Microcarriers are commonly used as a support matrix allowing for the growth of cells to be implanted or injected into an organism. Microcarriers generally have a spherical shape with particle size depending on the final use. Microcarrier preparations can vary depending on the material and physical parameters such as shape, porosity and stiffness.

Microcarriers are especially used as a three-dimensional culture and are typically seeded with cells such as chondrocytes, mesenchymal cells (MSC), stem cells, etc, before being implanted, surgically or also by injection when the particle size is suitable to such a route of administration.

Said three-dimensional culture provide larger surface for cellular attachment, protection against mechanical stresses and simulation of in vivo environment.

Disc degeneration disease (DDD) is diagnosed in more than 40% of patients who had low back pain. Degeneration of the intervertebral disc (IVD) is a progressive and chronic process which has been recently treated on a cell-based regeneration approach. Because of the lack of sustainable sources of autologous IVD cells, the preferred cell type for cell-based therapies of DDD are mesenchymal stem cells (MSCs) that are multipotent precursor cells which maintain the replacement pool for tissue regeneration of the musculo-skeletal system, not only muscle, tendons and ligaments, but also adipose, bone and cartilage tissues. Solid and gel-like microcarriers are used in the treatment of IVD.

One major inconvenience encountered with said three-dimensional culture therapy is the enzymatic degradation of the microcarrier support which promptly intervenes from the moment of its implantation.

There is therefore a need for long lasting three-dimensional cultures comprising microcarriers.

Objects of the invention

It is an object of the invention to provide novel collagen microcarriers which are less prone to enzymatic degradation.

It is another object of the invention to provide novel collagen microcarriers which can be injected in the intervertebral disc (IVD), which are long lasting in the site of injection thus allowing a convenient therapy of DDD.

It is another object of the invention to provide a process for the preparation of novel injectable and long lasting collagen microcarriers, which offer extensive attachment sited for cells and prevent extrusion from the intervertebral disc when under pressure. It is another object of the invention to provide cell cultures comprising the above microcarriers and their use in therapy, especially in the treatment of DDD.

Description of the Figures

Figures 1 and 2 show the size ranges of the width and length of the microcarriers of the invention.

Figure 3 shows SEM micrographs of control (A-B) and samples treated with EDC/NHS(C-D) and Riboflavin (E-F). The use of glutaraldehyde (GA) as fixative (Protocol A) (A, C and E) produced cross-linking of collagen microcarriers compared to the samples treated without fixatives (Protocol B)(B, D and F).

Figure 4 shows suspensions of control samples microcarriers observed under a stereomicroscope (A-B). The use of a mixer produced small fragments of frayed appearance, very heterogeneous in shape and size. Figure 4C showed images of microcarriers examined by LM (in transmitted light) indicating that, using this approach, no morphological differences were evidenced between control and samples treated with cross-linking substances.

Figure 5 shows (A-B) Images of microcarriers examined by LM (in transmitted light) representing examples respectively of large (A) and small structures (B) present in the examined suspensions. (C) Representative examples of the iTEM software measurement, concerning the major and minor lengths of each microcarrier.

Figure 6 shows the morphology and organization of microcarriers analyzed by SEM. After the treatment with a mixer, control microcarriers were identified as small masses, mainly formed by filiform elements that intertwined to form real skeins. Some regions with more compact appearance were present (A-B, arrows). The treatment with cross-linking substances revealed a slight increase in the compacting of microcarriers, especially referring to the filiform elements of the matrix (C-F).

Figure 7 shows a cross-section of bovine intervertebral discs without and with injected microcarriers (stained) from which it can be seen that microcarriers cannot be extruded post injection.

Description of the invention

According to one of its aspects, the invention is directed to collagen microcarriers being made of irregular agglomerates of filamentous collagen elements.

According to a preferred embodiment, the invention is directed to collagen microcarriers being made of irregular agglomerates of filamentous collagen elements, wherein at least 40%, preferably at least 44%, of said microcarriers has their length and width between 100 and 500 μπι when examined under a microscope. Details of the microscope evaluation are given in the experimental section of this specification.

The term "collagen" or also "starting collagen" herein includes all types of pharmaceutically acceptable collagen, i.a. bovine and equine collagen. Equine collagen, especially equine collagen type 1, is preferred. According to a most preferred embodiment, the starting collagen is collagen presently marketed under the trademark Biopad® and disclosed in Advances In Skin & Wound Care, vol. 24 No. 5, 2011 which is herewith incorporated as reference.

The expression "irregular agglomerates" herein designates uneven masses, made of collagen filamentous fibers, which have not a well defined spherical shape.

The expression "filamentous collagen elements" herein designates collagen fibers which intertwine to form the irregular agglomerates.

According to a preferred embodiment of the invention, when examined under a microscope, the microcarriers of the invention show the following size distribution:

- at least 70 % has a major length between 100 and 1000 μιη;

- at least 44 % has a major length between 100 and 500 μιη;

- at least 70% has a minor length between 100 and 500 μιη.

According to a most preferred embodiment of the invention, when examined under a microscope, less of 5% of the major length of the microcarriers exceeds 2000 μιη. According to another most preferred embodiment of the invention, when examined under a microscope, less of 20% of the major length of the microcarriers between 1000 and 2000 μΓΐΐ . According to another most preferred embodiment of the invention, when examined under a microscope, less of 1% of the minor length of the microcarriers exceeds 1000 μιη.

According to another most preferred embodiment of the invention, when examined under a microscope, about 15% of the minor length of the microcarriers is between 500 and 1000 μΓΐΐ .

A representative example of the size ranges of the width and length of the microcarriers of the invention is depicted in the graphs of Figures 1 and 2.

The microcarriers of the invention are prepared by processing a "starting collagen". According to the invention, the "starting collagen" is preferably a solid collagen foam, advantageously made of heterologous collagen type I, derived from equine tendons. Such a collagen is commercially available, for instance under the trademark "Biopad®".

The starting collagen used to prepare the microcarriers of the invention may be native collagen or it may also being optionally cross-linked. According to a preferred embodiment, the starting collagen is cross-linked with a peptide (a growth factor or other cell-stimulating molecule)... advantageously cross-linked by the EDC/NHS (1- ethyl-3-(3-dimethylaminopropyl)carbodiimide/N-hydroxysuccini mide) method or with riboflavin/UV method. Both cross-linking methods are well known to the skilled in the art.

The collagen microcarriers of the invention can be prepared by triturating/homogenizating the starting collagen.

So, according to another of its aspects, the invention is directed to a process for the preparation of microcarriers which comprises

- homogenizing the starting collagen along with a sterile, phosphate buffered saline solution (PBS), to achieve microcarriers having the above defined appropriate size; and

- optionally, lyophilizing the microcarriers thus obtained.

According to a preferred embodiment, the starting collagen is homogenized with a rotor blade homogenizer. Details of the homogenizing process are given in the experimental section. Preferably, the ratio starting collagen/PBS are 50/1 (p/v), i.e. for instance, 250 mg of collagen/5 ml of PBS.

The microcarriers of the invention are particularly useful as three-dimensional culture and can be implanted or, preferably, injected into an organism.

The term "organism" herein indicates the organism of a mammal living being, preferably a human being.

To that purpose, the microcarriers of the invention can be seeded with any type of cells useful for implantation/injection into an organism, such as chondrocytes, mesenchymal cells (MSC), stem cells, lipocytes, etc. According to a preferred embodiment, the microcarriers of the invention are seeded with mesenchymal cells (MSC) and are injected into IVD.

According to a preferred embodiment, after the cells are seeded, they are grown according to the methods well known in the art, advantageously under hypoxic conditions (5% oxygen culture). Details of representative and preferred seeding and growth processes are given in the experimental section of this description.

According to another aspect, the invention relates to a three-dimensional cell culture comprising the microcarriers of the invention, especially MSC cultures. Said cultures are suitable to be implanted or, preferably injected, into an organism. Advantageously, said cultures are injected in IVD to treat DDD.

Thus, according to another of its aspects, the invention relates to three-dimensional cell culture comprising the microcarriers of the invention for their use in therapy, especially for their use in the treatment DDD by injection in the IVD.

In order to be injected, the cultures of the invention are Culture inventions must be in sterile solution which must provide injectability, additives are not necessarily needed but can be added if desired.

According to another of its aspects, the invention relates to a method for the treatment of DDD which comprised implanting or, preferably, injecting in the IVD of a mammal in need thereof, an effective amount of the three-dimensional culture of the invention.

The advantages provided by the microcarriers of the invention are apparent.

The specific size of the microcarriers allows for their injection, especially in the IVD. The injectability property of the microcarriers of the invention provide them the advantage of easy handling and delivery in tissues using a needle. Furthermore, microcarriers are a versatile compromise between solid and gel-like supports for cell growth, combining properties of the two scaffold types. For instance, the fact that microcarriers are suspended in a solution allows a very efficient diffusion of oxygen and nutrients to all cells, as well as efficient removal of metabolic waste products. Besides fibrous rigidity and intertwined morphology of the microcarriers, microcarriers might provide resistance to extrusion and potentially support in the loaded environment characteristic for the IVD. Under compression loading conditions, gel-like supports are less stable than solid scaffolds, due to the release of water from the material structure and direct stress on the polymeric network. In addition, injection of gel-like scaffolds in IVD face the problem of the high water pressure inside the IVD which could lead to extrusion of the hydrogel including the cells from the site of injection. The solid structure and the heterogeneous texture of microcarriers provide a strong constrain against reflux after injection and can effectively promote the retention of cell-constructs into the IVD, while the high pressure expels only the injection solution without cells. So, the properties of the microcarriers of the invention provide a real technical progress with respect of the prior art microcarriers.

It is another subject matter of the invention a three-dimensional cell culture as herein defined and claimed, characterized in that, even under high pressure, hold at least 80%, preferably 90% or more than 90% of the cells back. Indeed, as it can be seen from Figure 7, microcarriers cannot be extruded post injection.

Experimental Section

Example 1

Preparation of microcarriers

Microcarriers are prepared starting from pure, native and structurally unaltered heterologous collagen type I, derived from equine flexor tendon (Biopad, Eur ore search). 250 mg of collagen was cut in pieces of approx. 1 cm per side and transferred into a rotor blade homogenizer. 5 ml of PBS were added and the homogenization is started. One 15 steps cycle lasts 40 seconds. Rotations are as follows:

- l lOO RPM FwR

- 400 RPM RR

- 1800 RPM FwR

- 700 RPM RR

- 2300 RPM FwR

- 900 RPM RR

- 2900 RPM FwR

- 1200 RPM RR

- 3400 RPM FwR

- 1500 RPM RR

- 3700 RPM FwR

- 2000 RPM RR

- 4000 RPM FwR

FwR = Forward blade rotation; RR = Reverse blade rotation; RPM = Rotation per minute

The cycle above is repeated 3 times.

The microcarriers thus obtained are use as such or lyophilized for long term storage. Example 2

Morphological analysis of the microcarriers

Microcarriers obtained as in Example 1 were analyzed by Light Microscopy (LM), Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM).

The samples examined are listed below:

1 - Collagen (control)

2 - Collagen "cross-linked" by EDC/NHS with a growth factor

3 - Collagen "cross-linked" by riboflavin with a growth factor

The investigations were carried out i) to define the dimensional ranges of the microcarriers, ii) to study their morphological and ultrastructural characteristics, and iii) to verify the preservation of the native structure of the collagen at the end of the treatments Light Microscopy

To characterize the dimensional ranges of microcarriers, suspensions in PBS of the different samples were observed at a stereomicroscope (Olympus). For each sample, some images were captured. They were representative of areas in which microcarriers were numerous, but adequately dispersed to be able to carry out a proper interpretation, and micrographs were analyzed using the iTEM software (Olympus). Because the heterogeneity in forms, to assess the size classes, the parameters utilized were the major and minor lengths of each fragment. For measuring, 500 microcarriers, each sample, were taken into consideration.

Electron Microscopy

Aldehyde-based fixatives, including the glutaraldehyde (GA), are commonly used during samples preparation for Scanning (SEM) and Transmission (TEM) Electron Microscopy.

Since this reagent produces a bifunctional crosslinking of amino groups, it has found use in the stabilization of biomaterials based on collagen, thus reducing the enzymatic turnover, but proving highly cytotoxic. For this reason it has been replaced by other substances, such as EDC/NHS and riboflavin, used in the present study.

In order to develop a suitable protocol for samples preparation, and to study the effect of cross-linking produced by the use of GA on the examined samples (already treated with substances which represent crosslinking stabilizers), various protocols were applied for the different approaches:

Protocol A (SEM approach): the microcarriers of collagen were fixed according to conventional EM procedures, with a first phase of aldehydic fixation by immerging the samples in a solution of 3% glutaraldehyde in 0.1M cacodylate buffer, pH 7.2 for 2 hours at 4°C. Samples were then washed in 0.1M cacodylate buffer at pH 7.2, two washes of 30 minutes each at 4 °C, and post- fixed in 1% osmium tetroxide in 0.1M cacodylate buffer at pH 7.2 for 2 hours at 4 °C and then washed, two washes of 30 minutes each in distilled water. The samples were then dehydrated using increasing concentrations of ethanol from 30% to 100 %, each step of 15 minutes and dried in a Critical Point Drier (Emitech) using liquid C02. The dried samples were attached to a stub by a metallic conductive double-sided tape and were coated with gold using the evaporator Union Balzer MED 010 company. The observation was made by the Scanning Electron Microscope Jeol JSM 601 OLA.

Protocol B (SEM approach): the samples were prepared without any type of fixation and were directly dehydrated using increasing concentrations of ethanol, from 30% to 100%, each step of 15 minutes. Samples were then dried, coated with gold and observed as described for Protocol A.

Morphological analysis by SEM showed that the use of GA as fixative (Protocol A) produced cross-linking of collagen microcarriers compared to the samples treated without fixatives (Protocol B). Therefore, we proceeded to analyze by SEM only the unfixed samples (Protocol B).

In order to investigate the ultrastructural characteristics of the microcarriers by TEM, an aldehyde fixation was required. Considering SEM results, a different preparation procedure was carried out for TEM approach. The samples were fixed with 4% paraformaldehyde in cacodylate buffer 0.1M, pH 7.2 for 2 hours at 4 °C. After fixation the samples were washed in 0.1M cacodylate buffer at pH 7.2, two washes of 30 minutes each at 4 °C and immersed in a solution of 1% tannic acid in 0.1M cacodylate buffer at pH 7.2. They were then washed in the same buffer and post- fixed in 1%) osmium tetroxide in 0.1M cacodylate buffer at pH 7.2 for 2 hours at 4 °C. After two washes of 30 minutes each in distilled water, samples were dehydrated in ethanol at increasing concentration and then infiltrated in resin using solutions of increasing concentration of ethanol and resin. The samples were immersed in these solutions and kept under stirring in a rotator at 4 °C. LRWhite (London Resin) was used, a low viscosity resin which can easily penetrate into the sample. After the infiltration, the samples were immersed in pure resin inside suitable gelatin capsules and let to polymerize in oven at 50 °C for 2 days (the polymerization of this resin takes place in the absence of oxygen, and for this reason capsules with suitable sealing were used). Embedded material was then sectioned at the Reichert Ultracut ultramicrotome using a diamond knife. Sections of 1 micron were collected for LM on glass slides, stained with toluidine blue and observed under a light microscope from Zeiss Axioskop2 plus. The images were acquired using a video camera Axiocam MRC equipped by the Axiovision4 software. For TEM, sections of 60 nm were collected on formvar-coated copper grids, stained with uranyl acetate and lead citrate and observed with a Transmission Electron Microscope JEOL 1200 EX II. The images were acquired by the Olympus SIS VELETA CCD camera equipped by the iTEM software.

Analysis on the size of microcarriers

Observations under a stereomicroscope of the control and treated samples showed as the use of a mixer produced small fragments of frayed appearance, very heterogeneous in shape and size. No morphological differences were evidenced between the samples by this approach.

As reported, considering the variability of the structures, the analysis by the iTEM software was performed by measuring the length and width of each microcarriers. Data on 500 microcarriers for each sample were obtained. No differences were highlighted between the samples, as the same dimensional ranges of microcarriers were represented both in control and treated samples.

As represented in Figure 1, the minimum value concerning the major length of the microcarriers was 50,60 μιη whereas the maximum value was 2.876,50 μιη. Within these size limits, six dimensional ranges were identified: < 100 μπι / 100-500 μπι / 500-1000 μιη / 1000-1500 μιη / 1500-2000 μιη / >2000 μιη. The highest percentage of microcarriers (44%) falls in the class 100-500 μπι.

As regards the minor length of microcarriers (Figure 2), the minimum value was 48,59 μπι whereas the maximum 1.249,02 μπι. Also in this case, the majority of the microcarriers (77%) falls in the 100-500 μπι dimensional range, and were not detected microcarriers with minor length greater than 1500 μπι. Also the 1000-1500 μπι size range was represented by only two microcarriers (0%) out of 500 analyzed structures.

Morphological and ultrastructural analysis of microcarriers

Scanning Electron Microscopy (SEM) allowed a more detailed study on the morphology and organization of microcarriers. It was evidenced that, after the treatment with the mixer, collagen composing microcarriers not maintained the regular architecture in lamellae, with parallel fiber as in the scaffold of origin. In control samples, the various fragments, both those of larger dimensions, and the smallest ones, were identified as small masses, mainly formed by filiform elements that intertwined to form real skeins . he untreated microcarriers presented also some regions with more compact appearance. Morphological examination of the samples treated with cross-linking substances revealed a slight increase in the compacting of microcarriers, especially referring to the filiform elements of the matrix, compared to the control.

The ultrastructural analysis on sections of embedded samples by LM and TEM confirmed the results obtained by SEM. The examination of sections stained with toluidine blue evidenced in fact that collagen composing microcarriers lost the laminar organization and assumed a disassembled appearance forming a very irregular texture. In all the examined samples, microcarriers were made by collagen no longer assembled in lamellar structures (regions with a less intense blue), although were still present elements that have maintained the compact organization of the laminae, and which were represented by the structures of a more intense blue in LM images. These elements were represented by the filamentous structures evidenced by SEM analysis. The comparison between control and treated samples indicated that, the use of cross-linking substances produced a slightly more dense and compact matrix, especially for microcarriers treated with EDN- HS.

The observation of ultrathin sections by TEM allowed a more detailed analysis of equine collagen type I microcarriers. In control samples, both in laminar structures and in the regions where laminae were disassembled, collagen was present in the form of fibrils and well-organized fibers, with the banding typical of its native structure. These results demonstrated that the treatment with a mixer did not compromise the maintenance of the native structure of the collagen present in the scaffold of origin. With regard to the microcarriers treated with cross-linking substances, ultrastructural analysis by TEM confirmed the results previously described, highlighting how in these samples was present a higher packing of the collagen matrix compared to the control sample. In particular, the microcarriers treated with EDC- HS showed regions in which the collagen, either of fibrillar nature or organized into fibers, was strongly packaged and compact. It is possible to conclude that after the treatment of the equine collagen Type I scaffold with a mixer, microcarners were obtained in the form of fragments of frayed appearance, with different shapes and sizes. Considering the microcarners variability, the analysis was performed by measuring the lenght and width of each structures. The major length ranged from 50,60 μιη to 2.876,50 μιη. Six dimensional ranges were identified: < 100 μιη / 100-500 μιη / 500-1000 μιη / 1000-1500 μιη / 1500-2000 μπι / >2000 μπι. The highest percentage of microcarriers (44%) falls in the class 100-500 μπι. As regards the minor length of microcarriers, the minimum value was 48,59 μπι whereas the maximum 1.249,02 μπι. The majority of the microcarriers (77%) falls in the 100-500 μπι dimensional range. No differences between control and treated samples were evidenced, as the same dimensional ranges of microcarriers were detected.

SEM analysis detailed the morphological characteristics of microcarriers which were evidenced as small masses, mainly formed by intertwining filiform elements. The regular organization of the scaffold of origin was lost and collagen assumed a disassembled appearance forming a very irregular texture. LM and TEM observation confirmed such results, and although few lamellar structures were still present, collagen composing microcarriers were mainly disorganized and no longer assembled in lamellae.

Samples treated with cross-linking substances were characterized by a slight increase in the compacting of the collagen matrix compared to the control. Ultrastructural investigations confirmed SEM data, indicating that the use of EDC/NHS and Riboflavin originated a slight cross-linking of the collagen matrix, especially for microcarriers treated with EDC/NHS, that resulted more packaged and compact if compared to Riboflavin.

In addition, in all the examined samples, collagen was present in the form of fibrils and well-organized fibers, with the banding typical of its native structure, demonstrating that the treatment with a mixer firstly, and secondly the treatments with cross-linking substances did not compromise the maintenance of the native structure of the collagen present in the scaffold of origin.

Example 3 MSCs Isolation and Culture

Fresh bone marrow (BM) samples (20 ml) were obtained from the iliac crest of the donors during surgery after informed consent and approval by the ethics committee of canton Lucerne. MSCs were isolated from BM of seven donors and split in two groups: cells from four donors (average age: 56 ± 9 years) were used to compare MSC chondrogenesis between passage 0 (PO) and 2 (P2), while cells from three donors (average age: 45 ± 1 year) were used to assess the influence of different oxygen levels and type of cultures. Cells from this group were expanded for three passages (P3) in culture.

The BM aspirates were diluted in 15 ml 3.8% sodium citrate and 20 ml phosphate buffered saline (PBS) and then filtered through a 100 μπι cell strainer to remove clots(Falcon, BD Bioscience). Mononuclear cells were separated by Ficoll gradient centrifugation (density 1.077 g/mL; GE Healthcare) in a Leucosep tube (Greiner) at 800 g for 15 minutes, washed with PBS, centrifuged again at 210 g for 10 min, re- suspended in PBS and counted using trypan blue dye in a single use Neubauer chamber (C-Chip Typ Neubauer, Zeiss). Cells were plated in tissue culture flasks (TPP) in non-hematopoietic (NH) Stem Cell Media (Miltenyi) at 37°C in a humid atmosphere containing 5% C0 ₂ . After 2 days, non-adherent cells were discarded, whereas adherent cells were cultured in growing medium consisting in DMEM/F12 + GlutaMAX, supplemented with 10% foetal bovine serum (FBS), (100 units/mL) penicillin / (100 mg/mL) streptomycin, 2.5 ng/ml Amphotericin B (all GIBCO) and 5 ng/ml recombinant basic Fibroblast Growth Factor (bFGF, Peprotech) with medium changed 3 times a week.

Example 4

Preparation of three-dimensional cell cultures

Individual microcarrier particles were separated in a 8-wells chamber slide (Lab- Tekll), washed with PBS and fixed by pure methanol (Applichem). After three washing cycles, cell-microcarrier particles were stained with 0.01 % SYBR green (Invitrogen) solution (1 : 10'000 in dH ₂ 0) and examined by fluorescence microscopy (Fluorescence Microscope U-TB190, Olympus). Comparison of different cell densities showed an optimal ratio between cells and collagen of 4xl0 ⁴ cells/5 mg collagen, which was used in all experiments involving cells at P2 and P3. Freshly isolated cells from bone marrow (PO) were seeded on microcarriers at the density of 2xl0 ⁶ cells/5 mg collagen.

Dynamic cultures were performed using a tube roller mixer placed inside the incubator. Falcon tubes containing MSC-microcarriers were submitted to constant rotation around the longitudinal axis. In proliferation and differentiation assays, the tubes rotated 3 times per minute.

To induce chondrogenesis, MSC-microcarriers were pre-cultured for seven days in growing medium to allow optimal spreading of cells on scaffolds, and then maintained for 21 days in chondrogenic medium. Chondrogenic medium consisted in DMEM/F12 + GlutaMAX, supplemented with 2,5% FBS, 40 ng/mL dexamethasone (Applichem), 50 μg/mL ascorbate-2-phosphate (Sigma), 50 μg/mL L-proline (Sigma), 100 U/mL penicillin - 100 mg/mL streptomycin, 2.5 ng/ml Amphotericin B, IX Insulin (10 μg/ml) - Transferrin (5.5 μg/ml) - Selenium (0.67 ng/ml) -X Supplement (ITS, Gibco), and 10 ng/ml a growth factor-. All media was replaced three times per week.

Quantitative determination of cell number on microcarriers was carried out by PrestoBlueTM (Invitrogen) cell viability indicator. PrestoBlueTM reagent was used according to the manufacturer's instructions and the following protocol. PrestoBlueTM reagent was added to samples (1 : 10 dilution), incubated for 1 hour at 37°C and florescence intensity was quantified by a multimode detector (DTX 880, Beckman Coulter) at 595 nm. Results referred to a standard curve, previously prepared with cells derived from the same donor.

Immunohistochemical analysis was used to detect aggrecan, collagen type I and type II accumulation. Constructs were harvested at 21 days of culture in chondrogenic medium, embedded in Neg-50 compound for 30 minutes, frozen at -80°C and subsequently sectioned at 20 μπι using a cryostat (CM 1850, Leica).

Endogenous peroxidase was quenched by 3% H ₂ O ₂ in PBS at room temperature, and washed with PBS. Before incubation with anti-collagen type II antibody, sections were pre-treated with 2500 U/mL hyaluronidase (Sigma-Aldrich) in PBS at 37°C, while sections for immunodetection with anti-aggrecan antibody were pre-treated with chondroitinase ABC (0.25U/ml, Sigma- Aldrich) in 0.1M Tris 0.03M acetate buffer pH 6.5, at 37°C. Non-specific background was blocked with PBS containing 1 mg/mL BSA, 10% FBS and 0.1% Triton (Applichem) followed by overnight incubation at 4°C with monoclonal mouse antibodies against collagen type I (1 :20; M-38, Development Studies Hybridoma Bank) -collagen type II (1 :20; II-II6B3, Development Studies Hybridoma Bank) and aggrecan (1 : 10Ό00; AHP0022, Biosource) in blocking solution. After washing with PBS, sections were incubated with a secondary biotinylated goat anti-mouse antibody (1 :200; B0529, Sigma), and then with streptavidin-horseradish peroxidase (1 :200; S2438, Sigma) at room temperature. Aggrecan, collagen type I and II were visualized by reaction with 0.075% solution of 3-amino-9-ethylcarbazole (AEC, Applichem) in 0.01% H ₂ 0 ₂ . Sections were mounted with 70% Glycerol (Applichem) and examined by light microscopy.

Histological detection of sulphated glycosaminoglycan (GAG) accumulation was carried out by Alcian Blue staining. Sections were stained overnight with 0.4% alcian blue (Fluka) dissolved in 0.01% H ₂ S0 ₄ and 0.5M guanidine hydrochloride (Fluka). Next, sections were washed for 30 minutes in 40% DMSO and 0.05M MgCl ₂ . Finally, sections were mounted with 70% glycerol and examined by light microscopy. Cell distribution on microcarriers was visualised by Giemsa (Applichem) staining. Sections were firstly fixed in pure methanol, washed in PBS and incubated with Geimsa staining solution. Prior examination by light microscopy, excess of Giemsa solution was removed by abundant washes in dH20.

Total RNA was isolated from MSC-microcarrier constructs after 21 days in culture. Constructs were homogenized using Dispomix in RNA lysis buffer of Aurum Total Mini Kit (Bio-Rad), following the manufacturer's instructions with the modification of adding 2 μΐ polyacryl carrier (LucernaChem) in the kit lysis buffer. cDNA was prepared using VILO cDNA Synthesis Kit (Invitrogen).

Real-time (RT)-PCR reactions were carried out in triplicates with the primers listed in Table 1 at a concentration of 250 nM, 5 μΐ cDNA template, and IQ SYBR Green Supermix (Bio Rad). Specific products were amplified by a quantitative PCR system (CFX96™ Real Time System, BioRad). Real-time PCR was carried out with the following settings: denaturation 95°C - 3 minutes (1 cycle), 95°C - 15 sec, 64°C-20 sec and 72°C-20 sec (40 amplification cycles). PCR reactions were carried out in triplicates in a final volume of 25 μΐ in 96-well plates (Bio Rad). Melting curve analysis was performed after each reaction. Gene expression differences were determined using the 2 ^~AACt method and the results were normalized to the expression of GAPDH.

Data were expressed as the mean ± SD. Non-parametric Mann- Whitney- Wilcoxon U test for dependent variables was used to determine significant differences between samples, as a normal distribution of the data could not be guaranteed in this data set. For all tests, p<0.05 was considered significant. Data analysis was performed with SPSS 14.0 for Windows (SPSS Inc.).

MSC number and morphology in microcarriers are positively influenced by hypoxia After 21 days in chondrogenic culture, MSC-microcarriers fused together forming a single construct, as opposite to day 0 where MSC-microcarriers were unbound. Depending on culture conditions, constructs were characterized by diverse sizes: compared to 21% oxygen culture (normoxic control), constructs at 10%, 5% and 2% oxygen levels were larger. Construct in normoxia had a size of approximately 2 mm diameter, against the 3.5-4 mm of the constructs in hypoxia.

This difference in sizes correlated to the number of cell retained within constructs. At the beginning of the differentiation assay (after one week) cell proliferation was significantly higher in the microcarriers cultured in hypoxic environment. Compared to normoxia (-150Ό00 cells), the increase in cell number corresponded to 152% (-230Ό00 cells) in 10% oxygen (p<0.05), 179% (-260Ό00 cells) in 5% oxygen(p<0.05) and 134% (-200Ό00 cells) in 2% (p<0.05) oxygen. Similarly, at the end of differentiation (three weeks), the number of cells per culture was significantly higher under hypoxic conditions compared to normoxic, and the relative amount of cells was equivalent to the beginning of the differentiation assay.

Hypoxia upregulates collagen type II expression in MSC-microcarriers

After 21 days in culture, quantitative real-time PCR was used to assess expression of the most important extracellular matrix molecules of the IVD, namely collagen type I (COL1) and type II (COL2), and aggrecan (ACAN) by MSCs on microcarriers. Expression of COL1 and AC AN by MSCs in 5% oxygen culture was comparable to other culture conditions and results showed no differences. Oxygen at 5% and 2% levels promoted the expression of COL2, especially in 2% hypoxia where the expression increased twofold, while COL2 expression was almost undetectable in normoxia and 10% oxygen groups (p<0.05),

Cells were homogenously spread in the constructs under both normoxic and hypoxic (5%) conditions as shown by Giemsa staining. MSC-constructs in 5% oxygen culture accumulated more proteoglycan compared to normoxia as evaluated by histological staining of construct sections with alcian blue for proteoglycan (GAG) accumulation. The localization of collagen type II, aggrecan and collagen type I was determined by immunostaining, which showed that in MSC-constructs differentiated in 5% hypoxia, collagen type II and aggrecan accumulation was undoubtedly higher than in normoxia, while the amount of collagen type I staining showed similar intensity between the two groups. The signal intensity of collagen type II was higher compared to that of aggrecan.

Dynamic cultures promote cell proliferation on microcarriers. but inhibit chondrogenesis

Dynamic culture had beneficial impact on cell proliferation, before differentiation assay. After one week, the number of cells retained in MSC-microcarriers cultured in 5% hypoxia was 178% (-250Ό00 cells, p<0.05) compared to normoxia (-150Ό00 cells), and cell number was further increased to 227% (-340Ό00 cells, p<0.05) in dynamic culture. Effects of dynamic cultures on chondrogenic differentiation of MSC-microcarrier in 5% hypoxia culture were compared to static cultures in both 5% hypoxia and normoxia. After 21 days of differentiation, MSC-microcarrier aggregates had dissimilar sizes, depending on culture condition, for example under static condition, aggregate in hypoxia was double size compared to normoxia (4 mm Vs. 2 mm diameter). However when MSC-microcarriers were differentiated under dynamic conditions, the increase in size seen in static hypoxic culture was hindered and the aggregates had sizes similar to those seen in normoxia.

On the other hand, dynamic culture had a beneficial impact on cell proliferation, before differentiation assay. After one week, the number of cells retained in MSC- microcarriers cultured in 5% hypoxia was 178% (-250Ό00 cells, p<0.05) compared to normoxia (-150Ό00 cells), and cell number was further increased to 227% (-340Ό00 cells, p<0.05) in dynamic culture. At the end of differentiation, cell number was unchanged in the static cultures compared to the beginning, while in dynamic culture cell number dropped fourfold - down to 44% (-80Ό00 cells, p<0.05) - reflecting the differences observed in construct sizes.

Dynamic cultures affected also gene expression of MSC-microcarriers at the end of differentiation. Under static condition, hypoxia strongly promoted expression of COL2, as opposite to normoxia (p<0.05), but the advantageous effect of hypoxia was reversed under dynamic conditions where COL2 expression was again reduced (p<0.05). Expression of ACAN and COL1 showed no changes between static and dynamic cultures.

Comparison of freshly isolated P0) and early passages P2) MSC chondrogenic potential in normoxia and hypoxia

Cells proliferation rate of freshly isolated mononuclear cells from bone marrow extracts in normoxia and 5% hypoxia was compared between microcarriers and monolayer cultures for 14 days. In monolayer, the number of MSCs was significantly higher compared to microcarriers, independently of oxygen levels. In normoxia, the number of cells grown on microcarriers was only a quarter (-40' 000 cells) of the number of cells in monolayer (-190Ό00 cells, p<0.05). Hypoxia increased cell number in both type of cultures but did not close the gap: cells in hypoxia were up to 38% (-70' 000 cells,) and in monolayer up to 183% (-350Ό00 cells, p<0.05).

After one week in culture, MSC at P0 and P2 were seeded on microcarriers and differentiated for 21 days in normoxia and 5% hypoxia. Then at one week of differentiation, the number of MSCs at P2 (-100Ό00 cells) was threefold the number of cells at P0 (30Ό00 cells, p<0.05). Hypoxia increased non-significantly cell number in both type of cultures up to 162% at P0 (55Ό00 cells) and 130% at P2 (130Ό00 cells). At the end of differentiation, cell number at P0 decreased to approximately one third of the amount of cells counted at the beginning, in both normoxia (-10Ό00 cells) and 5% hypoxia (20Ό00 cells). On the other hand, cells number at P2 slightly increased after differentiation of a factor of approximately 140% (-130Ό00 cells in normoxia Vs. -180Ό00 cells in 5% hypoxia) compared to the beginning of the assay.

Gene expression analysis of MSCs after differentiation showed that cells at PO were unable to express the expected amounts of COL2 while the cells at P2 were appropriate for the task. Hypoxia increased COL2 mRNA expression up to 50-fold (p<0.05) compared to normoxia. Similar gene expression pattern was observed for ACAN: in hypoxia, expression of ACAN by cells at PO was not detectable, while at P2 it was increased almost threefold compared to normoxia (but without reaching statistical significance). In normoxia, ACAN expression at PO was 40% lower compared to P2 (p<0.05). COL1 expression was identical in normoxia and hypoxia samples at P2. In comparison to P2, cells at P0 expressed more COL1, especially in normoxia where the expression was almost threefold higher (p<0.05).