PHOTOCROSSLINKABLE HYDROGELS FOR CARTILAGE TISSUE ENGINEERING

Abstract

For millions of people, damaged cartilage is a major source of pain and disability. As those people often discover upon seeking medical treatment, once damaged, cartilage is very difficult to repair. Finding better clinical therapies for damaged cartilage has generated a huge amount of research interest, and has led to the studies detailed in this thesis. This work was broadly motivated by the aim of advancing our understanding of cartilage tissue engineering – that is, the creation of new cartilage from a combination of biomaterials, cells and growth factors.

Biomaterials are a key part of cartilage tissue engineering therapies. Historically the main purpose of using biomaterials in these therapies was to provide structural support for cells while the damaged tissue regenerated. More recently, it has been recognised that the composition and properties of the biomaterial influence the cell responses and quality of the newly formed tissue. This has led to the development of biomaterials for the specific application of cartilage tissue engineering.

In this thesis, the focus was on hydrogel biomaterials derived from natural polymers. The polymers, including gelatin, hyaluronic acid (HA) and chondroitin sulfate (CS), were chemically modified to become gelatin methacrylamide (Gel-MA), hyaluronic acid methacrylate (HA-MA) and chondroitin sulfate methacrylate (CS-MA) respectively, allowing them to be crosslinked by UV light in the presence of living cells. Using these materials, we investigated in vitro the viability of the encapsulated cells, the cell phenotype, and the quantity, composition and mechanical properties of the extracellular matrix they produced.

Compared to encapsulation in other widely used hydrogels such as alginate, chondrocytes produced substantially more cartilage matrix when encapsulated in Gel-MA, and the matrix was much stiffer than that in other materials. However, chondrocyte redifferentiation in Gel-MA was poor. Characterisation of cell-free Gel-MA hydrogels showed that the compressive modulus properties can be controlled over a wide range – approximately 5 to 180 kPa – and that by increasing the viscosity of Gel-MA by supplementing it with hyaluronic acid, Gel-MA structures with defined a architecture and porosity could be printed.

To address the poor redifferentiation in Gel-MA, biomimetic constructs containing HA-MA and/or CS-MA were evaluated. By including a small quantity of HA-MA in Gel-MA hydrogels (9.5% Gel-MA and 0.5% HA-MA), chondrocytes redifferentiated to their normal phenotype to a greater extent, with more rounded cell morphologies and chondrogenic gene expression patterns. Collagen type II and aggrecan were distributed throughout the gels more evenly in the presence of HA-MA. Crucially for cartilage, the developed mechanical properties of constructs with HA-MA were greatly improved compared to Gel-MA controls.

The effect of HA-MA on the developed mechanical properties of tissue-engineered cartilage constructs was further investigated using a range of HA-MA concentrations. The developed mechanical properties were highly dependent on HA-MA concentration, with higher HA-MA concentrations leading to stiffer constructs. In addition to stiffness, the failure strength was increased in gels with HA-MA. Collagen type X deposition, which is an unwanted but commonly produced protein in cartilage repair tissue, was observed in all gels. This reinforces the existing evidence that shows that prolonged growth factor exposure combined with the reactive oxygen species that are generated during crosslinking may induce chondrocyte hypertrophy.

Gel-MA hydrogels were modified with processed cartilage, tissue or tendon tissue extracts (all equine origin). The tissues were digested with pepsin and modified with photocrosslinkable groups to allow the extracts to form stable hydrogels or be incorporated into Gel-MA hydrogels. Mesenchymal stromal cells (MSCs) responded more favourably to the tissue extracts than chondrocytes. Chondrocytes showed increased catabolic processes in the presence of tissue extracts, which is consistent with other research using collagen fragments. In summary, hydrogels formed from mixtures of Gel-MA and HA-MA showed significant promise as materials for tissue engineering models, and would be interesting materials to investigate in in vivo settings.