Characterisation of Regional Human Meniscal Progenitor Cells

Background The surgical treatment of meniscus injury has represented a clinical challenge for decades. Stimulating meniscus regeneration using transplanted meniscal progenitor cells has been suggested as a promising new strategy. However, there is a lack of studies which decisively identify and characterise progenitor cell populations in human meniscus tissues. Methods In this study, donor-matched progenitor cells were isolated via selective fibronectin adhesion from the avascular (PAvas) and vascular (PVas) regions of the meniscus and chondroprogenitors (PChs) from articular cartilage (n=5 donors). In addition, whole mixed populations of cells (MAvas, MVas, MChs) from the same regions were obtained by standard isolation techniques for comparison. The colony formation efficacy of PAvas, PVas and PChs was monitored using Cell-IQ ® live cell imaging. Proliferation rates of progenitors were compared with their mixed population counterparts. Cell surface markers indicative of mesenchymal stromal cells (MSCs) profile and progenitor markers were characterised by flow cytometry in all populations. The chondrogenic capacity was assessed via pellet culture assays and measuring chondrogenic gene expression levels, GAG/DNA content and morphology. showed significantly higher positivity for CD49b and CD49c compared to their mixed population counterparts and PChs had a higher positivity level of CD166 compared to mixed chondrocytes. Collagen types II and X expression was significantly downregulated in pellets formed by progenitor populations. GAG/DNA analysis demonstrated that progenitor cells generally produced more GAG than mixed populations. Our study demonstrates that the human meniscus contains meniscal progenitor populations in both the avascular and vascular regions. Meniscal progenitors derived from the vascular region exhibit enhanced proliferative and chondrogenic characteristics compared to those from the avascular region; this may associate with the enhanced meniscal healing potential in the vascular region. These findings build on the body of evidence which suggests that meniscal progenitors represent an attractive cell therapy strategy for meniscal regeneration.


Introduction
Menisci play a key role in joint congruence, dispersing load and protecting the articular cartilage surface of the femur and tibia 1 . Although treatment of meniscal tears has developed dramatically with such treatments as meniscus repair and replacement strategies, these surgical interventions provide limited protection against the progression of osteoarthritis (OA) 2,3 . Therefore, there is a demand to develop other novel treatments to improve meniscus repair and prevent or delay the onset and progression of OA.
The meniscus is composed of an outer "vascularised" zone that contains elongated fibroblastlike cells and an inner "avascularised" zone that contains rounded fibrochondrocytes. It has long been known that in the vascular region, tears of the meniscus tend to successfully repair themselves after a surgical procedure, whereas the inner avascular region has a low healing potential 4 . Mobilisation and homing of endogenous progenitor cells from the vascular zone of the meniscus may be responsible for some of the natural healing noted in the tissue following injury 5 . Indeed the enhanced regeneration of the vascular region of the meniscus might be due to the presence of CD34 and CD166 immunopositive progenitor cells visible via histological analysis in the blood vessels 6 . Further studies suggest the presence of progenitor cells in the meniscus promote repair of injured menisci in bovine, rabbit and mouse models [7][8][9] . Our previous work has demonstrated that there are fewer blood vessels carried by "tree-like" collagen fibres in the vascular zone of more degenerative menisci than in more healthy menisci 10 . These results suggest that there is a subpopulation of progenitorlike cells in the vascular region, but there is a lack of studies that decisively identify and characterise progenitor cell populations in different regions of the human meniscus .
Fibronectin-coated flasks have been commonly used to extract chondroprogenitors from articular cartilage and these progenitors have been investigated for their capability in terms of cartilage regeneration 11 . Chondroprogenitors are an ideal candidate for cell-based tissue engineering cartilage repair strategies because they are known as a relatively undifferentiated population of chondrocyte precursors, which are less likely to become hypertrophic and terminally differentiate sooner than their counterpart mature chondrocytes 12 . In this study, an established chondroprogenitor isolation protocol 11 was used to obtain and characterise the progenitors and whole mixed populations from donor-matched avascular and vascular regions of the meniscus, as well as the two cellular fractions from articular cartilage from the same individual.

Patient Information
The provision of written informed consent was obtained from each patient prior to surgery.
Favourable ethical approval was given by the National Research Ethics Service (NRES number 11/NW/0875) and all experiments were performed following relevant guidelines and regulations. Human meniscus (five patients) and cartilage (four patients) tissue samples were harvested from four patients undergoing total knee replacement (TKR) and one patient undergoing above-knee amputation (Table 1). The general workflow is shown in Figure 1.  above-knee amputation, IHC: immunohistochemistry. Scale bars represent 250 μm.

Progenitor Cell Isolation
The meniscus was dissected longitudinally into three parts: the inner avascular zone, the middle and the outer vascular zone. The middle portion was discarded and only the extreme inner and outer avascular and vascular zones were used to derive meniscal cells in order to ensure distinct regions. Additionally, full-depth macroscopically normal human articular cartilage from femoral condyles was used to isolate chondrocytes and their progenitors.
Cells extracted from the three tissue types (avascular & vascular menisci and articular cartilage) were cultured under two conditions. The first was for the isolation and growth of a mixed population of cells (avascular meniscal mixed cells (MAvas), vascular meniscal mixed cells (MVas) and mixed chondrocytes (MChs)), which were each plated at a density of 5000 cells/cm 2 . The second was for the isolation and growth of progenitor cells (avascular meniscal progenitors (PAvas), vascular meniscal progenitors (PVas) and chondroprogenitors (PChs)), which were subjected to a fibronectin selective adhesion as previously described 11 .
Isolated mixed populations from each tissue fraction (1000 cells/well) were seeded onto the coated wells in triplicate for 20 mins at 37℃ in DMEM/F12 medium, 10% Foetal Bovine Serum (FBS) (Gibco, USA), 1% P/S and 0.1 mM ascorbic acid (Sigma-Aldrich, UK) (culture medium). After 20 minutes, medium and non-adherent cells were removed and fresh culture medium was added to the remaining adherent cells and incubated at 37°C, 5% CO2.

Cell-IQ ® Live Cell Imaging
After three days of culture in a 6 well plate, progenitor cells were imaged using a Cell-IQ ® phase contrast Live Imaging Platform (CM Technologies, Tampere, Finland) to monitor colony formation. Spare wells and surrounding areas in 6 well plate were filled with distilled water to keep the plate humidified. At least one colony was monitored in each well and each colony was imaged every 30 minutes for 48 hours. The recorded images were analysed using the Cell-IQ ® Analyser (CM Technologies, Tampere, Finland) software in order to measure the live cell number in each colony every 30 minutes.

Growth Kinetics
The number of monoclonal colonies (defined as a cluster of more than 32 cells which represents a population of cells derived from more than 5 population doublings of a single cell) in each well was counted under light microscopy after 5 days of culture, which was considered as the initial number of progenitors that had adhered to the plate 11 . Each type of progenitor population was then trypsinised and cultured in culture medium supplemented with 1ng/ml transforming growth factor beta 1 (TGF-β1) (PeproTech, UK), 5ng/ml fibroblast growth factor 2 (FGF-2) (PeproTech, UK). Population doubling times (PDTs) were calculated using the formula: PDT = (t2-t1) x ln(2)/ln(n2/n1), where t1= the time of cell seeding, t2 = the time of cell harvest and n = the cell population at the matching time points.
PDTs at passages 0 to 2 and passage 2 to 3 were compared between mixed populations and progenitor cells, as progenitors could not be counted with sufficient accuracy at P0-1.

Flow Cytometry
Prior to chondrogenic differentiation, 120,000 avascular meniscal cells, vascular meniscal cells and chondrocytes from both mixed population and progenitor groups were resuspended in a PBS buffer consisting of 2% (w/v) bovine serum albumin (BSA; Sigma-Aldrich). Non-specific antibody binding was blocked using a PBS buffer composed of 10% (v/v) human immunoglobulin (Grifols, Spain) at 4°C for 1 hour. Immunopositivity for 9 surface molecules which are indicative of mesenchymal stromal cell (MSC) the International Society for Cellular Therapy (ISCT) profile (CD14, CD73, CD90, CD105), progenitor markers (CD44 CD166) or cell adhesion molecules (CD29, CD49b, CD49c) were evaluated in all five donors.

Chondrogenic Differentiation Assay
The chondrogenic potential of the three donor-matched populations in both mixed populations and progenitor groups was assessed at passage 2 using a well-established 3D pellet culture protocol 13  was used to construct a standard curve, which was plotted using the following equation: (A530nm/A590nm)-(A530nm blank/A590nm bank). The total GAG content of each pellet was calculated from the standard curve. The DNA content was measured spectrofluorometrically using a PicoGreen dsDNA Assay kit (Invitrogen) according to the manufacturer's instructions. Finally, the GAG content was normalised to the corresponding DNA content per pellet.

Sectioning and Immunohistochemical Staining of Chondrogenic Pellets
Three pellets from each cell population were cryosectioned at 7 m thickness.  15 .

Statistical analyses
GraphPad Prism (Version 8.30, San Diego, California, USA) was used for statistical analysis.
Two-way ANOVA with a multiple comparisons test was used to analyse flow cytometry, population doubling time, pellet RNA gene analysis, pellet GAG/DNA assay and pellet collagen staining intensity. Data were presented as mean ± standard deviation (SD) in the graphs and text.   (Figure 3).

Cell-IQ ® Live Cell Imaging Analysis
Four of five patients' progenitor colonies of PAvas, PVas and PChs were monitored for 48 hours in the Cell-IQ ® live cell imager. For each progenitor cell type from each patient, two or three colonies were selected for colony proliferation rate analysis. Figure 4 showed the results of individual colony proliferation data. Colonies with less than 32 cells beyond 48 hours in culture were not considered to be progenitor colonies and so were excluded; these comprised three of 12 colonies for PAvas, four of 11 colonies for PVas and three of 10 colonies for PChs which were characterised as non-progenitor colonies. After excluding these nonprogenitor colonies (10 of 33 colonies, 30.3%), proliferation data from progenitor colonies only was compared for PAvas, PVas, PChs fractions ( Figure 5). The comparison analyses ( and PChs (P<0.0001), whilst the PChs colony proliferation rate was significantly slower than the rate of proliferation for PAvas colonies (P=0.0026).  progenitor chondrocytes, CI: confidence interval.

Cell Surface Marker Immunoprofiles
Flow cytometry analyses ( Figure 6) revealed that all cell populations were over 95% immunopositive for the ISCT MSC markers CD73, CD90 and CD105, as well as other matrix adhesion markers (CD29 and CD44). CD14 was present on all cell populations, ranging on average from 5.18% to 10.75% positivity. There was no significant difference noted for any of these cell surface markers when comparing mixed populations and progenitor cells.     Figure 9). When the immunostaining was assessed semi-quantitively, there was more staining for collagen type I observed in pellets formed by all the progenitor populations compared with the mixed populations, but the difference was not significant ( Figure 10A). In contrast, a trend for weaker collagen type II staining was detected in all progenitor pellets compared with mixed population pellets ( Figure 10B), while only MChs pellets were found to demonstrate significantly stronger staining for collagen type II compared to PChs pellets (P=0.0383).   were presented as mean ± standard deviation.

Discussion
Cell-based therapies for meniscus tissue engineering are likely to represent key future meniscus regeneration strategies 17 . Recent studies have supported the hypothesis that meniscus progenitor cells are the most effective cell type for meniscus regeneration, thought to be due to their tissue specificity and histocompatibility 18 . However, the characteristics of human meniscus progenitor cells have not previously been comprehensively investigated.
Morphologically, the primary progenitors, PAvas and PChs, displayed characteristic cobblestone shaped morphologies, whereas PVas had an elongated fibroblast-like morphology. In the mixed populations, cells presented with more extensive cytoplasmic processes compared to their progenitors. Chondrocytes that display cytoplasmic process could be considered to have undergone a hypertrophic change, which is akin to changes observed in late-stage OA cartilage 19 . Samples used in this study were derived from latestage OA TKR samples, which might explain this distinct morphological feature noted in the mixed populations. However, the progenitor cells isolated from these OA tissues retained a typical proliferative fibroblastic morphology throughout the culture period assessed.
Clonogenicity is a key feature of all types of stem cells derived from various sources including neural 20 , hematopoietic 21 , embryonic stem cells 22 and epidermal stem cells 23 Table 3 summarises the meniscus progenitor cell isolation protocols used across previous studies. The table shows that the procedures used vary widely from FACS (fluorescence-activated cell sorting) sorting, selective adhesion, low seeding density and tissue explant isolation. All of the proposed methodologies have successfully produced colonies. However, which of these protocols produced progenitor cells with a higher colony forming efficiency is unclear.  31 and in our work collagen type X gene expression levels were found to be significantly downregulated in progenitor cells compared to the mixed populations. Together this might indicate that mixed population cells undergo hypertrophic differentiation as part of the progression of OA, whereas their progenitor counterparts are less committed to differentiating and more stem cell like with their lower expression of SOX9 and type X collagen. In addition, we found that progenitor cells generally produced higher amounts of GAGs compared with mixed population cells in terms of GAG/DNA analyses (significant difference only found between MVas and PVas). These findings suggest that the progenitor population in the meniscus is a suitable cell source for use in rebuilding the proteoglycan-rich avascular zone of damaged menisci, which represents a key challenge for meniscus repair in the clinic 4 . Interestingly, we found that the collagen type II staining intensity of PChs was In conclusion, our study demonstrates that the human meniscus contains meniscal progenitor populations in both the avascular and vascular regions based on clonogenicity and chondrogenic differentiation capacity. Our results also suggested that meniscal progenitors derived from the vascular region of the meniscus exhibit enhanced reparative characteristics which likely associate with the better meniscal healing potential previously observed in the vascular region. The findings of this study build on the body of evidence which suggests that meniscal progenitors represent an attractive cell therapy strategy for the enhancement of meniscal repair and regeneration.

Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding authors on reasonable request.