A point mutant of human sphingosine kinase 1 with increased catalytic activity.

Sphingosine kinase (SK) catalyses the formation of sphingosine 1-phosphate, a lipid second messenger that has been implicated in mediating such fundamental biological processes as cell growth and survival. Very little is currently known regarding the structure or mechanisms of catalysis and activation of SK. Here we have tested the functional importance of Gly(113), a highly conserved residue of human sphingosine kinase 1 (hSK), by site-directed mutagenesis. Surprisingly, a Gly(113)-->Ala substitution generated a mutant that had 1.7-fold greater catalytic activity than wild-type hSK (hSK(WT)). Our data suggests that the Gly(113)-->Ala mutation increases catalytic efficiency of hSK, probably by inducing a conformational change that increases the efficiency of phosphoryl transfer. Interestingly, hSK(G113A) activity could be stimulated in HEK293T cells by cell agonists to a comparable extent to hSK(WT).

An oncogenic role of sphingosine kinase.

Sphingosine kinase (SphK) is a highly conserved lipid kinase that phosphorylates sphingosine to form sphingosine-1-phosphate (S1P). S1P/SphK has been implicated as a signalling pathway to regulate diverse cellular functions [1-3], including cell growth, proliferation and survival [4-8]. We report that cells overexpressing SphK have increased enzymatic activity and acquire the transformed phenotype, as determined by focus formation, colony growth in soft agar and the ability to form tumours in NOD/SCID mice. This is the first demonstration that a wild-type lipid kinase gene acts as an oncogene. Using a chemical inhibitor of SphK, or an SphK mutant that inhibits enzyme activation, we found that SphK activity is involved in oncogenic H-Ras-mediated transformation, suggesting a novel signalling pathway for Ras activation. The findings not only point to a new signalling pathway in transformation but also to the potential of SphK inhibitors in cancer therapy.

Human sphingosine kinase: purification, molecular cloning and characterization of the native and recombinant enzymes.

Sphingosine 1-phosphate (S1P) is a novel lipid messenger that has important roles in a wide variety of mammalian cellular processes including growth, differentiation and death. Basal levels of S1P in mammalian cells are generally low, but can increase rapidly and transiently when cells are exposed to mitogenic agents and other stimuli. This increase is largely due to increased activity of sphingosine kinase (SK), the enzyme that catalyses its formation. In the current study we have purified, cloned and characterized the first human SK to obtain a better understanding of its biochemical activity and possible activation mechanisms. The enzyme was purified to homogeneity from human placenta using ammonium sulphate precipitation, anion-exchange chromatography, calmodulin-affinity chromatography and gel-filtration chromatography. This resulted in a purification of over 10(6)-fold from the original placenta extract. The enzyme was cloned and expressed in active form in both HEK-293T cells and Escherichia coli, and the recombinant E. coli-derived SK purified to homogeneity. To establish whether post-translational modifications lead to activation of human SK activity we characterized both the purified placental enzyme and the purified recombinant SK produced in E. coli, where such modifications would not occur. The premise for this study was that post-translational modifications are likely to cause conformational changes in the structure of SK, which may result in detectable changes in the physico-chemical or catalytic properties of the enzyme. Thus the enzymes were characterized with respect to substrate specificity and kinetics, inhibition kinetics and various other physico-chemical properties. In all cases, both the native and recombinant SKs displayed remarkably similar properties, indicating that post-translational modifications are not required for basal activity of human SK.

Expression of a catalytically inactive sphingosine kinase mutant blocks agonist-induced sphingosine kinase activation. A dominant-negative sphingosine kinase.

Sphingosine kinase (SK) catalyzes the formation of sphingosine 1-phosphate (S1P), a lipid messenger that plays an important role in a variety of mammalian cell processes, including inhibition of apoptosis and stimulation of cell proliferation. Basal levels of S1P in cells are generally low but can increase rapidly when cells are exposed to various agonists through rapid and transient activation of SK activity. To date, elucidation of the exact signaling pathways affected by these elevated S1P levels has relied on the use of SK inhibitors that are known to have direct effects on other enzymes in the cell. Furthermore, these inhibitors block basal SK activity, which is thought to have a housekeeping function in the cell. To produce a specific inhibitor of SK activation we sought to generate a catalytically inactive, dominant-negative SK. This was accomplished by site-directed mutagenesis of Gly(82) to Asp of the human SK, a residue identified through sequence similarity to the putative catalytic domain of diacylglycerol kinase. This mutant had no detectable SK activity when expressed at high levels in HEK293T cells. Activation of endogenous SK activity by tumor necrosis factor-alpha (TNFalpha), interleukin-1beta, and phorbol esters in HEK293T cells was blocked by expression of this inactive sphingosine kinase (hSK(G82D)). Basal SK activity was unaffected by expression of hSK(G82D). Expression of hSK(G82D) had no effect on TNFalpha-induced activation of protein kinase C and sphingomyelinase activities. Thus, hSK(G82D) acts as a specific dominant-negative SK to block SK activation. This discovery provides a powerful tool for the elucidation of the exact signaling pathways affected by elevated S1P levels following SK activation. To this end we have employed the dominant-negative SK to demonstrate that TNFalpha activation of extracellular signal-regulated kinases 1 and 2 (ERK1,2) is dependent on SK activation.

Targeting and insertion of C-terminally anchored proteins to the mitochondrial outer membrane is specific and saturable but does not strictly require ATP or molecular chaperones.

A distinct class of proteins contain a C-terminal membrane anchor and a cytoplasmic functional domain. A subset of these proteins is targeted to the mitochondrial outer membrane. Here, to probe for the involvement of a saturable targeting mechanism for this class of proteins, and to elucidate the roles of chaperone proteins and ATP, we have utilized an in vitro targeting system consisting of in vitro-synthesized proteins and isolated mitochondria. To establish the specificity of targeting we have used a closely related protein pair. VAMP-1A and VAMP-1B are splice variants of the vesicle-associated membrane protein/synaptobrevin-1 (VAMP-1) gene. In intact cells VAMP-1B is targeted to mitochondria whereas VAMP-1A is targeted to membranes of the secretory pathway, yet these isoforms differ by only five amino acids at the extreme C-terminus. Here we demonstrate that, in vitro, VAMP-1B is imported into both intact mitochondria and mitochondrial outer-membrane vesicles with a 15-fold greater efficiency than VAMP-1A. We generated and purified bacterially expressed fusion proteins consisting of the C-terminal two-thirds of VAMP-1A or -1B proteins fused to glutathione S-transferase (GST). Using these fusion proteins we demonstrate that protein targeting and insertion is saturable and specific for the VAMP-1B membrane anchor. To elucidate the role of cytosolic chaperones on VAMP-1B targeting, we also used the purified, Escherichia coli-derived fusion proteins. (33)P-Labelled GST-VAMP-1B(61-116), but not GST-VAMP-1A(61-118), was efficiently targeted to mitochondria in a chaperone-free system. Thus the information required for targeting is contained within the targeted protein itself and not the chaperone or a chaperone-protein complex, although chaperones may be required to maintain a transport-competent conformation. Moreover, ATP was required for transport only in the presence of cytosolic chaperone proteins. Therefore the ATP requirement of transport appears to reflect the participation of chaperones and not any other ATP-dependent step. These data demonstrate that targeting of C-terminally anchored proteins to mitochondria is sequence specific and mediated by a saturable mechanism. Neither ATP nor chaperone proteins are strictly required for either specific targeting or membrane insertion.

A splice-isoform of vesicle-associated membrane protein-1 (VAMP-1) contains a mitochondrial targeting signal.

Screening of a library derived from primary human endothelial cells revealed a novel human isoform of vesicle-associated membrane protein-1 (VAMP-1), a protein involved in the targeting and/or fusion of transport vesicles to their target membrane. We have termed this novel isoform VAMP-1B and designated the previously described isoform VAMP-1A. VAMP-1B appears to be an alternatively spliced form of VAMP-1. A similar rat splice variant of VAMP-1 (also termed VAMP-1B) has recently been reported. Five different cultured cell lines, from different lineages, all contained VAMP-1B but little or no detectable VAMP-1A mRNA, as assessed by PCR. In contrast, brain mRNA contained VAMP-1A but no VAMP-1B. The VAMP-1B sequence encodes a protein identical to VAMP-1A except for the carboxy-terminal five amino acids. VAMP-1 is anchored in the vesicle membrane by a carboxy-terminal hydrophobic sequence. In VAMP-1A the hydrophobic anchor is followed by a single threonine, which is the carboxy-terminal amino acid. In VAMP-1B the predicted hydrophobic membrane anchor is shortened by four amino acids, and the hydrophobic sequence is immediately followed by three charged amino acids, arginine-arginine-aspartic acid. Transfection of human endothelial cells with epitope-tagged VAMP-1B demonstrated that VAMP-1B was targeted to mitochondria whereas VAMP-1A was localized to the plasma membrane and endosome-like structures. Analysis of C-terminal mutations of VAMP-1B demonstrated that mitochondrial targeting depends both on the addition of positive charge at the C terminus and a shortened hydrophobic membrane anchor. These data suggest that mitochondria may be integrated, at least at a mechanistic level, to the vesicular trafficking pathways that govern protein movement between other organelles of the cell.

Regulated exocytosis in vascular endothelial cells can be triggered by intracellular guanine nucleotides and requires a hydrophobic, thiol-sensitive component. Studies of regulated von Willebrand factor secretion from digitonin permeabilized endothelial cells.

To study the intracellular events leading to regulated exocytosis in human umbilical vein endothelial cells (HUVEC) the plasma membrane of HUVEC was selectively permeabilized with digitonin while retaining secretory function. Fusion of Weibel-Palade bodies, the secretory organelle of HUVEC, with the plasma membrane was detected by assaying the media for von Willebrand factor (vWF). The secretion from permeabilized cells faithfully reflects that in intact cells by a number of criteria. First, in the presence of calcium, permeabilized HUVEC secreted vWF with the same kinetics and to the same extent as intact cells stimulated with secretagogue. In addition, the vWF secreted by permeabilized cells after stimulus was exclusively the processed mature form found in Weibel-Palade bodies. Release required micromolar levels of calcium. In addition, GTPgammaS could also stimulate release by a parallel pathway. Both calcium- and GTPgammaS-stimulated secretion required a thiol-sensitive component. The hydrophobic thiol alkylating agent U73122 inhibited calcium-dependent and GTPgammaS-stimulated secretion. Surprisingly, N-ethylmaleimide, a hydrophilic alkylating agent, did not inhibit secretion. The N-ethylmaleimide-sensitive fusion protein (NSF), a protein implicated in a variety of vesicle fusion events, did not appear to be the target of U73122. These data strongly suggests the participation of a non-NSF, membrane-associated protein in regulated secretion in endothelial cells. Further, there appear to be two parallel pathways leading to secretion in HUVEC, one stimulated by elevated levels of calcium and the other mediated by a GTP-binding protein.

The regulated degradation of a 3-hydroxy-3-methylglutaryl-coenzyme A reductase reporter construct occurs in the endoplasmic reticulum.

The rate-limiting enzyme in cholesterol biosynthesis, 3-hydroxy-3-methylglutaryl-coenzyme A (HMG CoA) reductase, is regulated at a number of levels. One important mechanism is regulation of the half-life of the protein by a controlled proteolytic system. This comes about in response to downstream products of the sterol biosynthetic pathway. Little is known about this system, including where in the cell this regulated degradation occurs. HMG CoA reductase resides in the endoplasmic reticulum. To localize the site of regulated degradation of HMG CoA reductase, we used a construct that fuses the N-terminal membrane-anchoring domain of HMG CoA reductase in-frame with beta-galactosidase as a reporter domain (HM-Gal). HM-Gal has previously been shown to reproduce faithfully the degradative properties of native HMG CoA reductase (Chun et al. (1990) J. Biol. Chem. 265, 22004-22010). CHO cells transfected with DNA encoding HM-Gal were exposed to mevalonic acid, which enhances the rate of HMG CoA reductase degradation several fold, and leads to the reduction of the steady state levels of HM-Gal by 80-90%. To accumulate HMG CoA reductase at the site of degradation, cells were simultaneously treated with N-acetyl-leucyl-leucyl-norleucinal (ALLN), which inhibits the protease responsible for reductase degradation. HM-Gal was localized morphologically by immunofluorescence and biochemically by measuring beta-galactosidase activity in Percoll gradients of cellular homogenates. Using either technique HM-Gal localization was indistinguishable from that of ER markers in both control cells and in cells treated to accumulate HMG CoA reductase at the site of degradation.(ABSTRACT TRUNCATED AT 250 WORDS)

The efficiency and kinetics of secretion of apolipoprotein A-I in hepatic and non-hepatic cells.

Apo A-I, the major protein component of high density lipoprotein (HDL), is synthesized by hepatic and intestinal cells and assembled with lipids to produce, in as yet incompletely understood ways, a mature HDL particle. For many secreted proteins only a portion of newly synthesized polypeptides are secreted, with the remainder being degraded at intracellular sites. For example apolipoprotein B secretion is controlled by the extent of intracellular degradation of the protein. Here we have systematically examined whether there is significant intracellular degradation of nascent apo A-I. We find that in two hepatic cell types, primary cultures of hepatocytes from cynomolgus monkey and HepG2 hepatocarcinoma cells, essentially all apo A-I that is synthesized is eventually secreted. A non-hepatic cell line, Chinese hamster ovary cells transfected with the apo A-I gene, secreted somewhat less (65%) of the apo A-I synthesized. In a careful kinetic analysis, the rate of apo A-I secretion was found to be identical between the three cell types. This indicates that the mechanisms governing secretion are conserved among the different cell types. Further, the rate of secretion was the same for apo A-I in a lipid-poor form and in a form found associated in the medium with sufficient lipid to promote flotation in density gradients. The kinetic analysis indicates that there are two rate limiting steps to apo A-I secretion from the cell. It has previously been suggested that, for most proteins, exit from the endoplasmic reticulum is the rate limiting step in the secretory pathway.(ABSTRACT TRUNCATED AT 250 WORDS)

Toxin resistance and reduced secretion in a mouse L-cell mutant defective in herpes virus propagation.

The mouse L-cell mutant gro29 was selected originally for its inability to propagate herpes simplex virus; it shows severe defects in virus egress and the transport and processing of viral glycoproteins after infection. In this report, we show that uninfected gro29 cells display pleiotropic changes in protein secretion, oligosaccharide processing, and sensitivity to the toxins ricin and modeccin. Specifically, the rate of secretion of a nonglycosylated protein, human growth hormone, was reduced 70% in gro29 cells compared with the parental L cells. A direct measurement of the transport capacity of Golgi membranes in a cell-free assay suggests that gro29 cells contain less functional Golgi than parental cells. Despite this deficiency, N-linked oligosaccharides were processed efficiently in mutant cells, although there were differences in the structure of the mature forms. Lectin intoxication assays revealed that gro29 cells were cross-resistant to killing by the cytotoxic lectins ricin and modeccin, but not to wheat germ agglutinin, Ricinus communis agglutinin RCA120, or leucoagglutinin. Fluorescence labeling using fluorescein-conjugated lectins showed that uninfected gro29 cells expressed relatively few ricin-binding molecules, suggesting a possible mechanism for toxin resistance. These studies provide evidence that the processes of protein secretion, lectin intoxication, and herpes virus maturation and egress may share a common cellular component.

The activity of Golgi transport vesicles depends on the presence of the N-ethylmaleimide-sensitive factor (NSF) and a soluble NSF attachment protein (alpha SNAP) during vesicle formation.

An assay designed to measure the formation of functional transport vesicles was constructed by modifying a cell-free assay for protein transport between compartments of the Golgi (Balch, W. E., W. G. Dunphy, W. A. Braell, and J. E. Rothman. 1984. Cell. 39:405-416). A 35-kD cytosolic protein that is immunologically and functionally indistinguishable from alpha SNAP (soluble NSF attachment protein) was found to be required during vesicle formation. SNAP, together with the N-ethylmaleimide-sensitive factor (NSF) have previously been implicated in the attachment and/or fusion of vesicles with their target membrane. We show that NSF is also required during the formation of functional vesicles. Strikingly, we found that after vesicle formation, the NEM-sensitive function of NSF was no longer required for transport to proceed through the ensuing steps of vesicle attachment and fusion. In contrast to these functional tests of vesicle formation, SNAP was not required for the morphological appearance of vesicular structures on the Golgi membranes. If SNAP and NSF have a direct role in transport vesicle attachment and/or fusion, as previously suggested, these results indicate that these proteins become incorporated into the vesicle membranes during vesicle formation and are brought to the fusion site on the transport vesicles.

Vesicle fusion in protein transport through the Golgi in vitro does not involve long-lived prefusion intermediates. A reassessment of the kinetics of transport as measured by glycosylation.

The well-characterized cell-free assay measuring protein transport between compartments of the Golgi [Balch, W. E., Dunphy, W. G., Braell, W. A., & Rothman, J. E. (1984) Cell 39, 405-416] utilizes glycosylation of a glycoprotein to mark movement of that protein from one Golgi compartment to the next. Glycosylation had been thought to occur immediately after vesicles carrying the glycoprotein fuse with their transport target. Therefore, the kinetics of glycosylation were taken to reflect the kinetics of vesicle fusion. We previously isolated and raised monoclonal antibodies against a protein (the prefusion operating protein, POP) which is required in this assay at a step after vesicles have apparently been formed and interacted with the target membranes, but long before glycosylation takes place. This was therefore presumed to be a reaction involving targeted but unfused vesicles. Here we report that POP is identical to uridine monophosphokinase, as revealed by molecular cloning. We show that POP is not active in transport per se but instead enhances the glycosylation used to mark transport. This indicated that, contrary to previous assumptions, glycosylation might lag significantly behind vesicle fusion. We directly show this to be true. This alters the interpretation of several earlier studies. In particular, the previously reported existence of a late, prefusion intermediate, the "NEM-resistant intermediate", can be seen to be due to effects on glycosylation and not indicative of true fusion events.

Primaquine blocks transport by inhibiting the formation of functional transport vesicles. Studies in a cell-free assay of protein transport through the Golgi apparatus.

The lysosomotropic amine primaquine has previously been shown to inhibit both secretory and recycling processes of cells in culture. We have used a cell-free assay that reconstitutes glycoprotein transport through the Golgi apparatus to investigate the mechanism of action of primaquine. In this assay, primaquine inhibits protein transport at a half-maximal concentration of 50 microM, similar to the concentration previously reported to disrupt protein secretion in cultured cells. Kinetic analysis of primaquine inhibition indicates that its point of action is at an early step in the vesicular transport mechanism. Primaquine does not inhibit the fusion of vesicles already attached to their target membranes. Primaquine irreversibly inactivates the membranes that form transport vesicles (donor), but not the membranes that are the destination of those vesicles (acceptor). Morphological data indicate that primaquine inhibits the budding of vesicles from the donor membranes. Once formed, the vesicles are refractile to primaquine action, and their attachment to and fusion with acceptor membranes proceeds unimpeded. In addition to illuminating the mechanism of action of primaquine, this study suggests that the selective action of this agent will make it a useful tool in the study of the formation of transport vesicles.

Analysis of protein transport through the Golgi in a reconstituted cell-free system.

The processes which transport membrane proteins between compartments of the Golgi apparatus have been reconstituted in vitro using isolated Golgi fractions. This cell-free system allows a detailed analysis of protein transport not possible in intact cells. Transport of the membrane glycoprotein (G protein) of vesicular stomatitis virus (VSV) is measured from a "donor" to an "acceptor" Golgi fraction. The donor Golgi fraction is prepared from VSV-infected Chinese hamster ovary (CHO) mutant cells deficient in the glycosylation enzyme N-acetylglucosamine transferase I. "Acceptor" is prepared from uninfected wild-type CHO cells. Transport is measured by the addition of N-acetylglucosamine to G protein, which can occur only upon movement of G protein from donor to acceptor. Transport requires physiological pH and osmolarity, is dependent on nucleotide triphosphates, and is mediated by proteins both from cytosol and on the Golgi membranes. Protein movement is inhibited by the non-hydrolyzable GTP analogue, GTP gamma S. The process of transport proceeds through the budding, pinching off, targeting, and fusion of transport vesicles. In this system these vesicles are initially coated with a non-clathrin coat and are targeted with this coat intact. Several of the proteins which mediate transport have been characterized, and isolated to homogeneity. The successful development of this assay has led to the formulation of cell free assays for protein transport between other compartments. Comparison of these systems indicates that some common mechanisms of vesicular movement are used in transport between a variety of membrane compartments.

Glycolipid and glycoprotein transport through the Golgi complex are similar biochemically and kinetically. Reconstitution of glycolipid transport in a cell free system.

Glycolipid transport between compartments of the Golgi apparatus has been reconstituted in a cell free system. Transport of lactosylceramide (galactose beta 1-4-glucose-ceramide) was followed from a donor to an acceptor Golgi population. The major glycolipid in CHO cells is GM3 (sialic acid alpha 2-3 galactose beta 1-4-glucose-ceramide). Donor membranes were derived from a Chinese hamster ovary (CHO) cell mutant (Lec2) deficient in the Golgi CMP-sialic acid transporter, and therefore contained lactosylceramide as the predominant glycolipid. Acceptor Golgi apparatus was prepared from another mutant, Lec8, which is defective in UDP-Gal transport. Thus, glucosylceramide is the major glycolipid in Lec8 cells. Transport was measured by the incorporation of labeled sialic acid into lactosylceramide (present originally in the donor) by transport to acceptor membranes, forming GM3. This incorporation was dependent on ATP, cytosolic components, intact membranes, and elevated temperature. Donor membranes were prepared from Lec2 cells infected with vesicular stomatitus virus (VSV). These membranes therefore contain the VSV membrane glycoprotein, G protein. Donor membranes derived from VSV-infected cells could then be used to monitor both glycolipid and glycoprotein transport. Transport of these two types of molecules between Golgi compartments was compared biochemically and kinetically. Glycolipid transport required the N-ethylmaleimide sensitive factor previously shown to act in glycoprotein transport (Glick, B. S., and J. E. Rothman. 1987. Nature [Lond.]. 326:309-312; Rothman, J. E. 1987. J. Biol. Chem. 262:12502-12510). GTP gamma S inhibited glycolipid and glycoprotein transport similarly. The kinetics of transport of glycolipid and glycoprotein were also compared. The kinetics of transport to the end of the pathway were similar, as were the kinetics of movement into a defined transport intermediate. It is concluded that glycolipid and glycoprotein transport through the Golgi occur by similar if not identical mechanisms.

The molecular control of transport vesicle fusion.

The fusion of transport vesicles with the appropriate target membrane in constitutive transport is a complex and well-controlled process. Many of the molecular details of the reactions that result in this control are being revealed through the use of cell-free assays of protein transport as well as by the study of the molecular genetics of secretion in yeast. Kinetic analyses have indicated that several structural intermediates are formed after transport vesicles attach to their destination, but before they fuse with the appropriate membrane. Proteins that mediate the formation and processing of these intermediates have been identified. Included among these are small molecular weight GTP-binding proteins. This intricate set of reactions may ensure the fidelity of transport and guard the integrity of the organelles along the transport pathway.

Low temperature blocks exit of pro-opiomelanocortin from the endoplasmic reticulum but not subsequent delivery to the site of prohormone processing.

The effect of reduced temperature on the delivery of the prohormone pro-opiomelanocortin (POMC) to the site of prohormone processing was investigated in the mouse anterior pituitary cell line AtT20. At 20 degrees C processing was substantially inhibited and was almost completely arrested at 18 degrees C. Earlier studies with membrane glycoproteins indicated that at these temperatures protein movement was blocked at the level of exit from the Golgi apparatus. In contrast it was found here that the inhibition of processing at reduced temperature was due to the retention of POMC in the endoplasmic reticulum. When POMC was allowed to progress to the Golgi before temperature was reduced, subsequent processing was only slightly retarded by incubation at 18 degrees C. This indicates either that Golgi exit is not inhibited at this temperature, or that the processing apparatus exists in the Golgi. A surprising incidental result was that when held in the endoplasmic reticulum at low temperature POMC is apparently subject to post-translational N-linked glycosylation.

Identification of a 25-kD protein from yeast cytosol that operates in a prefusion step of vesicular transport between compartments of the Golgi.

We have identified a 25-kD cytosolic yeast protein that mediates a late, prefusion step in transport of proteins between compartments of the Golgi apparatus. Activity was followed using the previously described cell free assay for protein transport between Golgi compartments as modified to detect late acting cytosolic factors (Wattenberg, B. W., and J. E. Rothman. 1986. J. Biol. Chem. 263:2208-2213). In the reaction mediated by this protein, transport vesicles that have become attached to the target membrane during a preincubation are processed in preparation for fusion. The ultimate fusion event does not require the addition of cytosolic proteins (Balch, W. E., W. G. Dunphy, W. A. Braell, and J. E. Rothman. 1984. Cell. 39:525-536). Although isolated from yeast, this protein has activity when assayed with mammalian membranes. This protein has been enriched over 150-fold from yeast cytosol, albeit not to complete homogeneity. The identity of a 25-kD polypeptide as the active component was confirmed by raising monoclonal antibodies to it. These antibodies were found to specifically inhibit transport activity. Because this is a protein operating in prefusion, it has been abbreviated POP.

Yeast and mammals utilize similar cytosolic components to drive protein transport through the Golgi complex.

Vesicular transport between successive compartments of the mammalian Golgi apparatus has recently been reconstituted in a cell-free system. In addition to ATP, transport requires both membrane-bound and cytosolic proteins. Here we report that the cytosol fraction from yeast will efficiently substitute for mammalian cytosol. Mammalian cytosol contains several distinct transport factors, which we have distinguished on the basis of gel filtration and ion-exchange chromatography. Yeast cytosol appears to contain the same collection of transport factors. Resolved cytosol factors from yeast and mammals complement each other in a synergistic manner. These findings suggest that the molecular mechanisms of intracellular protein transport have been conserved throughout evolution. Moreover, this hybrid cell-free system will enable the application of yeast genetics to the identification and isolation of cytosolic proteins that sustain intracellular protein transport.