Biomedical-grade, high mannuronic acid content (BioMVM) alginate enhances the proteoglycan production of primary human meniscal fibrochondrocytes in a 3-D microenvironment

Alginates are important hydrogels for meniscus tissue engineering as they support the meniscal fibrochondrocyte phenotype and proteoglycan production, the extracellular matrix (ECM) component chiefly responsible for its viscoelastic properties. Here, we systematically evaluated four biomedical- and two nonbiomedical-grade alginates for their capacity to provide the best three-dimensional (3-D) microenvironment and to support proteoglycan synthesis of encapsulated human meniscal fibrochondrocytes in vitro. Biomedical-grade, high mannuronic acid alginate spheres (BioLVM, BioMVM) were the most uniform in size, indicating an effect of the purity of alginate on the shape of the spheres. Interestingly, the purity of alginates did not affect cell viability. Of note, only fibrochondrocytes encapsulated in BioMVM alginate produced and retained significant amounts of proteoglycans. Following transplantation in an explant culture model, the alginate spheres containing fibrochondrocytes remained in close proximity with the meniscal tissue adjacent to the defect. The results reveal a promising role of BioMVM alginate to enhance the proteoglycan production of primary human meniscal fibrochondrocytes in a 3-D hydrogel microenvironment. These findings have significant implications for cell-based translational studies aiming at restoring lost meniscal tissue in regions containing high amounts of proteoglycans.


Results
Morphological characterization of alginate spheres containing human meniscal fibrochondrocytes. Human meniscal fibrochondrocytes were first encapsulated in various alginates as designed in Fig. 1 to determine the best suited compound that provides a favorable 3-D microenvironment for proteoglycan synthesis based on an evaluation of the morphology of the resulting spheres over an extended period of time. Application of biomedical-grade alginates always resulted in spheres that were spherically more uniform and could be prepared with a greater reproducibility compared with nonbiomedical-grade alginates (Fig. 2). The mean sphere diameters ranged between 2.69 ± 0.02 mm (BioLVM), 2.71 ± 0.03 mm (LVG), 2.74 ± 0.02 mm (BioMVG), 2.74 ± 0.03 mm (LVM), 2.81 ± 0.02 mm (BioMVM), and 2.83 ± 0.02 mm (BioLVG). Spheres composed of BioLVM displayed the smallest diameter, significantly different than those of spheres made of BioLVG (P = 0.001), BioMVM (P < 0.001), or BioMVG (P = 0.007). The diameters of all other spheres were not significantly different on day 3 (P > 0.05). After 21 days in vitro, the diameters of the LVG, LVM, BioLVG, and BioMVG spheres significantly decreased, with BioLVG and LVG spheres showing the most pronounced decrease (4.8% and 3.8%, respectively) (P < 0.001). There was no significant difference in the diameters of the BioLVM and BioMVM spheres over time (P > 0.05) (Fig. 3).
Morphology, viability, and proliferation of human meniscal fibrochondrocytes in alginate spheres. Histological evaluation of human meniscal fibrochondrocytes embedded in the different alginate spheres on day 3 after encapsulation mainly revealed cells with an oval or round morphology and were homogenously distributed without detectable differences between groups (Fig. 5).
On day 1 post-encapsulation, all spheres contained 7.48 ± 0.58 × 10 3 viable human meniscal fibrochondrocytes (Fig. 7). After 3 weeks of culture, the numbers of viable cells significantly decreased in all alginates except for LVM. The highest decline in the numbers of viable cells was seen upon encapsulation in BioLVM and BioLVG alginate (35.2% and 29.2%, respectively). The LVG and LVM spheres contained the highest numbers of viable cells at this time point. Live/dead staining of entire spheres containing human meniscal fibrochondrocytes revealed a similar pattern (Fig. 8). After 3 days post-encapsulation, a high number of human meniscal fibrochondrocytes was viable in all types of alginates (green fluorescence), while only a reduced number of cells was dead (red fluorescence) (Fig. 8). After 3 weeks, the highest amount of dead cells was stained when using BioLVM and BioLVG, in contrast to LVG or LVM spheres that showed the lowest amount of red fluorescence signal.
DNA and proteoglycan contents of alginate spheres containing human meniscal fibrochondrocytes. The DNA contents of human meniscal fibrochondrocytes encapsulated in the different constructs were estimated after 3 and 21 days post-encapsulation (Fig. 9). On day 3, the BioLVM spheres had the highest content of DNA among the alginates, e.g. 2.7-fold higher than with LVG spheres (P > 0.05) (Fig. 9). After 21 days, the highest decreases in the DNA contents were observed in spheres with high mannuronic acid content (LVM, BioLVM, and BioMVM), although only significant differences were noted with LVM (P < 0.001). The DNA contents of the cells embedded in alginate spheres with high guluronic acid content (LVG, BioLVG, and BioMVG) significantly increased from day 3 (P < 0.001). An histomorphometrical analysis of PCNA-positive cells on day 21 revealed a higher proliferation of cells embedded in BioLVM and BioMVG (P < 0.001).
A comparative estimation of the initial proteoglycan contents did not reveal significant differences between alginates (Fig. 10). After 21 days, the contents in the LVG, LVM, BioLVG, BioMVG, and BioMVM spheres were   significantly reduced (P < 0.001) (Fig. 10). When expressed per DNA on day 3 post-encapsulation, the highest amounts of proteoglycans were detected with BioLVG and LVG alginates (Fig. 11). After 21 days, the amounts of proteoglycans per DNA significantly increased only when human meniscal fibrochondrocytes were embedded in ultrapure medium viscosity high mannuronic acid alginate (BioMVM) (P < 0.05). In contrast, the proteoglycans per DNA were significantly reduced over time when cells were cultured in BioMVG (P < 0.001) (Fig. 10). No such differences were observed in all other types of alginates.
A meniscal explant culture model to study the transplantation of alginate spheres containing meniscal fibrochondrocytes into meniscal defects. The central (inner) parts of the meniscus contain the most proteoglycans, supporting its weightbearing function 31 . As the data revealed that proteoglycan production was optimal with BioMVM, we created a meniscal explant culture model to study the transplantation of alginate spheres containing meniscal fibrochondrocytes into areas where proteoglycan content is of major   importance. Transplantation of alginate spheres containing human meniscal fibrochondrocytes was possible, and the spheres remained in place without evidence for loss or fragmentation for 3 days in vitro. Histological examination of transverse sections of the composite cultures on day 3 revealed the close proximity of the alginate spheres with the meniscal tissue adjacent to the defect (Fig. 12).

Discussion
Culture of human meniscal fibrochondrocytes in alginate allows to maintain their physiological state within the hydrogel network and may thus be of high value for cell transplantation approaches. In the present study, we tested the suitability of different alginates to provide the best 3-D microenvironment for human meniscal fibrochondrocytes. First, the data demonstrated that the purity of the alginate affects the shape of the resulting spheres, with spheres based on biomedical-grade alginate with high mannuronic acid content being spherically the most uniform. A decrease in the size of all spheres was noted over time, with biomedical-grade high mannuronic acid content (BioLVM and BioMVM) spheres showing the lowest reduction. The data next indicate that the purity of the alginates does not affect the viability of the encapsulated human meniscal fibrochondrocytes. A significant decrease in the number of viable cells was reported over time in all types of alginates tested being more pronounced in BioLVM and BioLVG alginates. Of note, only cells encapsulated in BioMVM alginate produced and retained significant amounts of proteoglycans per cell, suggesting that BioMVM may be the best suited type of alginate to support proteoglycan production in primary human meniscal fibrochondrocytes in 3-D culture.
The 3-D environment better supports the phenotype and proliferative activities of meniscal fibrochondrocytes compared with monolayer culture 29 . However, specific effects of the 3-D microenvironment upon the ability to maintain their phenotype have been only rarely studied 10,29 . Culture of meniscal fibrochondrocytes in alginate spheres increased the synthesis of proteoglycans 32 , cell numbers, and transgene expression of genetically modified cells 33 . In good agreement with previous work 27, 28 , we observed here a relationship between the size and shape of alginate spheres and the composition and the purity of the alginate used. The shape of the spheres is essential for the functional survival of encapsulated cells 34,35 , as fragmented spheres or those containing many satellites are associated with protrusion of cells 36 and inflammatory responses 37 . Controllable swelling properties are indispensable features of alginate spheres 38 . The use of purified alginates in the present study minimized imperfections and led to more uniform spheres. These results support previous studies reporting a higher shrinkage during gel formation in low guluronic alginate 38 . The decrease of the size of all alginate spheres is in contrast with earlier observations which showed that a softer and less porous structure leads to the disintegration of spheres rich in mannuronic acid residues 38,39 but are in good agreement with other findings 27 and may be explained by differences in the experimental setup of testing spheres without or with encapsulated cells 35 .
Embedded human meniscal fibrochondrocytes remained viable and metabolically active as previously noted for articular chondrocytes 27,28 . Interestingly, the purity of the alginates did not affect the cell viability. These findings are in good agreement with previous work describing a decrease in meniscal cell proliferation over time upon encapsulation in alginate 29 or agarose hydrogels 40 , although they are in contrast with our previous observations when human articular chondrocytes were encapsulated in the same type of alginates 27 . This reduced cell proliferation rate may thus be attributed to a restriction of cell spreading when meniscal cells are induced to acquire a  round morphology within the hydrogel network 40 due to their dual morphology similar to either fibroblasts or chondrocytes 10,29 .
The meniscal proteoglycans in the ECM are chiefly responsible for the viscoelastic compressive properties, a pivotal factor in its shock absorber function 29,41 . In addition, they maintain the hydration grade of the tissue forming a basis to provide to the meniscal tissue a high capacity to resist compressive loads through compressive stiffness 29,42 . Noteworthy, the central parts of the menisci contain the highest glycosaminoglycan concentrations, besides the meniscal attachments 31 . The present study provides insight into the ability of proteoglycans synthesis by meniscal fibrochondrocytes upon alginate encapsulation. A pattern of production similar to explant cultures was observed, indicating a phenotype resembling the native situation 32 . Interestingly, differential synthesis of proteoglycans was also reported in encapsulated meniscal fibrochondrocytes derived from different meniscal regions 32 . Noticeably, only meniscal fibrochondrocytes embedded BioMVM alginate produced and retained significant amounts of proteoglycans over the 21-day culture period. These data are of high relevance for cell transplantation approaches based on alginates into meniscal areas with a high proteoglycan content.
Finally, the meniscal defect explant culture model employed here may be used to test the effect of different defect and alginate sizes and compositions on cellular behaviour in a relatively natural environment in a standardized and reproducible manner with the view of improving strategies for meniscal repair. For example, interactions of paracrine factors (secreted by the meniscal tissue) with the transplanted fibrochondrocytes within the alginate or vice versa (e.g. secreted following gene transfer into meniscal fibrochondrocytes 33,43 ) may be investigated 33,43 . If the defects are of a smaller diameter and filled with constructs allowing for the migration of fibrochondrocytes, mechanisms of cell-based meniscal defect repair could be explored.
A possible limitation of this study is the lack of in vivo data. Implantation of selected alginate spheres in experimentally created meniscal lesions in translational minipig 44 or sheep models 45,46 is the next step that will provide more information about functionality of such hydrogels in a clinically relevant in vivo environment. Also, contractile markers determining the response of meniscus to injury such as α -smooth muscle actin (α -SMA) may be studied using this explant culture model, e.g. using human meniscal explants 46 . On the other hand, using meniscal fibrochondrocytes from healthy donors may exclude any influence of possible previous pharmacological treatments such as intra-articular injections of steroids.
Altogether, biomedical-grade, high mannuronic acid content alginate enhanced proteoglycan production of primary human meniscal fibrochondrocytes in 3-D hydrogel culture. These data are of promising value for cell-based translational studies aiming at restoring lost meniscal tissue in regions containing high amounts of proteoglycans.

Materials and Methods
Materials. All reagents were purchased at Invitrogen/GIBCO (Karlsruhe, Germany) unless otherwise indicated. L-cystein, Na 2 EDTA, calf thymus DNA, were from Sigma (Munich, Germany). Collagenase type I (activity: 232 U/mg) was purchased at Biochrom (Berlin, Germany). Dimethylmethylene blue was obtained from Serva (Darmstadt, Germany). Chondroitin-6-sulfate from shark cartilage was purchased at Fluka (Munich, Germany). Plasticware was from Falcon (Becton Dickinson, Pont de Claix, France). The PCNA (proliferating cell nuclear antigen) antibody was from Santa Cruz Biotechnology (Heidelberg, Germany). Biotinylated secondary antibody and the ABC reagent were from Vector Laboratories (Alexis Deutschland GmbH, Grünberg, Germany). The investigation was performed at the Center of Experimental Orthopaedics, Saarland University (Saarland, Germany).
Isolation and cultivation of human meniscal fibrochondrocytes. Human meniscal fibrochondrocytes were isolated from normal human adult menisci obtained from knee joints treated by total knee arthroplasty from (n = 4, average age 65.5 years ± 13.2, range 46-75 years) as previously described 33,43 . The study was approved by the Ethics Committee of the Saarland Physicians Council (Ärztekammer des Saarlandes, Ethik-Kommission, No. 67/12). All patients provided informed consent before inclusion in the study, and all procedures were in accordance with the Helsinki Declaration. Menisci with degeneration on gross examinations were excluded. Menisci were washed, diced into 4 × 4 mm pieces and transferred to DMEM with 100 U/ml penicillin G and 100 μ l/ml streptomycin (basal medium) containing 2% fetal bovine serum (FBS) (growth medium) and 0.01% (w/v) collagenase at 37 °C in a humidified atmosphere with 5% CO 2 for 16 h. Isolated cells were filtered through a 100 μ m mesh to remove undigested matrix, washed twice with phosphate-buffered saline (PBS), seeded into 75-cm 2 tissue culture flasks and kept at 37 °C. Cells were encapsulated in alginate spheres at passage 2.
Encapsulation of human meniscal fibrochondrocytes in alginate spheres. Cells were trypsinized, washed, and resuspended in sterile-filtered 1.2% alginate in 0.15 M NaCl at a density of 10 6 cells/ml as previously described 27,28,33 . The cell suspension was extruded through a 21-gauge needle (Braun, Melsungen, Germany) into 102 mM CaCl 2 under constant shaking and allowed to polymerize for 10 min. After one wash in PBS and three washes in DMEM, the spheres were maintained at 37 °C in a humidified atmosphere of 5% CO 2 in growth medium that was changed twice per week (Fig. 1).
Microscopic evaluation of the alginate spheres. The morphology of the alginate spheres using each type of alginate were estimated on days 3 and 21 post-encapsulation using an inverted optical microscope (CKX-4; Olympus; Hamburg, Germany). The dimensions of the alginate spheres were measured using a computer-based image analysis (n = 4) 27,47,48 . Images of the whole spheres were acquired by a solid-state CC-12 digital camera (Soft Imaging System) mounted on an inverted microscope analyzed with the analySIS Five Program (Soft Imaging System Corp., Munster, Germany).
Cell proliferation and viability. On days 0, 1, and 21 after encapsulation, individual spheres were dissolved in 100 μ l of 55 mM sodium citrate and 90 mM NaCl (pH 6.8) for 20 min at room temperature. Meniscal fibrochondrocytes were counted using a Neubauer chamber and cell viability was determined by trypan blue exclusion (n = 4) 28 . Parallel cell viability within the spheres at 3 and 21 days post-encapsulation was qualitatively assessed using fluorescence staining with propidium iodide and acridine orange using a fluorescence microscope (Olympus CKX41). After removing the medium spheres were washed carefully with PBS twice and incubated with acridine orange/propidium iodide staining solution (10 μ g/ml each in PBS) for 10 min at room temperature 49,50 . Live cells are visualized in green (acridine orange) versus dead cells in red (propidium iodide).
Total sulphated glycosaminoglycan content of alginate spheres. On days 3 and 21 post-encapsulation, alginate spheres were dissolved as described above and the released meniscal fibrochondrocytes were incubated overnight in 125 μ g/ml papain in 1x PBE (100 mM sodium phosphate buffer, 10 mM Na 2 EDTA, pH 6.5) 28 . Proteoglycans were measured spectrophotometrically by binding to the dimethylmethylene blue dye using chondroitin-6-sulfate to generate a standard curve (n = 6) 51,52 . The DNA contents were determined by the Hoechst 33258 assay [53][54][55] . All data were normalized to the total protein contents assessed by a BCA assay (Pierce, Limburg, Germany), and to the DNA contents [53][54][55] . All measurements were performed on a GENios spectrophotometer/fluorometer (Tecan, Crailsheim, Germany).
Histological and immunohistochemical evaluation of the alginate spheres. Spheres based on the different alginates with encapsulated meniscal fibrochondrocytes (n = 4) were harvested after 3 an 21 days of in vitro culture, fixed in 4% buffered formalin, dehydrated in graded alcohols, and embedded in paraffin [53][54][55] . Paraffin-embedded sections (5 μ m) were stained with hematoxylin and eosin (H&E) and safranin O according to routine protocols 56 . Expression of PCNA was detected by immunohistochemistry using a specific primary antibody as described elsewere [53][54][55] . To control for secondary immunoglobulins, sections were processed with omission of the primary antibody. Samples were examined under light microscopy (Olympus BX 45). The percentage of cells positive for PCNA immunostaining was measured using eight serial histological and immunohistochemical sections for each parameter, test, and replicate condition 53-55 . Transplantation of alginate spheres containing human meniscal fibrochondrocytes in a meniscal explant culture model. Human menisci were obtained from normal human adult menisci from knee joints treated by total knee arthroplasty (n = 5), and maintained in growth culture medium. Sections of medial Scientific RepoRts | 6:28170 | DOI: 10.1038/srep28170 menisci (approximately 5-mm thick) were generated using a #21 scalpel blade. Using a 2-mm diameter dermal biopsy punch (Kai Medical, Tokyo, Japan) that was applied strictly perpendicular to the cut meniscal surface, one cylindrical meniscal defect was created in the central (inner) parts of each meniscal section, and one BioMVM alginate sphere containing meniscal fibrochondrocytes was transplanted in a press-fit fashion into the defect and cultured for 3 days in growth medium at 37 °C and 5% CO 2 (Fig. 1). Composite cultures were then subjected to macroscopic and histological examination with H&E and safranin O/fast green staining 46,56 . Statistical analysis. Data are given as mean ± standard of the mean (SEM). OneWay ANOVA was used in multiple comparisons. The Student's t-test was used to detect significant differences when two groups were compared. P values of < 0.05 were considered significant. Analyses were conducted using Origin 8 (OriginLab Corporation, Northampton, MA, USA).