CHARACTERIZATION OF P28, A NOVEL ERGIC/CIS-GOLGI PROTEIN, REQUIRED FOR GOLGI RIBBON FORMATION PH MEASUREMENTS IN THE EARLY SECRETORY PATHWAY IN VIVO

Abstract

The secretory pathway of mammalian cells consists of several compartments. Transport between these organelles is accomplished via vesicular carriers or maturation. For non abundant proteins it is thought that transport receptors help the proteins to exit the ER in an effective way. The best characterized mammalian cargo receptor is ERGIC-53, which transports blood coagulation factor V and VIII, cathespin C and Z as well as alpha1-antitrypsin. It localizes to the ER Golgi intermediate compartment (ERGIC) at steady state, but cycles between the organelles of the early secretory pathway. In S. cerevisiae Emp24p, a member of the p24 protein family, was revealed to be a cargo receptor for Gas1p and invertase. I characterized mammalian p28, a γ subfamily member of p24 proteins, which was shown to localize to the ERGIC similar like ERGIC-53. It accumulates therein after Brefeldin A treatment indicating that it cycles. As p24 proteins are known to assemble into complexes I used immunoprecipitation experiments to study the interaction of p28 with other p24 protein members. P28 specifically interacts with p23 and p25. Next, to study the function of the protein, I performed siRNA mediated knockdown experiments. In the absence of p28 the Golgi ribbon is disrupted. However, Golgi ministacks still localize to the perinuclear region. To search for the cause of Golgi fragmentation in p28-depleted cells, I analyzed anterograde and retrograde transport of ts045 VSV-G fusion proteins in p28 knockdown cells. Strikingly, they display normal anterograde as well as retrograde transport. Additionally, the association of COPI coat components with Golgi membranes is thought to be required for compartmentalization. However coatomer redistribution does not seem to be the cause for Golgi fragmentation when p28 is depleted, since I revealed comparable βCOP stainings in control and p28 knockdown cells. Next, I compared my knockdown phenotype with others knowing to result in disruption of Golgi integrity. While knockdowns of ER exit machinery components, SNARE proteins or some tethers give different phenotypes, depletion of GM130 results in a similar phenotype than knocking down p28. Therefore I investigated the distribution of GM130 in p28-depleted cells and assessed an unchanged localization of GM130 to the Golgi. Next, I decided to investigate the Golgi phenotype in more detail. Treating cells with BFA leads to the mergence of Golgi and ER membranes. Subsequent washout of the drug allows the Golgi to reform out of the ER. Lack of p28 rendered the Golgi stacks incompetent to establish a compact ribbon while reforming. Additionally, FRAP experiments revealed that the ministacks are not linked laterally in p28 knockdown cells. Therefore, I concluded that p28 is besides coatomer, tethers and SNARE proteins, necessary for compact Golgi ribbon formation. Cargo receptors have to bind their specific cargoes in the ER recruiting them into forming vesicles. These bud off the ER membrane and transport their content to the ERGIC and further to the Golgi. At the ERGIC, cis-Golgi the cargo proteins have to be released from their receptors, which then recycle back to the ER. How is binding in the ER and release in post ER compartments accomplished? It was proposed that this process depends on pH. Therefore, I was interested to measured pH along the organelles of the early secretory pathway in vivo. I performed pH measurements utilized EGFP as pH sensor in vivo. I targeted EGFP to the ER, ERGIC and Golgi to answer the question if there is gradual acidification along these organelles. I adapted a null-point method to estimate pH ranges in the three organelles. In collaboration with the imaging facility of the Friedrich Miescher Institut (FMI, Basel) customized software was developed to analyze data obtained during pH experiments. Especially the small size and mobility of ERGIC structures required sophisticated data processing. Taken together, the pH of the ER (pH 6.9- pH 7.5) is neutral, the Golgi shows acidic pH (pH 6.4- pH 7.0) and the ERGIC (pH 6.5- pH 7.2) revealed an intermediate ER/Golgi pH. In conclusion, this thesis provides deeper insight into the pH characteristics of the early secretory pathway organelles and shows that the ERGIC protein p28 is required for Golgi integrity.