Osteoarthritis (OA) and diabetis mellitus type 2 (DMT2) are pathogenetically linked. Complement dysregulation contributes to OA and could be involved in DMT2. The inflammatory anaphylatoxin C5a is released during complement activation. This study aims to understand the specific responses of chondrocytes isolated from diabetic and non-diabetic rats exposed to C5a and/or the proinflammatory cytokine TNFα in vitro dependent on the glucose supply. Articular chondrocytes of adult Zucker Diabetic Fatty (ZDF) rats (homozygous: fa/fa, diabetic, heterozygous: fa/+, lean controls) were exposed to 10 ng/mL TNFα and 25 ng/mL C5a alone or in combination, both, under normo- (NG, 1 g/L glucose) and hyperglycemic (HG, 4.5 g/L glucose) conditions (4 or 24 h). Chondrocyte survival, metabolic activity and gene expression of collagen type 2, suppressors of cytokine signaling (SOCS)1, −3 and anti-oxidative hemoxygenase-1 (HMOX1) were assessed. The complement regulatory protein CD46 and cell nuclei sizes were analyzed. Chondrocyte vitality remained unaffected by the treatment. Metabolic activity was impaired in chondrocytes of non-diabetic rats under HG conditions. Collagen type 2 transcription was suppressed by TNFα under HG condition in chondrocytes from nondiabetic donors and under both conditions in those of DMT2 rats (24 h)

Except for DMT2 chondrocytes under HG (4 h), HMOX1 was generally induced by TNFα +/- C5a (NG, HG). C5a elevated HMOX1 only in chondrocytes of controls. The SOCS1/3 genes were increased by TNFα (NG, diabetic, non diabetic, 4 and 24 h). This could also be observed in chondrocytes of diabetic, but not of lean rats (24 h, HG). At 4 h, C5a induced SOCS1 only in non diabetic chondrocytes (NG, HG). Cytoprotective CD46 protein was suppressed by TNFα under NG condition. Nuclear volumes of chondrocyte were lower in chondrocytes from DMT2 rats compared to those from controls. The differential response suggests that chondrocytes are irreversibly compromised by DMT2.

Achnowledgement: The authors are grateful for the support by the “Stiftung Edoprothetik (S 04/21)”

The regenerative capacity of hyaline cartilage is greatly limited. To prevent the onset of osteoarthritis, cartilage defects have to be properly treated. Cartilage, tissue engineered by mean of bioactive glass (BG) scaffolds presents a promising approach. Until now, conventional BGs have been used mostly for bone regeneration, as they are able to form a hydroxyapatite (HA) layer and are therefore, less suited for cartilage reconstruction. The aim of this study is to compare two BGs based on a novel BG composition tailored specifically for cartilage (CAR12N) and patented by us with conventional BG (BG1393) with a similar topology. The highly porous scaffolds consisting of 100% BG (CAR12N, CAR12N with low Ca2+/Mg2+ and BG1393) were characterized and dynamically seeded with primary porcine articular chondrocytes (pACs) or primary human mesenchymal stem cells (hMSCs) for up to 21 days. Subsequently, cell viability, DNA and glycosaminoglycan contents, cartilage-specific gene and protein expression were evaluated. The manufacturing process led to a comparable high (over 80%) porosity in all scaffold variants. Ion release and pH profiles confirmed bioactivity for them. After both, 7 and 21 days, more than 60% of the total surfaces of all three glass scaffold variants was densely colonized by cells with a vitality rate of more than 80%. The GAG content was significantly higher in BG1393 colonized with pACs. In general, the GAG content was higher in pAC colonized scaffolds in comparison to those seeded with hMSCs. The gene expression of cartilage-specific collagen type II, aggrecan, SOX9 and FOXO1 could be detected in all scaffold variants, irrespectively whether seeded with pACs or hMSCs. Cartilage-specific ECM components could also be detected at the protein level. In conclusion, all three BGs allow the maintenance of the chondrogenic phenotype or chondrogenic differentiation of hMSCs and thus, they present a high potential for cartilage regeneration.

Although autografts represent the gold standard for anterior cruciate ligament (ACL) reconstruction, tissue-engineered ACLs provide a prospect to minimize donor site morbidity and limited graft availability. This given study characterizes the ligamentogenesis in embroidered poly(L-lactide-co-ε-caprolactone) (P(LA-CL)) / polylactic acid (PLA) constructs using a dynamic nude mice xenograft model. (P(LA-CL))/PLA scaffolds remained either untreated (co) or were functionalized by gas fluorination (F), collagen foam cross-linked with hexamethylene diisocyanate (HMDI) (coll), or gas fluorination combined with the foam (F+coll). Cell free constructs or those seeded for 1 week with lapine ACL ligamentocytes were implanted into nude mice for 12 weeks. Following explantation, biomechanical properties, cell vitality and content, histopathology of scaffolds (including organs: liver, kidney, spleen), sulphated glycosaminoglycan (sGAG) contents and biomechanical properties were assessed.

Implantation of the scaffolds did not negatively affect mice weight development and organs, indicating biocompatibility. All scaffolds maintained their size and shape for the duration of the implantation. A high cell viability was detected in the scaffolds prior to and following implantation. Coll or F+coll scaffolds seeded with cells yielded superior macroscopic properties when compared to the controls. Mild signs of inflammation (foreign-body giant cells, hyperemia) were limited to scaffolds without collagen. Microscopical score values and sGAG content did not differ significantly. Although remaining stable in vivo, elastic modulus, maximum force, tensile strength and strain at Fmax were significantly lower in the in vivo compared to the samples cultured 1 week in vitro, but did not differ between scaffold subtypes, except for a higher maximum force in F+coll compared with F samples (in vivo). Scaffold functionalization with fluorinated collagen foam provides a promising approach for ACL tissue engineering.

(shared first authorship)

Acknowledgement: The study was supported by DFG grants SCHU1979/9-1 and SCHU1979/14-1.

Successful anterior cruciate ligament (ACL) reconstructions strive a firm ligament-bone integration. Therefore, the aim of this study was to address in more detail the enthesis as the thriphasic bone attachment of the ACL using a tissue engineering approach. To establish a tissue-engineered enthesis-like construct, triphasic scaffolds embroidered from poly(L-lactide-co-caprolactone) and polylactic acid functionalized with collagen foam were colonized with osteogenically differentiated human mesenchymal stromal cells (hMSCs) and lapine (L) ACL fibroblasts.

These triphasic scaffolds with a bone-, a fibrocartilage transition- and a ligament phase were seeded directly after spheroid assembly or with 14 days precultured LACL fibroblast spheroids and 14 days osteogenically differentiated hMSCs spheroids (=longer preculture) and cultured for further 14 days. Cell survival was tested. Collagen type I and vimentin were immunolabeled and the content of DNA and sulfated glycosaminoglycan (sGAG) was quantified. The relative gene expression of tenascin C, type I and X collagens, Mohawk and Runx2 was analyzed.

Compared to the LACL spheroids the hMSC spheroids adhered better to the scaffold surface with faster cell outgrowth on the fibers. Collagen type I and vimentin were mainly detected in the hMSCs colonizing the bone zone. The DNA content was generally higher in the bone (hMSCs) than in the ligament zones and after short spheroid preculture higher than after longer preculture whereas the sGAG content was greater after longer preculture for both cell types. The longer precultivated hMSCs expressed more type I collagen in comparison to those only shortly precultured before scaffold seeding. Type I collagen and tenascin C were higher expressed in scaffolds directly colonized with LACL compared to those seeded after longer spheroid preculture. The gene expression of ECM components and transcription factors depended on cell type and preculturing condition.

Zonal colonization of triphasic scaffolds using the spheroid method is possible offering a novel approach for enthesis tissue engineering.

Introduction and Objective

Regeneration of cartilage injuries is greatly limited. Therefore, cartilage injuries are often the starting point for later osteoarthritis. In the past, various bioactive glass (BG) scaffolds have been developed to promote bone healing. Due to the fact that they induce the deposition of hydroxyapatite (HA) -the main component of bone matrix, these BG types are not suitable for chondrogenesis. Hence, a novel BG (Car12N) lacking HA formation, was established. Since BG are generally brittle the combination with polymers is helpful to achieve suitable biomechanic stability. The aim of this interdisciplinary project was to investigate the effects of biodegradable polymer Poly(D,L-lactide-co-glycolide) (PLLA) infiltration into a Car12N scaffold for cartilage tissue engineering.

Cell sheets are manufactured from a high-density cell layer stabilized by its own freshly produced extracellular matrix (ECM). They could serve as versatile scaffolds for tissue repair. Unfortunately, their production often remains time-consuming requiring weeks of culturing. Ligament cell sheets are so far barely available. Regarding musculoskeletal tissues exposed to high repetitive biomechanical forces, the stability of cell sheets is insufficient. It could help to combine them with a biomechanical competent scaffold e.g. produced by an embroidering technique. Hence, we wanted to (1) develop a very rapid strategy to produce ACL ligamentocyte sheets within 24 h by using a thermoresponsive polymer surface, (2) use the sheets for scaffold seeding and (3) reflect the fibrocartilaginous transition zone of an ACL enthesis by combining sheets of ligamentocytes with chondrocytes or chondrogenic precursor cells as a strategy for directed seeding of two cell types on topologically different scaffold areas.

Different cell numbers of lapine ACL ligamentocytes (L-ACLs), lapine articular chondrocytes (L-ACs) and human mesenchymal stromal cells (H-MSCs) were used for sheet formation. Experiments were performed with novel, self-assembled poly(glycidyl ether) (PGE) brushes based on random glycidyl methyl ether and ethyl glycidyl ether copolymers on polystyrene 12-well cell culture plates, which allow rapid sheet formation within 24 h. Uncoated plates served as controls. Temperature-triggered detachment was performed by 10 min incubation with PBS at ambient temperature before treatment with fresh warm PBS for 5 min at 37 degrees Celsius. Harvested cell sheets were transferred on polyglycolic acid (PGA) or embroidered poly-lactic acid / poly-co-caprolactone (PLA/P[LA-CL]) scaffolds, functionalized with collagen foam and fluorine gas treatment (prepared at the IPF in Dresden and the FILK in Freiberg). Cell distribution, growth, vitality and synthesis of ECM components were monitored up to 7 days. Cell numbers required for sheet preparation (3.9 cm2) depended strongly on the cell type (L-ACLs: 0.395 mio/cm2, L-AC: 0.342 mio/cm2, H-MSCs: 0.131 mio/cm2) and was highest for L-ACLs. The majority of cells survived sheet assembly, detachment, transfer onto the scaffolds and culturing. Cells migrated from the sheets into the scaffolds and spread through the scaffolds. L-ACLs and L-ACs produced ECM and maintained their phenotypes (type II collagen and sulfated glycosaminoglycans in L-AC sheets, decorin and tenascin C in L-ACL sheets). The presence and distribution of two cell types in scaffold cocultures (L-ACLs and H-MSCs) was proven by anti-human vimentin labeling. Hence, the PGE brush surface allows rapid formation (24 h) of cell sheets.

Ligament fibroblasts must be mechanosensitive and possess sufficient adaptability to a novel mechanomilieu ensuring the permanent load capacity of the tissue. Once mechanoreceptors are activated, the fibroblasts react with a specific signal transmission (mechanotransduction), which ultimately leads to an adaption of their cytoskeletal organization and protein synthesis. However, the cellular response of anterior cruciate ligament (ACL) fibroblasts to cyclic mechanical stretching is still unclear. Hence, this study should allow a deeper understanding of the reaction profile of mechanically stretched ACL cells in two- (2D) and three-dimensional (3D) biomaterial-free and biomaterial cultures with respect to cell survival, size, orientation, migration and distribution. For the 2D approach consisting of monolayers with 6000 lapine (L) ACL cells per cm2 and for the 3D cultures using preformed LACL cell spheroids (2.5–4/cm2) with 25.000 cells per spheroid, silicone chambers were coated with geltrex and statically colonized with the LACL cells for 24 h before cyclically stretched for 48 h (14 percent uniaxial stretch). A second approach using 3D scaffold cultures was performed which were seeded dynamically for 24 h with LACL cells before cyclically stretched in a novel custom-made mechanostimulator. The scaffolds [polylactic acid (PLA) and polycaprolactone (PCL)] were functionalized with 10 percent gas fluorination and a collagen foam. Scaffolds (120 mm2) were precolonized dynamically with an LACL cell suspension (1 mio cells/mL) for 24 h before stretched for 72 h (4 percent uniaxial stretch). Cell vitality and numbers were monitored. The cytoskeleton orientation was shown by cytochemistry (F-actin) and evaluated (ImageJ). Cell proliferation, based on the DNA content was measured. Cell viability in stretched samples (2D, 3D and scaffold) remained above 90 percent. Stretching on the silicone chambers led to increased cell counts, length and significantly higher colonized areas than in unstretched controls. Higher numbers of LACL cells migrated out of the 3D spheroids under stretching conditions. In response to intermittent stretching, cells oriented in a 70 degrees' angle against the stretch direction in silicone chambers, whereas cell arrangement was more compact on the threads of the scaffolds than in unstretched cultures. In summary, stretching induced a rapid (48 h) cell and cytoskeletal alignment in 2D as well as in 3D cultures. The natural ACL is characterized by a strongly uniaxial cell and extracellular matrix organization which might be achieved in tissue engineered constructs by a suitable cyclic stretching protocol in future.

Cartilage injuries often represent irreversible tissue damage because cartilage has only a low ability to regenerate. Thus, cartilage loss results in permanent damage, which can become the starting point for osteoarthritis. In the past, bioactive glass scaffolds have been developed for bone replacement and some of these variants have also been colonized with chondrocytes. However, the hydroxylapaptite phase that is usually formed in bioglass scaffolds is not very suitable for cartilage formation (chondrogenesis). This interdisciplinary project was undertaken to develop a novel slowly degrading bioactive glass scaffold tailored for cartilage repair by resembling the native extracellular cartilage matrix (ECM) in structure and surface properties. When colonized with articular chondrocytes, the composition and topology of the scaffolds should support cell adherence, proliferation and ECM synthesis as a prerequisite for chondrogenesis in the scaffold.

To study cell growth in the scaffold, the scaffolds were colonized with human mesenchymal stromal cells (hMSCs) and primary porcine articular chondrocytes (pACs) (27,777.8 cells per mm³) for 7 – 35 d in a rotatory device. Cell survival in the scaffold was determined by vitality assay. Scanning electron microscopy (SEM) visualized cell ultramorphology and direct interaction of hMSCs and pACs with the bioglass surface. Cell proliferation was detected by CyQuant assay. Subsequently, the production of sulphated glycosaminoglycans (sGAGs) typical for chondrogenic differentiation was depicted by Alcian blue staining and quantified by dimethylmethylene blue assay assay. Quantitative real-time polymerase chain reaction (QPCR) revealed gene expression of cartilage-specific aggrecan, Sox9, collagen type II and dedifferentiation-associated collagen type I. To demonstrate the ECM-protein synthesis of the cells, the production of collagen type II and type I was determined by immunolabelling.

The bioactive glass scaffold remained stable over the whole observation time and allowed the survival of hMSCs and pACs for 35 days in culture. The SEM analyses revealed an intimate cell-biomaterial interaction for both cell types showing cell spreading, formation of numerous filopodia and ECM deposition. Both cell types revealed initial proliferation, decreasing after 14 days and becoming elevated again after 21 days. hMSCs formed cell clusters, whereas pACs showed an even distribution. Both cell types filled more and more the pores of the scaffold. The relative gene expression of cartilage-specific markers could be proven for hMSCs and pACs. Cell associated sGAGs deposition could be demonstrated by Alcian blue staining and sGAGs were elevated in the beginning and end of the culturing period. While the production of collagen type II could be observed with both cell types, the synthesis of aggrecan could not be detected in scaffolds seeded with hMSCs.

hMSCs and pACs adhered, spread and survived on the novel bioactive glass scaffolds and exhibited a chondrocytic phenotype.

Cultured primary cells have a limited life span and undergo dedifferentiation. Tissue engineering (TE) approaches require high cell numbers, but availability of human derived cells is limited and animal cells show inter-species differences. The advantages of immortalized cells are delayed senescence and phenotypic stability. The present study was undertaken to validate key properties of immortalized human anterior cruciate ligament (ACL) fibroblasts in direct comparison with non-immortalized cells from the same donor to assess their applicability as TE model. Human ACL ligamentocytes (40 years old female donor) were either immortalized using repeated transient transfection with a simian virus SV40 plasmid or remained untreated. Both cell populations were analyzed for cell survival, DNA content, tendon marker, extracellular matrix (ECM) and cytoskeletal protein expression. Cell spheroids of both populations were seeded on scaffolds embroidered either from polylactic acid (PLA) threads alone or combined PLA- and PLA-co-caprolacton-(P(LA-CL)) threads, functionalized with fluor treatment and collagen foams. Cell survival on the scaffolds was monitored for up to 5 weeks. In contrast to non-immortalized ligamentocytes, immortalized cells reflected some chaotic and incomplete cell divisions, higher DNA content, numbers of dying cells and nucleoli, reduced vimentin and vinculin-associated focal adhesions. Analysed markers, other cytoskeletal and ECM components were similarly expressed. Compared to the non-immortalized ligamentocytes immortalized formed instable spheroids and died on the scaffolds after 21 d. Both cell populations reflected superior growth on the PLA-P(LA-CL) compared with PLA scaffolds. Immortalized cells share crucial properties with their non-immortalized counterparts, but TE is only possible for limited culturing periods.