WO2006131722A1

WO2006131722A1 - Stem cells and isolation method

Info

Publication number: WO2006131722A1
Application number: PCT/GB2006/002066
Authority: WO
Inventors: Brian Mark Thomson
Original assignee: Smith & Nephew, Plc
Priority date: 2005-06-09
Filing date: 2006-06-06
Publication date: 2006-12-14
Also published as: GB0511723D0

Abstract

Isolated human placental cells that are capable of forming cartilage and methods for isolating and purifying chondrogenic mesenchymal progenitor cells (MPC) from human placenta without the need for cell-separation reagents such as antibodies, lectins, fluorochromes or magnetic beads.

Description

Stem cells and isolation method

The invention relates to isolated human placental stem cells and methods of isolating same. In particular the invention relates to methods for isolating and purifying chondrogenic mesenchymal progenitor cells (MPC) from human placenta without the need for cell-separation reagents (e.g. antibodies, lectins, fluorochromes, magnetic beads etc.). The resulting isolated MPC can be used in the treatment of damaged connective tissue, e.g. cartilage. Early research into tissue engineering demonstrated that three- dimensional pieces of tissue, (e.g. cartilage), could be constructed from non-human mammalian cells and biocompatible polymer scaffolds. Unfortunately, difficulties associated with obtaining sufficient numbers of functionally active human cells prevented these laboratory findings from being converted into clinically useful or commercially viable therapies.

It was recognised that if cartilage tissue engineering were ever to become a practical reality, a new and abundant source of mature human chondrocytes would have to be found. Theoretical considerations suggested that MPCs, for example, the ancestral precursor cells for bone, cartilage and fat cells, could provide the basis for this required cell source.

However, practical experience showed that the chondrogenic activity of MPC cultures tended to be incomplete, variable and/or inhomogeneous. Observations by Lennon, (Developmental

Dynamics 219:50-62 (2000)), suggested that this could be due to the highly impure nature of even supposedly 'homogeneous' human mesenchymal isolates. We therefore sought methods to improve the yield and purity of human MPC isolation and purification techniques.

There is therefore a need to develop methods to isolate large numbers of human MPC from a readily available human tissue source at sufficient levels of purity to support in vitro chondrogenesis.

The existence of MPC in perivascular tissue such as placenta has been described in the published literature. Unfortunately MPC are only present as a small percentage of the total number of cells, necessitating the use of cell purification techniques. Existing MPC purification methods utilise reagents that bind to specific features on the cells' surfaces, (such as antibodies, bacteriophages, lectins, fluorochromes, magnetic particles and affinity substrates). Examples include the antibodies STRO-1 , CD90 and CD105.

However, no one single cell surface marker exclusively and unambiguously distinguishes MPC from all other cell types under all circumstances. Furthermore, the profile of markers expressed by the MPC changes when (i) the cells are removed from their normal tissue environment; (ii) when the cells are cultured in either monolayer or three dimensional environments and (iii) when the cells are 'trypsinised' in order to move them between culture vessels or prepare them for fluorescence activated cell sorting (FACS). It is therefore necessary to assess a panel of markers in order to identify the required MPC. Quantifying the expression of these panels of markers is technically demanding and involves treating the cells with sophisticated mixtures of biochemical reagents. This is undesirable because any method that binds reagents (for example antibodies and fluorochromes) to features on the cells' surface inevitably alters the nature of the cells and thereby introduces regulatory/ quality control concerns over any subsequent clinical application of the isolated cells. Furthermore the use of these cell labelling reagents creates commercial difficulties because of their expense and, in some instances, the 'batch variable' quality of the reagents. Thus, whilst the simultaneous analysis of multiple cell surface markers can be used in the laboratory to identify and purify MPC, the complexity, cost, commercial barriers and regulatory issues associated with these methods prevents them from being used on an industrial scale.

It is an aim of the present invention to provide an isolated population of cells from human placental tissue that are capable of foming cartilage tissue. Furthermore it is an aim of the present invention to provide a method of isolating cells from human placental tissue that does not involve the use of biochemical reagents that bind onto the cells to facilitate their isolation. Accordingly, to the present invention there is provided an isolated population of cells derived from full term human placental tissue capable of forming cartilage.

The cells are typically adherent.

The cells are preferably enriched for mesenchymal progenitor cells. Enriched is defined as a higher proportion of mesenchymal progenitor cells than is present in native placental tissue.

The cells are purified according to their physical properties and not upon the binding of exogenous reagents to specific features on the surface of the required cells. Suitable intrinsic properties upon which to purify the cells include size, granularity and autofluorescence. These properties are readily measurable, persist during cell isolation from the tissue, trypsinisation, passage and monolayer culture, and can be detected without the need for exogenous cell separation reagents.

According to the present invention there is provided a method for isolating an enriched population of mesenchymal progenitor cells from human placenta, comprising the steps of: a) Dissociating the placental tissue into a mixture of (i) individual cells and (ii) fragments of vascular tissue that consist of vessel-wall cells and residual extracellular matrix b) Culturing the dissociated placental tissue to isolate adherent cells c) Sorting the adherent cells according to granularity and autofluorescence d) Isolating adherent cells of below average autofluorescence and below average granularity compared with unsorted human placental cells

According to the present invention there is further provided a method for preparing purified human mesenchymal progenitor cells from full term human placenta by isolating mixed populations of adherent cells from the placental tissue and sorting these adherent cells into sub-fractions using a FACS machine according to the cells' granularity and autofluorescence, such that one fraction (the 'MPC enriched cells') has an increased percentage of cells displaying characteristic MPC properties, (such as the ability to form colonies of > 32 cells, cartilage and bone), whilst the remaining fraction(s) (the residual cells) are relatively depleted of cells with typical MPC properties.

The MPC enriched fraction(s) are therefore aptly defined as a population of nucleated human placental cells with the following two properties when analysed by FACS:

(i) An autofluorescence intensity (λex = 488nm; λem = 655~685nm) less than or equal to that of the 5^th most intense fluorescence intensity calibration bead supplied by DakoCytomation kit K0112. This level of autofluorescence corresponds to those placental cells with a lower than average level of autofluorescence at the above wavelengths when compared with an unsorted population of adherent human placental cells. (ii) A 'side scatter' (or granularity) property substantially equivalent to that of a 6μm reference bead (e.g. Molecular Probes; F-13838). This level of granularity corresponds to those cells with lower than average granularity when compared with unsorted, adherent human placental cells.

An MPC enriched fraction of full term adherent human placental cells (termed Ao cells) has both of the above distinguishing properties and can be of any size.

The cells may then be further purified on the basis of size. Isolation of cells of above average size yields a population of cells further enriched for MPC cells.

A more highly MPC-enriched cell fraction (termed Aι_arge cells) consists of cells with both the above properties and a diameter of greater than or equal to 18 μm when compared with reference beads from, for example, Duke Scientific Corporation when assessed by forward scatter on a FACS machine. This cell fraction consists of the cells that have above average diameters when compared with the initial, unsorted adherent placental cell population, (i.e. Aι_arg_e cells have diameters greater than or equal to the 50^th percentile of the diameters of the unsorted placental cells)

A yet more highly MPC-enriched fraction (termed AR5 cells) consists of cells with both the above properties and diameters between 18 and 30μm when judged by forward light scattering and compared with 10, 20 and 30μm reference beads (Duke Scientific Corporation). The diameter of this AR5 cell fraction ranges from the 50^th - 65^th percentile of the diameter range of the unsorted adherent placental cell population.

The isolated populations of MPC enriched cells described above are suitable for use in the repair of connective tissue defects (e.g. cartilage defects) or for the production of tissue-engineered grafts (e.g. tissue engineered cartilage grafts).

The invention will now be described, by way of example, with reference to the following description and figures.

Figure 1. Monolayer culture of mixed, adherent human placental cells.

Figure 2. Autofluorescence of placental cell isolates.

Figure 3. Classification of placental cell isolates by particle size.

Figure 4. Classification of placental cell isolates by particle granularity.

Figure 5. Detailed relationship between placental cell autofluorescence, granularity and cell diameter. Figure 6. Identification and isolation of the MPC enriched fractions of adherent human placental cells.

Figure 7. Quantification of the autofluorescence of the MPC-enriched placental cell fractions using fluorescence calibration beads.

Figure 8. Measurement of the size of the MPC-enriched placental cell fraction using size calibration beads.

Figure 9. Quantification of the 'granularity' of MPC-enriched placental cell populations.

Figure 10. Colony formation by A_Larg_e placental cells grown in monolayer culture.

FigureH . Colony formation by sorted and unsorted placental cells.

Figure 12. Cartilage formation by A_Large placental cells.

Figure 13. Production of cartilage-specific type Il collagen by pellet cultures of A_Large placental cells and unsorted placental cells.

Figure 14. Bone nodule formation by placental cells grown in osteogenic medium.

Isolation of mixed populations of adherent human placental cells.

A fresh human placenta was dissected to remove the umbilicus, large vessels and outer membranes and the remaining 'red tissue' finely minced. The minced tissue was suspended in a sterile mixture of 500 ml of low glucose DMEM containing penicillin, streptomycin, non-essential amino acids and L-glutamine plus 10% foetal calf serum; 0.05% w/v collagenase (Worthington type II) and 0.05% w/v dispase in a 2 litre Duran bottle and incubated on an orbital mixer for 16 hours at 37°C. The digested tissue and released cells were aliquoted into four equal portions and sedimented through 3 changes of PBS (3 x 2 litres x > 45 minutes). The washed placental digest was centrifuged, (Megafuge 1.0; 1500rpm; 5 minutes) and the resulting pellets re-suspended in 1 litre of low glucose DMEM containing penicillin, streptomycin, non-essential amino acids and L-glutamine plus 10% foetal calf serum. The resulting tissue suspension was aliquoted to 10 T175 flasks and the cultures incubated in a humidified atmosphere of 5% CO₂ until the tissue fragments adhered and cells migrated onto the plastic. Figure 1 shows a monolayer culture of mixed, adherent human placental cells. Two distinct cell populations can be discerned. In one population, (the MPC-depleted 'residual' cells; outlined arrows), the nucleus (white) is surrounded by a cytoplasm that contains numerous autofluorescent granules (grey speckles). In the other cell population, (the MPC enriched fraction; white arrows), the nucleus (white) is surrounded by a cytoplasm that is devoid of autofluorescent granules. An MPC enriched cell fraction can be obtained by collecting those cells that lack the autofluorescent cytoplasmic granules.

Preparation of placental cells prior to FACS sorting

FACS sorting buffer was prepared by mixing PBS (200ml); 1g of BSA and 800μl of 0.5 M EDTA and filter sterilising.

A T175 flask of primary placental digest was rinsed with PBS, the adherent cells trypsinised and transferred to a fresh T175 flask containing LDMEM/10%FCS (75 ml). Cultures were incubated for 16 hours in a humidified atmosphere of 5%CO2. (This step disperses the cell 'clumps' that form during the early stages of cell isolation and gives a more usable cell monolayer).

The cells were then rinsed in PBS, re-trypsinised (to give an unclumped single cell suspension), centrifuged (1500 rpm; 5 minutes), and the cell pellet resuspended in 5 ml FACS sorting buffer. The resulting cell suspension was passed through a fine syringe needle and stored on ice in a sterile polypropylene tube.

FACS sorting of mixed placental cell suspensions

Cells were analysed using a fluorescence-activated cell sorter that simultaneously evaluated each cell according to three criteria 1. Autofluorescence Unsorted human placental cells were found to have a reproducible distribution of autofluorescent intensities (see figure 1). This autofluorescence was an intrinsic property of the cells and did not require any exogenous fluorochromes or additional reagents. Autofluorescence was detectable across a wide range of wavelengths, but was usually assessed with λ > 600nm; (preferably λ_em = 655 - 685 nm). For routine measurements, λ_ex = 488nm, 20OmW and λ_em = 670-30 band-pass filter. Placental cells were classified as having either low autofluorescence, mid autofluorescence or high autofluorescence.

Figure 2 shows an example of autofluorescence of placental cell isolates. Placental cell autofluorescence (λex = 488nm; λem = 655 - 685 nm) was measured using a FACS machine. Particles were classified as having low, medium or high autofluorescence as shown. The fluorescence intensity of the MPC-enriched fractions is highlighted (light grey).

Figure 7 shows quantification of the autofluorescence of the MPC- enriched placental cell fractions using fluorescence calibration beads. The auto fluorescence intensity of the MPC-enriched placental cell fractions was compared to that of a commercially available set of calibration beads with 8 different fluorescent intensities (hatched peaks; DakoCytomation; K0112). The MPC- enriched fraction of placental cells (grey) had an autofluorescence (λex ~ 488nm; λem = 655 - 685nm) less than or equal to that of the fluorescence shown by the fifth brightest calibration beads.

2. Forward light scatter (FSC). The extent to which laser light passes through the sample in a FACS machine is a measure of the cells' size. Cells were clearly distinguishable from particles of debris

(produced during cell preparation) and were shown to have a range of sizes.

Figure 3 shows the classification of placental cell isolates according to their diameters (as assessed by their forward light scattering properties). Particles were classified as debris, small cells, medium cells or large cells. The size of the MPC enriched fraction, (Aι__ar_ge cells), is indicated by the arrow. The size range of the most highly M PC-enriched fraction (AR5 cells) is highlighted light grey. Figure 8 shows measurement of the size of the MPC-enriched placental cell fraction using size calibration beads. The size distribution of the entire population of placental cells, (as assessed by forward light scatter), is shown by the single line (u). The size of 10, 20 and 30μm calibration beads (Duke Scientific Corporation) is indicated by the diagonally hatched peaks. The MPC-enriched cell fraction hereinafter termed Aι_arge cells have diameters greater than or equal to that of the 20μm diameter calibration beads. The most highly enriched fraction, hereinafter termed AR5 cells have a size range greater than or equal to that of the 20μm diameter calibration beads but less than that of the 30μm diameter calibration beads (grey region labelled hMSC enriched fraction). The most MPC- enriched population has an estimated size range of 18 - 27μm diameter.

3. Sideways light scatter (SSC). The extent to which the laser light is reflected from the specimen perpendicular to the incident illumination is a measure of the cells' granularity. Figures 4, 5, 6 and 9 show that cells have a range of granularities and that cells with low autofluorescence have below average granularities. Figure 4 shows classification of placental cell isolates by particle granularity. Placental cell isolates were classified using a FACS machine according to their sideways light scattering properties

(SSC); i.e. granularity. Particles were classified as either debris, or as cells with low, medium or high granularity. The SSC profile of the MPC-enriched cell fraction is highlighted (light grey). Figure 9 shows quantification of the granularity of MPC-enriched placental cell populations. The granularity of adherent human placental cells was quantified by comparing the sideways light scattering properties of the cells with that of standard reference beads (for example Molecular Probes; F-13838) using a FACS machine. The side scatter values of the unsorted adherent placental cell population is shown by the single line 'u'. The MPC enriched fraction is shown in the grey-shaded peak labelled hMSC enriched fraction. These cells had granularities (i.e. side scatter values) in the lower 50% of the overall population. This corresponds to side scatter properties equivalent to more than or equal to that of the 6μm reference beads but less than the 15μm reference beads. Interpolation of the results indicates that the preferred MPC enriched placental cell fraction has an estimated side scatter value less than or equal to approximately 8μm in diameter.

Detailed analysis of the relationship between adherent human placental cell autofluorescence intensity, granularity (SSC) and cell size (FSC)

The detailed relationship between placental cell autofluorescence intensity, granularity (SSC) and cell size (FLS) underpins this invention and is shown in figure 5.

The upper panel of figure 5i shows a graph of the number of particles (i.e. cells plus debris) in a placental cell suspension versus autofluorescence intensity of the particles. The lower panel in figure 5i shows a scattergram of particle granularity versus particle size for the same placental cell sample. Two clearly distinct classes of particle are visible in the lower panel, namely 'debris', (lower left; in which the particles have low size and low to medium granularity) and 'cells' (upper right; in which the particles have a greater size and granularity than the debris particles). The nature of these two classes of particles was verified by light microscopy.

The upper panel of fig 5i shows a user-defined region of interest, Ri₁ that selects a sub-group of particles with a specific range of autofluorescent intensities. It is possible to use the FACS software to display where the particles in Ri (i.e. particles with a specified range of autofluorescent intensities) fall within the scattergram of granularity versus size shown in the corresponding lower panel. The pairs of upper and lower panels shown in figures 5ii - 5x indicate that the relationship between a particles autofluorescence and its position on the size versus granularity scattergram is non-random.

The upper panel of fig 5ii indicates that particles with the lowest autofluorescent intensities (labelled 'd') are contained within the 'debris' peak in lower panel 5ii (labelled 'd'). Shifting the region of interest Ri to the next highest band of autofluorescence, (see upper panel 5iii), highlights a population of particles ('a-i') that consists mostly of debris but which contains a small number of cells with very low granularity when compared with the overall placental cell population (labelled 'a-t' in the lower panel of figure 5iii).

The next highest band of autofluorescent intensities (see upper panel 5iv) identifies a key population of cells (labelled a₂) that combine a lower than average autofluorescence (upper panel 5iv) with a lower than average granularity (lower panel 5iv).

Similarly, the cells highlighted in the upper and lower panels of figure 5v show a population of cells (labelled az) with below average autofluorescent intensity and granularity.

In contrast the cells highlighted in figures 5vi, 5vii and δviii (i.e. those labelled bi, b₂ and bz) combine a moderate autofluorescence intensity with a moderate to high granularity. In general, these 'B' cells tend to have a greater than average cell diameter, (compare lower panels of 5vii and 5ix)

The remaining cells, (i.e. those highlighted in figures 5ix and 5x), have the highest autofluorescence, a high granularity and a smaller than average cell size (¹C cells; see lower panel 5ix).

The FACS machine allows cells to be 'gated' according to multiple parameters.

Placental cells were thereby divided into 3 groups:

Population Ao: Non-debris particles (i.e. particles outside R₂) with low autofluorescence, low granularity and small to large size.

Population B₀: Non-debris particles (i.e. particles outside R₂) with moderate autofluorescence and high granularity.

Population Co: Non-debris particles (i.e. particles outside R₂) with high autofluorescence, high granularity and below average size. Determination of the relative MPC content of population A₀, B₀ and C₀ cells using colony forming assays

The MPC content of these three cell classes was assayed by monitoring their colony forming potential. Figure 10 shows a colony formed by Aι_arge placental cells grown in monolayer culture. Colony formation is defined as the proliferation of a single cell to produce an aggregate of > 32 daughter cells.

Figure 11 shows the number of colonies formed by placental cells incubated in sparse monolayer culture for 10 days. The number of colonies formed was then counted, (i) shows colony formation by A₀ placental cells, (i.e. cells with low autofluorescence and low granularity); (ii) shows colony formation by Bo plus Co cells (i.e. cells with moderate to high autofluorescence and high granularity); and (iii) shows colony formation by unsorted placental cells. The Ao cells had significantly higher levels of colony forming activity than the unsorted placental cells, indicating that they were enriched for MPC compared to the starting cell population. In contrast, the B₀ and C₀ cells had significantly less colony forming potential than the unsorted cells, indicating that this fraction was depleted of MPC compared to the starting cell population.

Further subdivision of the class Ao placental cell population in order to increase the purity of the isolated MPC

Ao cells, (i.e. the putative human MPC containing placental cell fraction), were subdivided by size into 2 sub-fractions termed, Αι_arge' and 'Asmaii' 0-β. low autofluorescence, low granularity cells with above or below average forward scatter values respectively). The MPC were found to be more highly enriched within the TVarge' fraction. Further sub-division of the 'A_Large' fraction by size into three sub-fractions indicated that the human MPC were most highly enriched within the smallest third of the 'ALarge cell fraction (i.e. those cells with diameters judged by forward scatter to lie between the 50^th and 65^th percentile of the overall range of placental cell diameters). These cells were termed AR5 cells. Some MPC were also found in the middle third of the 'ALarge' fraction, (termed AR2 cells; i.e. those cells with diameters judged by FLS to lie between the 65^th and 80^th percentile of the overall range of placental cell diameters).

Figure 6 shows the identification and isolation of the MPC enriched fractions of adherent human placental cells by simultaneously assessing their autofluorescence (λex = 488nm; λem = 655 - 685 nm), size (FSC) and granularity (SSC). The MPC enriched fraction A₀ cells consists of nucleated placental cells which combine low autofluorescence (values in region R1 of the upper panel) with low granularity scores (values in R10 and R11 of the lower panel). The more highly MPC enriched fraction (ALarge cells) consists of nucleated placental cells which combine low autofluorescence (values in region R1 of the upper panel) with low granularity scores (SSC) and above average diameters (FSC) (contained within boxes A-i, A₂ and A₃ of the lower panel). The most highly MPC enriched fraction (AR5 cells) consists of nucleated placental cells which combine low autofluorescence (values in region R1 of the upper panel) with low granularity scores (SSC) and diameters (FSC) contained within box Ai of the lower panel.

Chondrogenesis by purified human placenta derived MPC and mixed placental cells

Mixed human placental cells, FACS-sorted placental ALarge cells and placental Asm_aii cells were trypsinised, counted and 2x10⁵ cells were added to sterile polypropylene tubes and centrifuged at 1500rpm/10 minutes to pellet the cells. Supernatant was discarded and 1ml of either standard medium (high glucose DMEM/10% FCS) or 'chondrogenic medium' (DMEM containing; 5ng/ml transforming growth factor beta 1 (TGF-β1), 10^"7 dexamethasone, 50μg/ml AA, 40μg/ml proline and 50μg/ml ITS+ Premix), was added to the tubes in duplicate. Cells were incubated with loosened tops and media changes were carried out every 4 days thereafter. Pellets were harvested at days 11 to 29, paraffin wax embedded and sectioned (5μm) prior to safranin O staining. Human cartilage tissue was also paraffin wax embedded, sectioned and stained to act as a positive control. All specimens were mounted using DPX histomount, observed microscopically and images taken. Results (figure 12c) show that no recoverable pellets were formed when mixed placental cells were incubated in standard medium for 29 days. Mixed placental cells cultured in 'chondrogenic medium' formed recoverable pellets but did not show any evidence of chondrogenesis after 29 days (figure 12d).

In contrast, Aι_arge cells formed pellets, but did not undergo chondrogenesis when cultured in standard medium for 29 days (figure 12a). When grown in chondrogenic medium for 29 days however, the A_Large cells showed uniform evidence of chondrogenesis in all but the periphery of the pellet cultures (white staining; figure 12b). Cartilage formation was first detectable after 8- 11 days and became more pronounced with time thereafter. A_smaii cells showed some evidence of chondrogenesis but it was less conspicuous than with the ALarge cells (data not shown). Chondrogenesis in these cultures was confirmed by immunostaining for the cartilage-specific matrix molecule, collagen type II. Figure 13 shows production of type Il collagen by pellet cultures of A_Large placental cells and unsorted placental cells, lmmunocytochemistry revealed that ALarge placental cells grown in chondrogenic medium produced the collagen Il (white; panel b). In contrast, cultures of unsorted placental cells grown in chondrogenic medium did not show immunocytochemical evidence of collagen Il production (panel d). Negative control sections (with irrelevant primary antibodies) are shown in panels (a) and (c) respectively. These results confirm that the A_Large placental cells contain a sufficiently high level of human MPC to support chondrogenesis. In contrast, the unsorted placental cells do not have a sufficiently high level of MPC to support chondrogenesis.

Cartilage formation by the FACS sorted placentally derived MPC used in these cultures resembled that produced by mature, functionally active chondrocytes from non-human mammalian cartilage sources, (e.g. young bovine articular cartilage), confirming that the human MPC in the A_Lar_ge fraction are highly chondrogenic.

Osteogenesis by purified human placenta derived MPC and mixed placental cells Figure 14 shows bone nodule formation by placental cells grown in osteogenic medium. Mixed human placental cells and FACS-sorted placental A_Smaιι and Aι_arg_e cells were trypsinised, counted and seeded into 24-well plates, containing either standard medium (low-glucose DMEM/10% HIFCS) or osteogenic medium (DMEM containing 10^"8 dex, 5mM BGP, 50μg/ml AA) in the presence and absence of demineralised bone matrix, (a source of osteogenic growth factors), at a density of 2x10⁵ cell/well. Cells were incubated at 37⁰C and media was changed every 3-4 days thereafter. At days 7 and 14 the cells in each condition were analysed for alkaline phosphatase expression using the Napthol staining procedure. Staining was observed microscopically and images were taken. Both the unsorted placental cells and the sorted, MPC enriched placental cells formed evidence of bone-nodule formation in these assays, the dark stained areas indicating the presence of the bone cell marker enzyme alkaline phosphatase. These results suggest (i) that osteogenic MPC exist in all three of these cell populations and (ii) that osteogenesis is less dependent upon MPC purity, (or less susceptible to inhibition by contaminant cells), than is the case for chondrogenesis.

The MPC isolation method described here could be extended by using the three parameters described here, (i.e. size, granularity and autofluorescence), in conjunction with other cell isolation methods known to those skilled in the art, (e.g. density gradient centrifugation, antibody markers, lectins etc.). Such methods could be used as preliminary screening steps prior to the FACS method described here or as additional parameters during a multi-parameter FACS sort or as post FACS techniques to refine the cell populations described here, (e.g. differential adhesion methods to further refine the MPC content of the Ao population).

The MPC enriched cell populations produced using this method may be culture expanded using standard medium (preferably LGDMEM/ 10% FCS), although additional supplements maybe added to improve cell growth, especially in the first days after cell sorting. Examples include bFGF (5ng/ml), increased FCS concentrations (20%) and 10% Ham's F-12. Such supplements and growth factors are well known to those skilled in the art of connective tissue cell culture. There does not appear to be any obligatory need for special medium formulations of the type reported by other authors, e.g. highly selected batches of FCS.

The cells produced using this method could be used in regenerative medicine to promote fracture callus formation in hypocellular osseous defects (e.g. large segmental bone defects; tumour resection sites; radiotherapy sites; orthopaedic implant revision surgery sites; or bone defect sites damaged by infections)

In cartilage repair the cells could be used to repair focal full and partial thickness articular cartilage defects; osteoarthritis resurfacing or to promote meniscal repair

MPC have also been reported to promote skin lesion repair, suggesting that these cells could be used to repair dermal wounds, (e.g. ulcers).

Other connective tissue defects that could be addressed using these cells include injuries to ligament, tendon, muscle and heart muscle.

It has been reported that MPC are immunosuppressive. It is possible therefore that these cells could be used to modulate immune responses, (e.g. transplant rejection or autoimmune disease).

Claims

What is claimed is:

1. An isolated population of adherent cells derived from human placental tissue that are capable of forming cartilage

2. An isolated population of adherent cells according to claim 1 that are enriched for mesenchymal progenitor cells

3. An isolated population of adherent cells according to claims 1 or

2 wherein the cells are purified according to their intrinsic physical properties

4. An isolated population of adherent cells according to claim 3 wherein the cells are purified according to autofluorescence and granularity

5. An isolated population of adherent cells according to claim 4 wherein the cells are of lower than average autofluorescence compared with unsorted human placental cells

6. An isolated population of adherent cells according to claims 4 or

5 wherein the cells are of lower than average granularity compared with unsorted human placental cells

7. An isolated population of adherent cells according to claims 4 to

6 wherein the cells are further purified according to size

8. An isolated population of adherent cells according to claims 1 to 7 where the cells are at least 18μm in diameter

9. An isolated population of adherent cells according to claims 1 to 8 where the cells are 18-30μm in diameter

10. Use of an isolated population of adherent cells according to claims 1 to 9 for the repair of tissue defects

11. Use of an isolated population of adherent cells according to claim 10 where the tissue is bone, cartilage, skin, meniscus, ligament, tendon, striated muscle, cardiac muscle or smooth muscle

12. Use of an isolated population of adherent cells according to claims 1 to 9 for the production of tissue-engineered devices

13. Use of an isolated population of adherent cells according to claims 1 to 9 for the repair of non-union fractures, large segmental bone defects, tumour resection sites, radiotherapy sites, orthopaedic revision sites or bone sites damaged by infection

14. Use of an isolated population of adherent cells according to claims 1 to 9 for the repair of focal full and partial thickness articular cartilage defects, osteoarthritis resurfacing or promotion of meniscal repair

15. Use of an isolated population of adherent cells according to claims 1 to 9 for repair of dermal wounds

16. Use of an isolated population of adherent cells according to claims 1 to 9 to modulate immune responses

17. A method for isolating an enriched population of mesenchymal progenitor cells from human placenta, comprising the steps of: a) Dissociating placental tissue into a mixture of (i) individual cells and (ii) fragments of vascular tissue that consist of vessel-wall cells and residual extracellular matrix b) Culturing the dissociated placental tissue to isolate adherent cells c) Sorting the adherent cells according to granularity and autofluorescence d) Isolating adherent cells of below average autofluorescence and below average granularity compared with unsorted human placental cells

18. A method for isolating an enriched population of mesenchymal progenitor cells according to claim 17 where the below average autofluorescence is defined as less than or equal to that of the 5^1h most intense fluorescence intensity calibration bead supplied by DakoCytomation kit K0112

19. A method for isolating an enriched population of mesenchymal progenitor cells according to claims 17 or 18 where the below average granularity is defined as a side scatter substantially equivalent to that of a 6μm reference bead

20. A method for isolating an enriched population of mesenchymal progenitor cells according to claims 17 to 19 where the enriched cell population is further selected for cells of above average size compared with unsorted human placental cells

21. A method for isolating an enriched population of mesenchymal progenitor cells according to claim 20 where the enriched cell population is selected for cells in the 50^th to 65^th percentile of the diameter range of the unsorted placental cell population

22. An isolated population of adherent cells according to claims 17 to 19 where the cells are at least 18μm in diameter

23. An isolated population of adherent cells according to claims 17 to 19 where the cells are 18-30μm in diameter

24. The use of cells isolated according to claims 17 to 23 for the repair of tissue defects

25. The use of cells isolated according to claim 24 where the tissue is bone, cartilage, skin, meniscus, ligament, tendon, striated muscle cardiac muscle or smooth muscle

26. The use of cells isolated according to claims 17 to 23 for the production of tissue-engineered devices

27. The use of cells isolated according to claims 17 to 23 for the repair of non-union fractures, large segmental bone defects, tumour resection sites, radiotherapy sites, orthopaedic revision sites or bone sites damaged by infection

28. The use of cells isolated according to claims 17 to 23 for the repair of focal full and partial thickness articular cartilage defects, osteoarthritis resurfacing or promotion of meniscal repair

29. The use of cells isolated according to claims 17 to 23 for repair of dermal wounds

30. The use of cells isolated according to claims 17 to 23 to modulate immune responses