Lysozyme self-assembles into amyloid networks that support cartilage tissue formation

Purpose: No cure is available for repair of damaged cartilage. Once damaged, cartilage will continue to degenerate, resulting in immobile and painful joints. Treatments are limited to symptom relief, while tissue engineering approaches fail to produce cartilage comparable to native articular cartilage in terms of strength and modulus. In this project, we investigated the use of self-assembling proteins as scaffold material for cartilage tissue engineering. These so-called amyloid networks consist of amyloid fibrils forming physical cross-links. The fibrils self-assemble from proteins by forming inter-protein β-sheets. Amyloid networks resemble the extracellular matrix of cartilage since the strength and Young’s modulus of the fibrils is comparable to those of collagen and the networks are hydrogels. We therefore hypothesized that amyloid networks can be used as scaffold material for cartilage tissue engineering. Methods: The proteins α-synuclein, β-lactoglobulin and lysozyme were incubated under self-assembly conditions. Formation of amyloid fibrils was confirmed by using the amyloid-specific fluorescent dye ThT. The change in secondary structure was measured with CD-spectroscopy and the morphology of the obtained structures was visualized with electron microscopy. The formation of amyloid networks was confirmed with electron microscopy and ThT. Bovine chondrocytes were isolated from calf knees using collagenase and used at passage 1. The cells were cultured in monolayer for 3 days in de presence of monomers or amyloid networks of the 3 proteins to study their effect on chondrocyte viability. The percentage healthy chondrocytes was quantified with the Calcein-AM dye and flow cytometry. The metabolic activity was measured with MTT assays. The relative change in gene expression was studied with qPCR for anabolic, catabolic and hypertrophy marker genes. Amyloid networks were mixed with chondrocytes and cultured in 3D for 5 weeks to investigate whether the networks allow cartilage extracellular matrix formation. Samples were sectioned and stained histologically for aggrecan and collagen type 2. Results: A-synuclein, β-lactoglobulin and lysozyme self-assembled into amyloid fibrils as confirmed by the fluorescent signal of ThT, the change of their natural secondary structure to a β-sheet rich conformation, and their fibrous morphology. These fibrils formed amyloid networks that remained ThT positive. These networks influenced the viability of bovine chondrocytes. The percentage healthy cells increased significantly in presence of amyloid networks as compared to the monomers or when no additional proteins were added, while the metabolic activity decreased significantly in the presence of amyloid networks. All amyloid networks had a limited effect on the expression of COL2A1 and ACAN as compared to controls without amyloid networks. Α-synuclein and β-lactoglobulin networks increased the expression of MMP1 and MMP13. After 5 weeks of culture, chondrocytes produced aggrecan when cultured in agarose gels. A previously used hydrogel for cartilage formation, RADA-16, was used as a positive control. Chondrocytes cultured in RADA-16 produced almost no aggrecan. A limited amount of aggrecan was produced in the α-synuclein and β-lactoglobulin amyloid network samples. Lysozyme networks increased aggrecan production compared to agarose gels. Conclusions: Amyloid networks are best known for their presence in several pathologies. However, these structures are used throughout nature as a functional biomaterial, also in humans. Here we show that several proteins can form amyloid networks and that these networks have a positive effect on chondrocyte viability and the formation of cartilage extracellular matrix when cultured in 3D. These results indicate that amyloid networks can be used as scaffold material for cartilage tissue engineering. Furthermore, the amyloid networks are easy to produce, perform better than the synthetic hydrogel RADA-16, and there appears to be a protein specific effect.