Caveolae are nanoscopic bulb-shaped invaginations of the plasma membrane involved in numerous cellular processes including lipid metabolism, mechanosensation, mechanoprotection and signal transduction. Caveola biogenesis involves two classes of proteins, caveolins [Caveolin1 (CAV1), caveolin2 (CAV2) and caveolin3 (CAV3)] and cavins [cavin1, 2,3 and 4]. Caveolins are integral membrane proteins with CAV1 and CAV2 expressed in all tissue types while CAV3 is a muscle specific isoform of caveolin. Cavins are peripheral membrane proteins that form a coat on the cytosolic surface of caveolae. CAV1 and cavin1 are both required for formation of caveolae in all cell types. In contrast, cavin2 co-operates in caveola formation in lung endothelia while cavin3 is involved in signalling through caveolae. Cavin4 is muscle specific isoform of cavin and mutations in cavin4 gene lead to hypertrophic cardiomyopathy. Loss and/or mutation in cavins and caveolins are associated with progression of muscle dystrophy and generalised lipodystrophy. The molecular mechanisms of disease development and progression upon caveola disruption are unclear due to a lack of structural information about the protein components of caveolae. Cavin proteins assemble into homo- and hetero-oligomeric complexes in the cytosol and associate with the plasma membrane in the presence of CAV1 to form characteristic caveola membrane invaginations. Caveolae respond to changes in membrane tension due to mechanical stretch or oxidative stress by releasing cavin complexes into the cytosol. The role of cavin proteins in the formation of caveolae and the importance of cavin release into the cytosol is a poorly understood aspect of caveola biology, which is crucial to understand molecular mechanisms underlying diseases related to caveolae. Based on this, the major aims of my thesis were to 1) understand the molecular mechanism of cavin complex assembly, and 2) investigate the fate of cavin complexes after release into the cytosol. In Chapter 2, I contributed to work that identified and characterised the structural domains in the cavin family proteins. The structural domain organisation of cavins consists of two α-helical regions (HR) called HR1 and HR2 domain flanked by three disordered regions DR1, DR2 and DR3. Crystal structures of cavin HR1 domains showed a trimeric coiled-coil organisation that is essential for homo- and hetero-oligomerisation of cavin proteins. The HR1 domain showed a preferential phosphoinositide (PI) binding while HR2 domain bound to phosphatidylserine in vitro. The trimeric organisation of the HR1 domain is essential for folding of the HR2 domain. Electron microscopy analysis of cavin proteins showed a characteristic rod like arrangement that was proposed to form the striated coat on the cytosolic face of caveolae. In Chapter 3, I characterised ubiquitylation as a major post-translational modification of cavin1 and identified functions of the phosphoinositide (PI) binding site of cavin1 in vivo. Cavin1 was shown to undergo proteasomal degradation, and lysine residues in the PI-binding region of cavin1 were shown to acts a major site for cavin1 ubiquitylation. Mechanical stretching of cells caused cavin1 release from caveolae membranes, leading to exposure of PI-binding lysine residues to cytosol that triggered ubiquitylation and subsequent cavin1 degradation. This novel regulatory mechanism helps to maintain low levels of cellular cytosolic cavin1 in absence of its putative cytosolic or nuclear signalling interacting proteins. Research work in Chapter 4 showed that the E3 ubiquitin ligase HUWE1 (HECT, UBA and WWE domain containing 1) ubiquitylates cavin1 in response to oxidative stress induced by the addition of hydrogen peroxide. Chapter 5 systematically characterised the distinct roles of cavin1 structural domains in cavin oligomeric complex assembly. First I showed that both N and C-terminal domains of cavin1 are required for its ability to generate caveolae, and appeared to promote formation of larger oligomers from core cavin1 trimers. Examining the recruitment of other cavins demonstrated the central importance of cavin1, where formation of hetero-oligomers with cavin2, cavin3 or cavin4 through the HR1 domain was found to be essential for their localisation to caveolae. Subsequently, fluorescence correlation spectroscopy analysis showed that a unique stretch of undecad repeat residues in the cavin1 HR2 domain plays a major role in cavin1 trimer–trimer association and localisation to caveolae. Interestingly, introducing these undecad repeats into the cavin2 HR2 domain increased membrane affinity and caused extensive membrane tubulation suggesting an important role for undecad repeats in membrane attachment and remodelling. These studies lead to a novel model for cavin1 higher order oligomerisation and hierarchical cavin coat assembly essential for caveola formation. Overall this thesis work has made a substantial contribution towards understanding the mechanism of cavin coat formation on caveolae and proposed a novel feedback mechanism through ubiquitylation to regulate the cytosolic pool of cavins acting as signalling molecules.