Caveolae are small plasma membrane invaginations present in many different cell types, which have been linked to diverse cellular functions, including cell signalling, membrane rearrangements and lipid regulation. The caveolae markers, members of the caveolin family of proteins, are essential for caveola formation and function. Recently, however, a protein named PTRF (Polymerase I and Transcript Release Factor) or cavin, originally identified as a nuclear factor that regulates transcription in vitro, was shown to be associated with caveolae in adipocytes. In the first chapter of this thesis, I have used the zebrafish Danio rerio to investigate the relation of PTRF/cavin to caveolae as well as caveola function in vivo. During zebrafish development, PTRF/cavin was highly expressed in the notochord in 18 h, 24 h and 35 h post-fertilization embryos, as detected by in situ hybrydization. Analysis of later development stages showed that PTRF/cavin is also present in the otic vesicle, brachial arches, and periderm. Disruption of PTRF/cavin expression, via morpholino-mediated inhibition, caused severely defective development of the notochord as well as heart edema, in a dose-dependent manner. PTRF/cavin knockdown embryos had curved notochords and were shorter than the controls. Examination of the notochord by electron microscopy showed that the number of caveolae was greatly reduced in PTRF/cavin-morpholino-injected embryos. Similar effects were observed when caveolin-1, the major protein of caveolae in non-muscle cells, was down-regulated. Altogether, these results indicated a role for PTRF/cavin during formation and/or stabilization of caveolae as well as an essential role for caveolae during zebrafish embryo development. Combined with results obtained in mammalian cells, these findings identify PTRF/cavin as the first component of a caveolar coat, required for caveola formation and function (Hill et al., 2008). We subsequently identified a family of PTRF/cavin-related proteins, the cavins, that all associate with caveolae. Using biochemistry, light microscopy, and FRET-based approaches we characterised PTRF/cavin and the new members of this family of proteins SDR/cavin-2, SRBC/cavin-3 and MURC/cavin-4. We have shown that the four members of the cavin family form a multi-protein complex that associates with caveolae. This complex can constitutively assemble in the cytosol and then associate with caveolin at the plasma membrane caveolae; interestingly, caveolin is essential for the plasma membrane translocation of the cavin complex, and in caveolin-1 knockout cells the four cavin proteins are restricted to the cytosol. PTRF/cavin-1, but not other cavins, can induce caveola formation in a heterologous system and is required for the recruitment of the cavin complex to caveolae. The four cavin proteins present distinct patterns of tissue expression, which suggests that caveolae may perform tissue-specific functions regulated by the composition of the cavin complex. MURC/cavin-4 is expressed predominantly in muscle and its distribution is perturbed in human muscle disease associated with caveolin-3 dysfunction, identifying MURC/cavin-4 as a novel muscle disease candidate caveolar protein. To functionally investigate the relation of cavins and caveolae, we explored a caveolar function in mechanosensation. Through the use of hypo-osmotic media, we induced membrane-stretch and showed that the increased membrane tension leads to dissociation of the caveolin-cavin module and caveola disassembly as observed by immunofluorescence and FLIM/FRET techniques. Once released from caveolae, caveolin was seen internalized in late endosomes and lysosomes. Cavin-1, on the other hand, was found to be diffused in the cytosol and from there it was translocated to the nuclear compartment. The nuclear translocation was observed in several different cell types, which suggests a universal role for nuclear cavin-1, and was independent of caveolin expression. Analysis of live cells using real-time FLIM/FRET showed that cells quickly respond to variations in membrane tension by dissociation/re-association of caveolin and cavin-1. Altogether, in the course of this project, I was able to show that cavin-1 is an essential regulator of caveola biogenesis in cultured cells and in vivo. Cavin-1 and the other members of the PTRF/Cavin family form a multiprotein complex that is recruited to caveolae by caveolin and coats plasma membrane caveolae. The association between cavin-1 and caveolin is crucial for caveolae assembly and this interaction has a role in the cellular sensation of plasma membrane tension. Under high membrane tensions, caveolin and cavin-1 dissociate with the consequent flattening of caveolae. Under these circumstances, caveolin is internalized into enlarged endosomes and lysosomes while cavin-1 is translocated to the nucleus, identifying for the first time a caveola- to nucleus signalling pathway. The exact role of nuclear cavin-1 under plasma membrane stretch is now amenable to analysis.

Increased secretion of very low-density lipoprotein (VLDL), a triglyceride-rich lipoprotein, by the liver causes hypertriglyceridemia, which is a major risk factor for the development of atherosclerosis. The rate of VLDL-secretion from the liver is determined by its controlled transport from the endoplasmic reticulum (ER) to the Golgi. The ER-to-Golgi transport of newly synthesized VLDL is a complex multi-step process and is mediated by the VLDL transport vesicle (VTV). Once a nascent VLDL particle is synthesized in the lumen of the ER, it triggers the process of VTV-biogenesis and this process requires coat complex II (COPII) proteins that mediate the formation of classical protein transport vesicles (PTV). Even though, both VTV and PTV bud off the same ER at the same time and require the same COPII proteins, their cargos and sizes are different. The VTV specifically exports VLDL to the Golgi and excludes hepatic secretory proteins such as albumin and the size of the VTV is larger (~ 100 -120 nm) than PTV to accommodate VLDL-sized particles. These observations indicate (i) the existence of a sorting mechanism at the level of the ER; and (ii) the involvement of proteins in addition to COPII components. This doctoral thesis is focused on identification of proteins regulating VLDL sorting into the VTV and involved in the biogenesis of the VTV. In order to identify proteins present exclusively in VTV, we have characterized the proteome of VTV, which suggest CideB (cell death-inducing DFF45-like effector b) and SVIP (small VCP/P97 interacting protein) as candidates, present in VTV but excluded from PTV. We further confirmed the finding by performing co-immunoprecipitation studies and confocal microscopy studies. CideB, a 26-kDa protein was found to interact with apolipoprotein iv B100 (apoB 100), the structural protein of VLDL. Moreover, CideB interacts with two of the COPII components, Sar1 and Sec24. VTV generation was examined after blocking CideB by specific antibodies and by silencing CideB in rat primary hepatocytes. Knockdown of CideB in primary hepatocytes showed significant reduction in VTV generation, however, CideB was concentrated in VTV as compared with the ER suggesting its functional role in the sorting of VLDL into the VTV. SVIP, a small (~ 9-kDa) protein was found to interact with Sar1, a COPII component that initiates the budding of vesicles from ER membrane. SVIP has sites for myristoylation and we found increased recruitment of SVIP on ER membrane upon myristic acid (MA) treatment. Sar1 that lacks sites for myristoylation also is recruited more on ER upon myristoylation indicating that SVIP promotes Sar1 recruitment on ER. Additionally, our data suggest that Sar1 interacts with SVIP and forms a multimer that facilitates the biogenesis of VTV. Interestingly, silencing of SVIP reduced the VTV generation significantly. Conversely, incubation with MA increased the VTV budding, suggesting recruitment of SVIP on ER surface facilitates the VTV budding. We conclude that SVIP recruits Sar1 on ER membrane and makes an intricate COPII coat leading to the formation of a large vesicle, the VTV. Overall, the data presented in this thesis, determines the role of CideB and SVIP in regulating VLDL sorting and VTV biogenesis.