liters and adjusted to pH 3.5 with few drops of a 2.5 N sodium hydroxide solution. 154.85 g of sodium chloride was added to the fluid to a concentration of 0.7M: after about 2 hours, a precipitate was formed consisting of very short (few mm) and thin fibers. The precipitate was collected on a cotton cloth by filtration under vacuum and subsequent centrifugation at 4500 rpm for 20 minutes, until a very concentrated suspension was obtained. An aliquot of the suspension (28 g) was dialyzed against water in a Spectrapore 1000 MW tubing for 65 hours at 4° C. At the end of dyalisis, the sample was brought to 100 g with 0.3% acetic acid and was designated C2/030717. Instead, other 300 g of the concentrated fiber suspension were treated with 20 g of sodium hydroxide and brought to 500 ml (1N sodium hydroxide solution). The ratio between starting cartilage (about 46 gm after removing the first aliquot of 28 g of concentrated fiber suspension) and sodium hydroxide was calculated to be 46 g/20 g=2.3. Fibres completely dissolved were kept in sodium hydroxide for 1 hour at about 23° C. After diluting with 1.5 liters of water and adjusting the pH to 3.5 with HCl, a new precipitation of fibers was obtained in about two hours. The new fibers, collected by centrifugation, were dyalized in Spectrapore 1000 MW tubings in water. A 90% proportion of the so obtained sample, already in the form of a gel that was not very viscous, was dissolved in a few milliliters of acetic acid at 0.3% final concentration and was designated C2/030718 (628 g). The other 10% was dissolved in a proportional amount of water: a gel was obtained also in this case, that was designated C2/030718H (about 50 g).
C2/030717 and C2/030718 collagens were electrophoretically (PAA 4-15%) identical to a commercial type II collagen that was taken as reference (SIGMA C-1188). C2/030718 was cleaner than C2/030717 which is shown by the low molecular weight electrophoretic bands which are completely removed by sodium hydroxide treatment.
Table 1 shows the characteristics of the 2 samples:
Free NH2 groups remain almost unchanged, while sodium hydroxide treatment removes a fraction of sugars and hexosamines. Table 2 shows several tests of treatment of type II collagen fibers with sodium hydroxide.
The table shows various ratios between the starting cartilage and sodium hydroxide in relation to the characteristics of the resulting collagen (presence of sugars, percentage of hexosamines, resistance of the molecule to collagenases). Sodium hydroxide treatment of type II collagen gives optimal results when the ratio between the starting tissue and sodium hydroxide is comprised between 2 and 3.5 and the ratio between grams of tissue/ml is comprised between 0.08 and 0.15. It should be noted that membranes from lots 030718 and 050415 are more resistant to the attack by collagenases at a concentration of 10 U/ml at 37° C.
Table 3 shows the evaluation of mechanical resistance to traction of the membrane obtained from collagen 030718 compared to that obtained from 050615 (C2A050617P100 membrane).
An amount of 80 g of ground equine tendon was placed in 1 liter of water, left to swell for two hours, homogenized in a Sterilmixer Lab homogenizer. PBI: 0.81 g of pepsin were added to the suspension that was then brought to 10 liters with water. The pH was adjusted to 2.5 and the suspension was left for 23 hours at 20° C.+/−2° C. After this time the suspension was again homogenized with a Sterilmixer and filtered on cotton cloth. The liquid filtrate was adjusted to pH 5.5 with 30% sodium hydroxide, resulting in precipitation of collagen fibers that were washed twice with water at pH 6. An amount of 304.55 g of wet fibers was recovered. From this material, 50 g were washed with water 2 more times and then dissolved in 0.3% acetic acid (they were brought to 180 g in order to obtain a 1% collagen concentration: the so-formed gel (050920 A) was repeatedly homogenized with a Sterilmixer and was degassed under vacuum and under mechanical stirring. Fibers left after removal of the 50 g aliquot (253.9 g) were treated with 1540 ml of 30% sodium hydroxide and brought to 12 liters (in order to obtain 1N sodium hydroxide concentration) for 1 hour. Under the conditions used, the ratio between grams of starting organ (tendon) and sodium hydroxide is about 0.5. This tendon/sodium hydroxide ratio turned out to be optimal for purification of type I collagen suitable for preparation of multimicrolamellar membranes.
After one hour incubation, the solution was brought to pH 5.5 with HCl, thus resulting in re-precipitation of fibers. These were drained, washed 3 times with water at pH 6 and then recovered with 0.3% acetic acid in an amount suitable to obtain 798 grams of 1% collagen gel (lot 050920 AN).
Human dermal fibroblasts isolated from skin biopsies of adult healthy subjects were cultured in plastic wells according to the conditions described in the section “materials and methods”. The cell culture medium was replaced according to a pre-arranged schedule, so that cells were always fed with fresh medium. The medium withdrawn from wells every 24 hours was collected and frozen at −20° C. Aliquots kept at −20° C. were thawed and pooled together: in total, 102.66 ml of culture medium were collected, containing products of cellular metabolism. The fluid was subjected to a process for the extraction of type I collagen, following the method described in example 2, with an appropriate resizing of the procedure. At the end of the process, 19.4 mg of wet fibers were obtained (which turned out to consist of type I collagen, by polyacrylamide gel electrophoresis) that were recovered in 50 microliters of 0.3% acetic acid: a viscous and homogeneous gel was formed in few minutes.
Human chondrocytes, isolated from joint cartilage of adult healthy subjects who have undergone several type of surgical interventions following traumas, were cultured in plastic wells according to the conditions described in the section “materials and methods”. The cell culture medium was replaced according to a pre-arranged schedule, so that cells were always fed with fresh medium. The medium withdrawn twice a week was collected and frozen at −20° C. Aliquots kept at −20° C. were thawed and pooled together: in total, 77.23 ml of culture medium were collected, containing products of cellular metabolism. The fluid was subjected to the process for extraction of type II collagen, following the method described in example 1, with the appropriate scale down. At the end of the process, 7.89 mg of very short and light wet fibers were obtained (which turned out to consist of type II collagen, by polyacrylamide gel electrophoresis) that were recovered in 25 microliters of 0.3% acetic acid: a clear, fluid and homogeneous gel was formed in few minutes.
All the trays used for preparation of the membranes of this example were exposed to an electrostatic field according to the procedure described in the section “Materials, methods and equipment”
a) Membrane with 9 Mg of Collagen/Cm2 Lot 40122/1
Membranes were prepared from a 1% collagen gel, equilibrated at 26° C.+/−2° C. An amount of 65.5 g of 1% collagen gel, lot 050920 AN, was poured in a square polystyrene tray having a side of 12 cm, and left to expand until a homogenous layer covered the bottom completely. The gel was spread with a 11.6 cm wide spatula made of the same material as the tray, until a homogenous layer covered the bottom completely. The edge of the spatula was passed several times on the gel. The thickness of this layer was 5.52 mm, as measured by an Electronic outside micrometer IP54 IDF. Time was recorded as T01 upon the first stratification. The tray was placed in thermostat at constant temperature of 26° C.+/−2° C., with ventilation set to the minimum, until its thickness was reduced to 2.85 mm after 48 hours, due to evaporation. At this point, an amount of gel identical to the first amount was again poured in the tray, covering completely and homogeneously the semi-solid surface of the previous layer: also in this case the gel was spread with a 11.6 cm wide spatula made of the same material as the tray, until a homogenous layer covered the bottom completely. The edge of the spatula was passed a few times on the gel. The total thickness of the new semi-solid membrane (6.36 mm) was recorded at the new T0 (T02) and subsequently controlled over time, monitoring the thickness every 24 hour, while the system was kept at 26° C.+/−2° C. without ventilation. At this point the membrane was kept under the same temperature conditions and without ventilation until complete exsiccation and spontaneous detachment from the bottom of the template.
b) Membrane with 14 Mg of Collagen/Cm2 Lot 40122/2
In a square polystyrene tray having a side of 12 cm, 65.5 gm of 1.56% collagen gel, lot 40121D, were poured. The membrane was prepared according to the procedure described in a).
c) Membrane with 18 Mg of Collagen/Cm2 Lot 40122/3
Membranes were prepared according to the procedure described in a), starting with 130 g of 1% collagen gel (lot 050920 AN), equilibrated at 26° C.+/−2 and poured in a square polystyrene tray having a side of 12 cm.
d) Membrane with 23 Mg of Collagen/Cm2 Lot 40122/4
Membranes were prepared starting from 100 g of 1.11% collagen gel, that was equilibrated at 26° C.+/−2° C. and poured in a square polystyrene tray having a side of 12 cm. The total thickness of the new semi-solid membrane (8.12 mm) was recorded at the new T0 (T02) and subsequently controlled over time, monitoring the thickness every 24 hour, while the system was kept at 26° C.+/−2° C. with minimal ventilation.
Onto the 4.26 mm thick sheet (72 hours after the last gel addition) a third amount of gel, having a weight identical to the previous two amounts, was homogenously distributed on the semi-solid surface of the underlying layer, always using a polystyrene spatula. Again the thickness of the semi-solid membrane was recorded as T03 (9.06 mm). At this point the membrane was kept under the same temperature conditions and without ventilation until complete desiccation and spontaneous detachment from the bottom of the template.
Examples a, b, c, d indicate that the characteristics of the membrane can be modulated by changing gel concentration and number of layers. The table below shows the absorption capacity of each membrane of example 3 within 24 hours. Table 4 shows the increase in thickness (mm) over time of the above described membranes dipped in water.
Membranes were prepared starting from a 1.06% collagen gel, lot 78802, prepared as in example n.2. The gel was subjected to deaeration (to exclude air bubbles) by passage in a suitable planetary mixer under vacuum and stirring for 24 hours, then it was kept under vacuum without stirring for additional 12 hours, followed by equilibration at 28° C.+/−2° C. An amount of 130 g of collagen gel was poured in square polystyrene trays having a side of 12 cm, exposed to an electrical field as described in materials and methods, paying attention not to form air bubbles. Using the above described spatula, the gel was distributed in each tray until a homogenous layer, with an average thickness of 10.52 mm, covered the bottom completely. The trays were kept at a constant temperature of 26° C.+/−2° C., with ventilation set to the minimum, until their average thickness was reduced to 5.55 mm after 46 hours, due to evaporation. At this point, an amount of gel identical to the first amount was again poured in each tray, following the procedure already described. The new average total thickness of the semi-solid membranes was 14.11 mm. The system was kept at 26° C. without ventilation until complete spontaneous detachment of the exsiccated membranes from the trays. By SEM analysis, the resulting membranes, with a homogeneous thickness of 0.25 mm, appear to be composed of several parallel microlamellae.
Membrane 40126 swells in water up to a 4-fold increase of thickness.
The graph in table 5 shows the increase in thickness of membrane 40126 in water compared to a membrane prepared according to EP1307247, with the same initial thickness (lot 40124).
Moreover, the multimicrolamellar membrane 40126 is more resistant to traction than 40124, as shown in the histogram in
Membranes were prepared starting from a 1% gel composed of type II collagen (lot 030718), equilibrated at 18° C.+/−2. An amount of 65.5 g of collagen gel was poured in a square polystyrene tray having a side of 12 cm, exposed to an electrical field, as described in materials and methods, and distributed with a spatula made of the same material as the tray, forming a 5.02 mm thick layer. The tray was incubated at a constant temperature of 24° C.+/−2° C., without ventilation, until the thickness was reduced to 1.89 mm in 48 hours, due to evaporation (the thickness was controlled every 24 hours). At this point an amount of gel equal to the first amount was again poured in the tray, covering completely and homogeneously the semi-solid surface of the previous layer, whose surface was rendered homogeneous with the polystyrene spatula described previously. The total thickness of the new semi-solid membrane (6.36 mm) was monitored over time, performing controls every 24 hours, keeping the system at 24° C.+/−2° C. without exsiccation until complete exsiccation and spontaneous detachment from the bottom of the tray.
The membrane (9 mg/cm2) was marked C2A-030718 (
The weight and the thickness of membrane C2A-030718 and of the reference type I collagen membrane C1-78908, comparable by collagen content per cm2, were assessed before and after water immersion for 1 hour.
Compared to a similar type I collagen membrane, the membrane C2A-030718 is very hygroscopic, as shown by the data in table 4.
A few mixtures of type I and type II collagen, in various proportions, have been prepared, starting from the type II collagen gel lot C2/030718 (see example n.2) and from the type I collagen gel lot 78791 (both of which are about 1%). The two gels were equilibrated at 22° C.+/−2, and mixed in several proportions:
5 g of C2A/030718+95 g of 78791=5% type II collagen
10 g of C2A/030718+90 g of 78791=10% type II collagen
20 g of C2A/030718+80 g of 78791=20% type II collagen
40 g of C2A/030718+60 g of 78791=40% type II collagen
60 g of C2A/030718+40 g of 78791=60% type II collagen
80 g of C2A/030718+20 g of 78791=80% type II collagen
From the above mentioned mixtures, a corresponding number of membranes were prepared by superimposition of two sequential gel loads, as described in example 7.
By SEM analysis, microlamellae appear large and distant one from the other: this structure allows an optimal absorption, that is three-fold higher compared to membranes consisting exclusively of type I collagen and one and half-fold higher compared to membranes consisting exclusively of type II collagen (table 7).
From table 7, it turns out that membranes obtained from mixtures of type I collagen and type II collagen do not have predictable behaviour when they are evaluated with respect to weight and thickness before and after imbibition in water for one hour. In particular, a membrane containing 5% type II collagen (C2A-5%) gives rise to a more compact and less hygroscopic membrane compared to a membrane composed exclusively of type I collagen (C1-100%), while 40% type II collagen (C2A-40%) renders the membrane even more suitable for fluid absorption than a membrane composed exclusively of type II collagen (C2A-100%).
A few mixtures of type II and type I collagen, in various proportions, have been prepared, starting from the type II collagen gel lot C2/050415 and from the collagen gel lot 78908 (both of which are about 1%). The two gels were equilibrated at 20° C.+/−2° C., and mixed in several proportions:
0 g of C2A/050415+100 g of 78908=100% type I collagen
25 g of C2A/050415+75 g of 78908=25% type II collagen
50 g of C2A/050415+50 g of 78908=50% type II collagen
75 g of C2A/050415+25 g of 78908=75% type II collagen
100 g of C2A/050415+0 g of 78908=100% type II collagen
From the so prepared mixtures, membranes were obtained that were prepared as in example 7.
Table 8 shows the weight and the thickness of mixed type I+II collagen membranes in different proportions before and after imbibition in water for one hour. Clearly, the more organized structure, which is obtained with a percentage of type II collagen of about 50%, leads to a better predisposition of the membrane for imbibition (see also the example 6, where the sample containing 40% type II collagen is morphologically more regular and organized and functionally more hygroscopic).
Membranes were prepared starting from 1.03% type II collagen gel (lot 050627, containing traces of type I collagen, prepared following the method described in example n. 1, by using a cartilage/sodium hydroxide ratio=about 2, calculated from the presumed amount of cartilage in the starting trachea) that was equilibrated at 22° C.+/−2° C. Also in this case, membranes were prepared with two subsequent gel additions according to the procedure described in example n.5. The membrane lot has been identified with the C2T/100/1 abbreviation. By scanning electron microscopy, also this membrane turned out to consist of parallel microlamellae.
The membrane was prepared in a round polystyrene plate of 6 cm diameter, starting from type II collagen gel lot 030718H, that was obtained by dissolving collagen fibers in water rather than acetic acid. The preparation method is the same as described in example 7.
The resulting membrane, 0.14 mm thick, is very fragile once hydrated, but is able to swell up to 1.22 mm thickness after 1 hour (almost 9-fold compared to the initial thickness). The swollen membrane is very sticky.
The membrane was prepared starting from 1.15% type I collagen gel, lot 78791, that was prepared as in example 2, and from 1.07% type II collagen gel, lot C204120, that was prepared as in example 1. An amount of 130 g of collagen was poured in a square polystyrene tray having a side of 12 cm, exposed to electrostatic field as described in materials and methods, paying attention not to form air bubbles. The gel was spread to form a homogenous layer, using a spatula made of the same material as the tray. The thickness of this layer was 10.46 mm. The tray was incubated at constant temperature of 24° C.+/−2° C., with ventilation set to the minimum, until its thickness was reduced to 5.39 mm after 72 hours, due to evaporation. At this point, 130 g of 1.07% type II collagen gel (lot C204120) were poured in the tray, covering completely and homogenously the semi-solid surface of the previous layer. Also in this phase the gel was stratified homogeneously with the polystyrene spatula. The new thickness of the semi-solid membrane (14.23 mm) was monitored over time, performing controls of the thickness every 24 hours, keeping the system at 26° C.+/−2° C., without ventilation, until complete exsiccation and spontaneous detachment from the bottom of the template. The membrane, termed C1+C2, becomes slightly sticky on the surface of type II collagen once it is dipped in water, and is more hygroscopic than a membrane composed exclusively by type II collagen, although it does not reach the level of hygroscopicity of a mixed membrane composed by a 50% mixture of type I and II collagens (C2A-50%).
Table 9 shows the data relative to the swelling in water of the bilayer type I+type II collagen membrane (C1+C2) compared with membranes consisting exclusively of type I collagen (C1-100%), type II collagen (C2A-100%), and the data relative to the swelling in water of the membrane obtained from a 50% mixture of type I and II collagen gels (C2A-50%).
With respect to various parameters that have been considered, membranes composed exclusively by type II collagen are much less resistant to traction than membranes composed exclusively by type I collagen.
Membranes obtained by mixing the two collagens in various proportions have unpredictable mechanical characteristics: the presence of type I collagen results in little and not always proportional (see Young's module) increase of membrane resistance.
From the data shown in table 10, it can be observed that:
1) the maximum deformability upon traction, that is the maximum degree of stretching, is 13 fold less for a membrane composed exclusively by type II collagen compared to a membrane composed exclusively by type I collagen. Increasing amounts of type I collagen results in a deformability that is slightly higher and non-proportional to the amount of type I collagen present in the membranes.
For a membrane composed exclusively by type II collagen, the Young modulus (an index of resistance but also of rigidity) is about 3000-fold lower compared to a membrane composed exclusively by type I collagen. For membranes composed by a mixture of the two types of collagen, and only those containing 50% and 75% type I collagen, the Young modulus reaches a maximum value that is 3-fold higher than the value of the type II collagen membrane.
2) yielding point, which is the maximum load that can be applied, appears to increase with increasing proportions of type I collagen, but anyway it is always lower than 70% in all the membranes that are different from those composed exclusively by type I collagen.
3) the differences among the various types of membrane are less marked for the breaking stress, a parameter measuring the load that causes rupture.
4) breaking strain follows a profile that is proportional to the content in type I collagen but is dramatically higher for a membrane composed exclusively by type I collagen than for the others (about 20-100 fold, depending on the ratio between the two types of collagen).
Membranes obtained from mixtures of type I collagen and type II collagen were evaluated for their microlamellar structure with a scanning microscope (SEM) at high magnification.
a shows a SEM analysis at 2500× magnification of the 100% type I collagen multimicrolamellar membrane (40126).
Images show that type I collagen microlamellae are thick and further stratified, while those of type II collagen are thin, frayed and disorderly arranged.
c shows a SEM analysis at 5030× magnification of a mixed multimicrolamellar membrane composed by 40% type II collagen (C2 40%). Mixing the two collagens in the proportion 40% type II-60% type I leads to a better organization of the membranes in ordered and defined microlamellae.
Type I collagen multimicrolamellar membranes, type I/II collagen mixed multimicrolamellar membranes and biphasic type I collagen+type II collagen membranes were tested in rat for a preliminary experiment on direct cartilaginous regeneration without a step of in vitro expansion of autologous chondrocytes, and they were compared with type I collagen monolayer membranes. The method is described in the section “Materials, methods and equipment”.
1st group: in the right knee, the type I collagen membrane, lot 40124, prepared according to EP1307247.
2nd group: in the right knee, the type I collagen multimicrolamellar membrane, lot 40126.
3rd group: in the right knee, the mixed type II/I collagen multimicrolamellar membrane (40% and 60%, respectively), lot C2A40.
4th group: in the right knee, the biphasic type I+II collagen membrane, lot C1+C2, with the type II collagen surface facing the lesion and the type I collagen surface facing the articular cavity. In this group, it was not necessary to fix the membranes with type I collagen gel, because the type II collagen surface, that was slightly gellified after immersion in physiological solution, was sticking to the lesion borders.
Lesions of the left knee (control) were evident in all animals. The lesion bottom did not appear very different at time 0, even if a little filling layer was clear in a few cases. On average, the reconstruction was 0.02 mm.
In the group treated with 40124, defects were evident and the cartilage appeared only partially reconstituted and irregular at the bottom of the lesion. On average, the thickness of the cartilage in the lesion was 0.05 mm.
In the group treated with 40126, defects were also evident but the cartilage appeared reconstituted to a certain extent, though irregularly in a significant portion of the lesion. On average, the thickness of the cartilage in the lesion was 0.12 mm.
In the group treated with C2A40, the defects were almost completely filled with tissue very similar to the surrounding cartilage. On average, the thickness of the cartilage in the lesion was 0.36 mm, and the surface was quite regular.
In the group treated with C1+C2, reconstitution was very similar to that observed with C2A40. On average, the thickness of the cartilage in the lesion was 0.30 mm, and the surface was quite regular.
The in vivo cartilage regeneration test was carried out according to the method of Pridie (Pridie K. H. A Method of resurfacing Osteoarthritic Knee Joints. J. Bone Joint Surg., 1959, 41-B, 618-619) both in rat and sheep.
Under conditions of surgical sterility, a longitudinal side cut of soft tissues and of the articular capsule was made in all animals, which had been subjected to general anaesthesia and, after medial dislocation of the rotula, the medial femoral condyles of the right and left femur were exposed in order to inflict the cartilaginous lesion. Deep perforations were made at the bottom of every lesion, to allow the penetration of mesenchymal cells coming from the bone underlying the site of lesion. The lesion of the left knee of each animal was not treated, while the lesion on the right was filled with the various membranes according to the group. In the rat, the membranes, after hydration in physiological solution, were fixed to the lesion borders with a thin layer of 2% type I collagen gel. In the sheep, instead, membranes were sutured. At last, the rotula was placed back in position, and the articular capsule and the skin were sutured.
After 3, 6 and 9 months, animals were euthanized, femurs were explanted and cartilaginous lesions were evaluated.
Ten adult sheep, that is after reaching skeleton maturity, were used in total. Three millimeter diameter osteochondral lesions were made bilaterally at the level of the condylus and of the femoral trochlea. On one side, lesions were treated by perforations of the bottom and insertion of a membrane prepared as described in the example no. 8 (40% type II collagen-60% type I collagen), fixed by mini-anchor with thread. As control, controlateral lesions were not treated. Animals were sacrificed at 3 and 9 months. In particular, the repair of an experimental type of complete traumatic lesion (osteochondral) was evaluated.
For microscopic analysis the samples, immediately after withdrawal, were fixed in 4% paraformaldehyde, 0.1 M phosphate buffer pH 7.4, decalcified in 4N formic acid and sodium citrate, embedded in paraffin and sectioned. Five-six micron thick sections were stained with hematoxylin-eosin, Alcian-PAS, Masson's Trichrome, Fullmer-Halmi.
In the controls at 3 months, lesions of the joint surface appear empty or partially filled with fibrin clots (
In the subsequent phases, the repair tissue coats the subchondral bone lamina and covers the lesion bed, like a cloth. The joint cartilage around the lesion shows early degenerative phenomena: disappearance of the columnar arrangement of the cellular component; notches from the more superficial layers deepen to the point to determine delamination and fibrillation of superficial and intermediate layers (
At 9 months, the lesion is filled by a repair tissue that is well anchored to the underlying bone and is composed by cartilaginous cells and by a hyaline cartilaginous matrix (
Results of the application of the multimicrolamellar membrane, compared to a control lesion, can be seen in
The relation between the newly formed cartilage and the surrounding tissues, at 9 months from application of the I/II membrane, are also evident in
The experiment shows that osteochondral lesions are repaired by filling the defect with a high number of osteochondral progenitor cells (mesenchymal, that is multipotent of stem type) coming from perforations of the subchondral bone. In addition to providing protection inside the lesion, the implant of collagen-based membranes according to the invention provides a three-dimensional support that is biologically suitable for invasion, proliferation and chondroblast differentiation of the above mentioned cells.
It was shown in vitro that chondrocytes proliferate and maintain their phenotype for a long time when they are cultured on multimicrolamellar membranes composed of type I and II collagen, and that pluripotent mesenchymal stromal cells are able to proliferate and differentiate as chondrocytes when they are cultured after appropriate separation on the same scaffolds.
Chondrocytes and adult mesenchymal stromal cells from bone marrow (MSC) were cultured on multimicrolamellar membranes made of type I collagen and of type II/I collagen in a ratio 40/60 (described above in example n.6 and n.8, respectively).
Cell proliferation was assessed by the MTT test, while differentiation was assessed by histochemical and immunohistochemical tests. Adult chondrocytes proliferated to a larger extent on the type II/I collagen membrane. MSC on type I collagen multimicrolamellar membrane proliferated and differentiated into cells that stained weakly with Safranin-O and produced type I collagen and chondroitin sulphate. Instead, on the type II/I collagen multimicrolamellar membrane, MSC differentiated in chondroblasts showing a clear immune reaction for the S-100 protein, chondroitin sulphate and type II collagen, which are typical reactions of cartilaginous cells (as described in: Grigolo B. et al., Biomaterials, 22, 2417, 2001).
Cartilage regeneration test in vivo upon implant of membranes obtained from collagen gels of the known art (comparative data) An in vivo study, similar to that described in example 16, was carried out in rabbits, using membrane 40046 (EP1307247). As it is clear when comparing the section shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 06425234.9 | Apr 2006 | EP | regional |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2007/053234 | 4/3/2007 | WO | 00 | 10/3/2008 |
