We have developed a cell-free assay that reproduces vesicular budding during exit from the Golgi complex. The starting preparation for the in vitro system was a rat liver stacked Golgi fraction immobilized on a magnetic solid support by means of an antibody against the cytoplasmic domain of the polymeric IgA receptor. Vesicular budding was ATP, cytosol, and temperature dependent and was inhibited by 1 mMN-ethylmaleimide. Budding was maximum within 10 min and originated preferentially from the trans-Golgi. Exocytic transport vesicles immunoisolated from the total budded population were enriched in the mature forms of secretory and membrane proteins destined to the basolateral plasma membrane and were depleted in lysosomal enzymes and galactosyltransferase activity. The finding that a major proportion (>70%) of newly synthesized, sialylated secretory and transmembrane proteins is contained in a single population of post-Golgi transport vesicles implies that, in a constitutively secreting cell, basolaterally destined proteins are sorted and packaged together into the same exocytic transport vesicles.The hepatocyte, a polarized epithelial cell, lacks storage granules or a known secretagogue, indicating that in this cell only a constitutive secretory pathway exists (1). This pathway delivers newly synthesized proteins to the blood and involves fusion of camer vesicles with the basolateral plasma membrane (PM). Transport of proteins into bile, mediated by fusion of vesicles with the apical PM, is by transcytosis rather than by direct traffic from the Golgi complex (2).Traffic of hepatocyte PM proteins seems to parallel that of secretory proteins. Bartles et al. (3) have presented evidence that PM proteins are initially transported to the basolateral PM, irrespective of their ultimate destination to either the basolateral or the apical PM. However, recent data suggest that sorting of apical and basolateral PM proteins can take place either intracellularly, at the trans-Golgi network (TGN), or at the basolateral PM depending on the protein and the cell type (4-6).Although it is likely that in constitutively secreting cells, PM and secretory proteins are transported from the Golgi to the cell surface in the same transport vesicles, there is currently no direct evidence for this in mammalian cells. Previous electron microscopic studies in hepatoma cells (7) indicated that a secretory protein (albumin) and a viral trans-membrane protein destined to the PM colocalize in structures within the Golgi region, some of which could represent exocytic transport vesicles.Several defined steps of biosynthetic protein traffic have been reconstituted in cell-free assays. Transport from the endoplasmic reticulum to the Golgi complex (8), between compartments of the Golgi complex (9), and from the Golgi complex to the plasma membrane (10, 11) has been investigated. In all of these studies the reporter molecules followed were viral proteins. We have designed a cell-free system that reconstitutes the formation and release of trans...