Osteosarcoma is the most common primary malignancy of bone. 5-year survival has remained stable at around 60-70% for 40 years. However, a number of patients will suffer from recurrent and/or metastatic disease representing a large unmet clinical need. CSPG4 is a transmembrane protein which is expressed on a number of progenitor cells and tumour types. Preliminary work had found CSPG4 present in osteosarcoma tumour samples. In this study, CSPG4 mRNA and protein expression was demonstrated in clinical samples and model cell lines. CSPG4 mRNA is overexpressed in osteosarcoma samples compared to mature osteoblast cells, the putative cell of origin for osteosarcoma. In a cohort of patients, CSPG4 protein expression was found on 86% of samples. Furthermore, CSPG4 expression was demonstrated in U2OS, MG63, HOS, HOS-MNNG and 143B osteosarcoma cell lines. CSPG4 protein expression was successfully deleted in 143B cells using CRISPR/Cas9 technology. Two stable CSPG4-negative cell lines were produced. CSPG4 expression was then reintroduced into negative cell lines, as well as the parental 143B cell line. This created a panel of 6 cell lines with differing CSPG4 expression. Furthermore, siRNA treatment of U2OS, MG63, 143B and U87MG cell lines reduced CSPG4 expression. These cells provided another panel with varying CSPG4 expression for in vitro investigation. In vitro experiments failed to demonstrate a role for CSPG4 in osteosarcoma tumorigenesis. The CRISPR/Cas9 cell panel found that CSPG4 expression did not influence cell proliferation, adhesion and spreading on fibronectin or collagen-I, cell migration, chemosensitivity or anchorage-independent growth. Similarly, the siRNA cell panel found that CSPG4 expression did not influence cell proliferation or anchorage-independent growth. In vivo experimentation did not demonstrate a role for CSPG4 in mediating osteosarcoma tumour growth or metastatic spread. Treatment with a sc-Fv antibody fragment failed to demonstrate specific toxicity of CSPG4-positive cell lines. These results indicate that CSPG4 plays no role in osteosarcoma tumour cell behaviour. However, due to its wide expression pattern it represents a viable therapeutic option for drug targeting.

Introduction: NG2/CSPG4 is a unique transmembrane chondroitin sulphate proteoglycan molecule expressed as a core protein and a chondroitin sulphate proteoglycan (CSPG) up to 400kD. NG2/CSPG4 mediates the communication between the extracellular and intracellular compartments through interactions with collagen VI, growth factors and the actin cytoskeleton. NG2/CSPG4 affects cell migration, spreading, apoptosis and proliferation processes. NG2/CSPG4 has been shown to be expressed in developing and adult cartilage where less is known of its function. I tested the hypothesis: NG2/CSPG4 is an important regulator of chondrocytes function and has the potential to be a therapeutic target for treatment of diseases of cartilage such as osteoarthritis and chondrosarcoma. To do this, I had the following aims: 1) investigate whether different types of chondrocytes show variation in the form or distribution of NG2/CSPG4 expression and 2) through a knockdown approach develop a model to study the functional roles of NG2/CSPG4 in human chondrocytes. Materials and Methods: JJ012, a chondrosarcoma cell line, chondrocytes derived from human articular cartilage and C20/A4 an immortalised chondrocyte cell line were used. NG2/CSPG4 expression was investigated by RT-PCR western blotting, flow-cytometry and immunocytochemistry. NG2/CSPG4 interaction with Golgi complex and endoplasmic reticulum (ER) was assessed by double immunofluorescence. Biochemical interactions were assessed by immunoprecipitation and mass spectroscopy. For NG2/CSPG knockdown, a viral transduction method was carried out using 5 different constructs. Different functional roles of NG2/CSPG4 were investigated. The role of NG2/CSPG4 in gene regulation was studied by shRNA knockdown of NG2/CSPG4 in JJ012 cells and RTPCR. Results: NG2/CSPG4 mRNA was detectable in all cells tested. Western blotting showed expression of only a 270kD core protein in JJ012 and C20A4 cells. Using two different anti NG2/CSPG4 antibodies human OA chondrocytes were seen to express multiple molecular weight forms differentially recognised with and without chondroitinase ABC pre-treatment. Expression of NG2/CSPG4 in JJ012 cells was predominantly membrane associated whilst in OA chondrocytes and C20A4 cells, additional, predominant punctuate cytoplasmic distribution was evident. In OA chondrocytes NG2/CSPG4 co-localised with the Golgi complex and ER. Immunoprecipitation and mass spectrometry data demonstrated associations between NG2/CSPG4 and both collagen VI and thrombopoietin in OA chondrocytes. A model of NG2/CSPG4 gene knockdown was achieved in JJ012 chondrosarcoma cell line, known as B3. B3 cells spread more and migrate less than JJ012 cells; with a significant difference observed in migration (after 10hours: the closed area was 81.4% for JJ012 and 54.6% for B3). There was no difference in cell adhesion to collagen I, II, VI and fibronectin. EGTA inhibited cell adhesion to fibronectin in dose dependent manner with no significant difference observed between both JJ012 and B3 cells. EDTA reduced adhesion of B3 cells but not JJ012 to fibronectin. A significant difference in cell proliferation was detected with no change in apoptosis. Following NG2/CSPG4 knockdown in JJ012 cells there was no difference in expression of aggrecan, collagen II and SOX-9. In contrast, B3 cells showed a decreased expression of MPP3 and ADAMTS-4, a complete loss of ADAMTS-5 and increased expression of MMP13. Conclusions: I have identified altered expression and multiple forms of NG2/CSPG4 in different types of chondrocytes and shown association of this molecule with type VI collagen and thrombopoietin. Creation of a chondrocyte cell line that has stable knockdown of NG2/CSPG4 allowed further investigation of NG2/CSPG function in chondrocytes. NG2/CSPG4 knockdown reduced the cellular migration and proliferation and increased the chondrocyte spreading. The adhesion mechanism in JJ012 appears to be calcium dependent. Loss of NG2/CSPG4 induced changes in expression of aggrecanases and MMPs. Altered expression or associations of NG2/CSPG4 with extracellular ligands or intracellular signalling cascades may be important in the pathogenesis of OA by regulating proteolytic activity or apoptosis related pathways. NG2/CSPG4 is a potential therapeutic target in degenerative and neoplastic diseases of cartilage.

Globally, breast cancer is the leading cause of death in the female population aged 45 and below with a breast cancer incidence reaching 18.1 million in the year 2018. Triple negative breast cancer (TNBC) is part of a group of cancers that lack the expression of Progesterone receptor (PR), Estrogen receptor (ER) and Human epidermal growth factor receptor 2 (HER2). TNBC is commonly associated with early stage metastasis with low survival rates as well as a high frequency of recurrence and proves to be problematic in both the young and elderly female populations. Conventional diagnostic methods for TNBCs include mammography, magnetic resonance imaging (MRI) and ultrasound while therapeutic methods include mastectomy and breast conserving surgery (coupled with radiation therapy). The lack of effective therapeutic options, poor prognostic value and high rates of metastasis, has made treatment of TNBC difficult. The major focus of this work was on the following tumour associated antigens (TAAs): CSPG4 (a transmembrane protein found in 50% of TNBC cases), EGFR (which is overexpressed in 13-76% of TNBCs), and MSLN (which is overexpressed in 67% of TNBCs) as potential targets for monospecific therapy. The evolution of antibody-based immunotherapy strategies has led to applications of single chain variable fragment (scFv) & single domain/nanobody (VHH) antibody formats for diagnostic and therapeutic purposes. In this work, these recombinant antibody fragments have been combined with SNAP-tag, a modified version of the human DNA repair enzyme O6-alkylguanine-DNA-alkyltransferase (AGT), which autocatalytically binds benzyl-guanine modified substrates such as fluorophores or small molecule toxins covalently in a 1:1 stoichiometry. In this study, the primary aim was the comparison of different antibody formats fused to SNAPtag and the potential of these biopharmaceuticals towards immunodiagnosis and therapy of TNBCs. First functionalities of two scFv SNAP fusion proteins and one VHH SNAP fusion protein previously not having been described are provided through binding analyses on receptor positive tumour cell lines. This was achieved by in-silico design and molecular cloning of genetically fused antiCSPG4(scFv), -MSLN(scFv), -MSLN(VHH), -EGFR(scFv) & -EGFR(VHH) to SNAP-tag. The final constructs were confirmed by Sanger sequencing and subsequently transfected into a mammalian vector system (HEK293T) for transient expression of the engineered fusion proteins. Full length protein purified from cell culture supernatant was analysed for diagnostic/therapeutic activities dependant on the substrate attached in the form of a fluorophore or small molecule toxin resulting in recombinant antibody-drug conjugates (ADCs). The study shows promise in providing new immunodiagnostic and therapeutic agents that are specific and less harmful than the current state of the art procedure