Search

Cystatin C is an important inhibitor of lysosomal and extracellular cysteine proteases. It is synthesized as a precursor containing a signal sequence that targets the protein to the secretory pathway. In retinal pigment epithelial (RPE) cells, cystatin C is abundantly expressed and secreted, suggesting a critical role in the regulation of the proteolytic activity in the extracellular space, which, if altered, may contribute to diseases such as Age-related Macular Degeneration (AMD). An amino acid substitution in the signal sequence of the precursor cystatin C generates the variant B, which has been linked to an increased risk of developing AMD and Alzheimer's disease (AD). This mutation has been reported to affect protein trafficking, resulting in decreased secretion and mitochondrial mislocalization. Recently, however, both its disease association as well as the effect on the protein trafficking have been questioned. The main aim of my study was to further investigate the trafficking and secretion of the mutated cystatin C. Recombinant wild-type and variant B cystatin C constructs, untagged or linked to different tags (EGFP, FLAG, HA), were transfected into RPE cell lines. Intracellular localization was investigated by imaging and subcellular fractionation. Protein trafficking and secretion were examined by analysing protein release into the culture media and the use of inhibitors of the secretory pathway. In addition, glycosylation sites were engineered to monitor transition through the Endoplasmic Reticulum (ER) and Golgi apparatus. Constructs encoding the signal sequence linked to EGFP were also used to evaluate the targeting properties of the variant B signal sequence. In the second part of my study, the efficiency of signal sequence cleavage was investigated by analysing the levels and localization of the precursor fraction by immunoprecipitation and imaging. The presence of alternatively cleaved forms and glycosylation modifications was also analysed, using immunodetection, mass spectrometry and deglycosylation treatments. In the last part, mechanisms of protein regulation, including degradation, internalization, dimerization were investigated to identify possible differences between the wild-type and variant B proteins. Interacting proteins were identified by mass spectrometry. My results showed no major differences in trafficking and regulation between wild-type and variant B proteins. Only the EGFP-tagged constructs resulted in mistrafficking to mitochondria, and this was the case for both cystatin C forms, suggesting that EGFP is not a suitable tag to study cystatin C and excluding mitochondrial mislocalization as a factor contributing to disease association. There were also no differences in secretion, regulation and interaction partners between wild-type and variant B. However, precursor processing for the variant B was slightly less efficient, resulting in a slightly higher fraction of non-processed precursor protein and an alternatively cleaved, possibly O-glycosylated form. The mutation in the signal sequence therefore appears to affect the precursor processing rather than targeting. These forms may accumulate over-time generating proteolytic imbalance and protein aggregates, which ultimately may lead to cellular dysfunction and degeneration associated with AMD and AD. Interestingly, inhibition of the ubiquitin-proteasome system or the autophagy-lysosome pathway, which become functionally impaired during aging and disease, led to a decrease or increase of cystatin C protein levels, respectively, suggesting a possible involvement in the altered levels of cystatin C in the structural and functional changes seen in age-related conditions.

Synovitis is a key mediator of osteoarthritis (OA) and in the perpetuation of cartilage degradation, and approximately 60% of lameness in the horse has previously been attributed to OA. Synovial inflammation is characterised by a mononuclear cell infiltration into the synovial membrane, with concurrent increases in catabolic cytokines predominantly produced by synovial macrophages. These cytokines cause changes in fibroblast-like synoviocyte gene expression, including the up-regulation of damaging cartilage matrix degrading proteinases and further cytokines. It was therefore hypothesised that the synovial membrane may represent a disease-modifying target. When referring to synovial macrophages and fibroblast-like synovioctyes collectively, the term synoviocytes is used. The general aims of this thesis were to design and test an anti-cytokine therapy; short interfering RNA (siRNA), and to analyse its potential with respect to decreasing target catabolic cytokines produced by synoviocytes. Delivery of this therapy to synoviocytes via biodegradable polymer microspheres was also examined. The effects of the therapy and therapydelivery system on synoviocyte and cartilage health, were assessed using in vitro co-culture systems, which were designed and used to elucidate interactions between the two predominant cell populations present during synovitis; synovial macrophages and fibroblast-like synoviocytes. Successful equine anti-IL-1ß siRNA was designed and tested on equine synovoicytes. These cells were also shown to successfully phagocytose 2- 6µm biodegradable polymer microspheres in a highly efficient manner. A novel multi-species in vitro model was created to quantify cell-specific responses to inflammatory stimulation, of both macrophages and fibroblastlike synoviocytes in co-culture. This model revealed that macrophages can stimulate fibroblast-like synoviocyte mitosis, and modulate its ADAMTS5 mRNA expression in an inflammatory environment. Using an inflammatory stimulated co-culture model involving a mixed culture of macrophages, fibroblast-like synoviocytes and cartilage explants, it was observed that the mixed synoviocyte culture can also act to moderate the inflammatory response. Synoviocytes were shown to exert a protective mechanism over cartilage glycosaminoglycan release, whilst driving the inflammatory response through the production of pro-inflammatory cytokines. This thesis has successfully provided the initial design and testing of an equine anti-IL1ß therapy and delivery vehicle in vitro, providing a platform for further experimentation to assess the efficiency of delivery and therapeutic potential when used together.

Disorders of cartilage and tendon account for a high incidence of disability and are highly prevalent co-morbidities within the ageing population; therefore, musculoskeletal disorders represent a major public health policy issue. Despite considerable efforts to characterise biochemical and biomechanical cues that promote a stable differentiated cartilage or tendon phenotype in vitro the benchmarks by which progress is measured are limited. Common regenerative interventions, such as autologous cartilage implantation, have a required period of monolayer expansion that induces a loss of the functional phenotype, termed dedifferentiation. Dedifferentiation has no definitive mechanism yet is widely described in both regenerative and degenerative contexts; in addition to stem cell transplantation and cell-seeding in three-dimensional scaffolds, dedifferentiation represents the third approach to the development of regenerative mechanisms for mammalian tissue repair. Cartilage and tendon show a number of common features in structure, develop, disease, and repair. The extracellular matrix is a dynamic and complex structure that confers the functional mechanical properties of cartilage and tendon. Dysregulation of production and degradation are critical to the pathophysiology of musculoskeletal disorders, therefore, reparative interventions require a stable, functional phenotype from the outset. Cartilage and tendon demonstrate a commonality in terms of function defining structure both being sparsely cellular with a preponderance of collagenous matrix. Parity of functionality with the pre- injury state after healing is rarely achieved for cartilage and tendon. Cartilage and  tendon also share common embryological origins. Common mesenchymal progenitor cells differentiate into many musculoskeletal tissues with diverse functions. Specialist sub-populations of tendon and cartilage progenitors enable formation of transitional zones between these developing tissues. The development of musculoskeletal structures does not occur in isolation, however, cartilage and tendon have not previously been considered together in a systems context. An integrated understanding of the differentiation of these tissues should inform regenerative therapies and tissue engineering strategies. Systems biology is paradigm shift in scientific thinking where traditional reductionist strategies to complex biological problems have been superseded by a holistic philosophy seeking to understand the emergent behavior of a system by the integrative and predictive modeling of all elements of that system. Whole transcriptome and proteome profiling studies are used to collect quantitative data about a system, which may then be exploited by systems biology methodologies including the analysis of gene and protein networks. Gene-gene co-expression relationships, which are core regulatory mechanisms in biology, are often not part of a comprehensive gene expression analysis. Many biological networks are sparse and have a scale-free topology, which generally indicates that the majority of genes have very few connections, whilst certain key regulators, or ‘hubs’, are highly interconnected. Co-expression networks may be used to define regulatory sub- networks and ‘hubs’ that have phenotypic associations. This approach allows all quantitative data to be used and makes no a priori assumptions about relationships in the system and, therefore, can facilitate the exploration of emergent behavior in the system and the generation of novel hypotheses. The ultimate goal of tissue engineering is the replacement of lost or damaged cells, and in vitro, to develop biomimetic (organotypic) structures to serve as experimental models. Tissues, and the strategies to functionally replicate them ex vivo, are complex and require an integrated, multi-disciplinary approach. Systems biology approaches, using data arising from multiple-levels of the biological hierarchy, can facilitate the development of predictive models for bioengineered tissue. The iterative refinement, quantification, and perturbation of these models may expedite the translation of well-validated organotypic systems, through legal regulatory frameworks, into regenerative strategies for musculoskeletal disorders in humans. In this thesis the systems under consideration are the major cell populations of cartilage and tendon (chondrocytes and tenocytes, respectively). They are described in three environmental conditions: native tissue, monolayer (two- dimensional), or three-dimensional models. There has been no systematic investigate of the global gene and protein profiles of cartilage and tendon in their native state relative to monolayer or three-dimensional cultures. There is no clear mechanistic description of the impact of in vitro environmental perturbations on the system or indeed the adequacy of these models as proxies for cartilage and tendon. A discovery approach using transcriptomic and proteomic profiling is undertaken to define a robust and consistent gene and protein profile for each condition. Differentially expressed elements are functionally annotated and pathway topology approaches employed to predict major signalling pathways associated with the observed phenotype. This study defines dedifferentiated chondrocytes and tenocytes in monolayer culture as expressing markers of musculoskeletal development, including scleraxis (Scx) and Mohawk (Mkx). Furthermore, there is reproducible synthetic profile convergence in monolayer culture between cartilage and tendon cells. Standard three-dimensional culture systems for chondrocyte and tenocytes fail to replicate the gene expression profile of cartilage and tendon. The PI-3K/Akt signaling pathway is predicted to be the predominant canonical pathway associated with de- and re-differentiation in vitro. Using novel, and publically available, transcriptomic data sets a meta-analysis of microarray gene expression profiles is performed using weighted gene co- expression network analysis. This is employed for transcriptome network decomposition to isolate highly correlated and interconnected gene-sets (modules) from gene expression profiles of cartilage and tendon cells in different environmental conditions. Sub-networks strongly associated with de- and re- differentiation phenotypes are defined. Comparison of global transcriptome network architecture was performed to define the conservation of network modules between a model species (rat) and human data. In addition to the annotation of an osteoarthritis-associated module in the rat a class-prediction analysis defined a minimal gene signature for the prediction of three-dimensional cultures from standard monolayer culture. Finally, proteomic and transcriptomic data sets are integrated by defining common upstream regulators (TGFB and PDGF BB) and unified mechanistic networks are generated for de- and re- differentiation. The studies collected in this thesis contribute to a wider understanding of cartilage and tendon tissue engineering and organotypic culture development. A clear mechanistic understanding of the regulatory networks controlling differentiation of cartilage and tendon progenitor cells is required in order to develop improved in vitro models and bio-engineered tissue that are physiologically relevant. The findings presented here provide practical outputs and testable hypotheses to drive future evidence-based research in organotypic culture development for musculoskeletal tissues.