In vitro protein translocation across the yeast endoplasmic reticulum: ATP-dependent posttranslational translocation of the prepro-alpha-factor

The in vitro synthesized precursor of the alpha-factor pheromone, prepro-alpha-factor, of Saccharomyces cerevisiae was translocated across yeast microsomal membranes in either a homologous or a wheat germ cell free system. Translocated prepro-alpha-factor was glycosylated, sedimented with yeast microsomal vesicles, and was protected from digestion by added protease, but was soluble after alkaline sodium carbonate treatment. Thus prepro-alpha-factor was properly sequestered within yeast microsomal vesicles, but was not integrated into the lipid bilayer. In marked contrast to protein translocation across mammalian microsomal membranes, translocation of prepro-alpha-factor across yeast microsomal membranes could occur posttranslationally. This reaction required protein components in the yeast microsomal fraction that could be inactivated by alkylation or proteolysis, was ATP-dependent, and was insensitive to the presence of a variety of uncouplers and ionophores.     

Secretion in yeast: translocation and glycosylation of prepro-alpha-factor in vitro can occur via an ATP-dependent post-translational mechanism

In an in vitro system comprising a yeast cell-free translation system, yeast microsomes and mRNA encoding prepro-alpha-factor, the translocation of this protein across the membrane of the microsomal vesicle and its glycosylation could b uncoupled from its translation. Such post-translational processing is dependent upon the presence of ATP in the system. It is not, however, affected by a variety of uncouplers, ionophores or inhibitors, including carbonyl cyanide m-chlorophenyl hydrazone (CCCP), valinomycin, nigericin, dinitrophenol (DNP), potassium cyanide (KCN) or N-ethyl maleimide (NEM). This mechanism of translocation is significant as it indicates that a protein of 18 000 daltons is capable of crossing an endoplasmic reticulum-derived membrane post-translationally. For the moment, this phenomenon seems to be restricted to prepro-alpha-factor in the yeast in vitro system. Neither invertase nor IgG chi light chain could be translocated post-translationally in yeast, nor was such processing observed for prepro-alpha-factor in a wheat germ system supplemented with canine pancreatic microsomes.     

Protein translocation across the yeast microsomal membrane is stimulated by a soluble factor

We have found that a soluble activity present in the postribosomal supernatant fraction of Saccharomyces cerevisiae stimulates posttranslational translocation of yeast prepro-alpha-factor across yeast microsomal membranes. Stimulation of translocation is not due to a nonspecific affect on ATP levels. The activity is likely to be due to protein(s) as it is destroyed by N-ethylmaleimide, protease, or heat treatment but not by incubation with RNase. Its apparent sedimentation coefficient is approximately 9.6 S.     

Secretion in yeast. Purification and in vitro translocation of chemical amounts of prepro-alpha-factor

The Saccharomyces cerevisiae mating pheromone precursor, prepro-alpha-factor, can be translocated across yeast endoplasmic reticulum membranes post-translationally in an in vitro system. This characteristic makes prepro-alpha-factor potentially useful as a probe in the biochemical dissection of the mechanism of this basic cellular process. Efforts have been limited by the inability to isolate sufficient quantities of such secretory protein precursors in a translocation-competent form. We report here the one-step purification of chemical amounts of translocation-competent prepro-alpha-factor using nickel ion affinity chromatography on nitrilotriacetate resin. An oligonucleotide encoding 6 histidine residues was inserted into a genomic clone encoding prepro-alpha-factor 5' of the naturally occurring translational stop codon by site-directed mutagenesis. The construct was expressed at high levels in a SecY- strain of Escherichia coli. The produced preprotein was solubilized in 6 M guanidine hydrochloride and bound to nitrilotriacetate resin. Prepro-alpha-factor was recovered at a purity in excess of 95% by elution with 0.25 M imidazole, 8 M urea, which competitively displaced the histidine affinity tag from the nickel column. The chemical amounts of prepro-alpha-factor obtained in this way were determined to be competent for translocation across yeast microsomal membranes and for subsequent modifications such as signal sequence cleavage and N-linked glycosylation.     

Reconstitution of protein translocation from solubilized yeast membranes reveals topologically distinct roles for BiP and cytosolic Hsc70

We reconstituted prepro-alpha-factor translocation and signal peptide processing using a yeast microsomal detergent soluble fraction formed into vesicles with soybean phospholipids. Reconstituted translocation required ATP, and was deficient when sec63 and kar2 (BiP) mutant cells were used as a source of membranes. Normal translocation was observed with vesicles reconstituted from a mixture of pure wild-type yeast BiP and a soluble fraction of kar2 mutant membranes. Two other heat-shock cognate (hsc) 70 homologs, yeast cytosolic hsc70 (Ssalp) and E. coli dnaK protein did not replace BiP. Conversely, BiP was not active under conditions where translocation into native ER vesicles required cytosolic hsc70. We conclude that cytosolic hsc70 and BiP serve noninterchangeable roles in polypeptide translocation, possibly because distinct, asymmetrically oriented membrane proteins are required to recruit each protein to opposing surfaces of the ER membrane.     

Signal recognition particle (SRP) stabilizes the translocation-competent conformation of pre-secretory proteins.

When affinity-purified proOmpA was diluted out of 8 M urea into a sample of yeast microsomes, it was translocated and processed in the absence of any cytosolic factors; an intact membrane and ATP were the only requirements. The translocation competence of proOmpA was lost, however, during a 15-h incubation at 0 degrees C. The competence was retained when trigger factor and a yeast cytosolic extract were present during incubations at 0 degrees C. The same reactions were carried out with affinity-purified prepro-alpha-factor, and the same results were obtained with the exception that trigger factor was not required. When the various cytosolic factors were replaced with SRP, the addition of yeast microsomes after 15 h resulted in the translocation and processing (and glycosylation) of both proOmpA and prepro-alpha-factor. Pancreatic microsomes were also used in this type of assay, and it was found that proOmpA (but not prepro-alpha-factor) could be translocated when diluted out of urea. In this case, as with yeast microsomes, translocation competence was maintained by SRP. These results show that in addition to a recognition and targeting function, SRP can stabilize the translocation-competent conformation of pre-secretory proteins in vitro for translocation across eukaryotic membranes.     

Post-translational transport of proteins into microsomal membranes of Candida maltosa.

We have isolated from the yeast Candida maltosa microsomal membranes that are active in the translocation of proteins synthesized in cell-free systems derived from C. maltosa, Saccharomyces cerevisiae or wheat germ. Translocation and core glycosylation of prepro-alpha-factor, a secretory protein, were observed with yeast microsomes added during or after translation. The signal peptide is cleaved off. Cytochrome P-450 from C. maltosa, the first integral membrane protein studied in a yeast system, is also inserted both co- and post-translationally into Candida microsomal membranes. Its insertion into canine microsomes occurs efficiently only in a co-translational manner and is dependent on the function of the signal recognition particle.     

Subcellular location of enzymes involved in the N-glycosylation and processing of asparagine-linked oligosaccharides in Saccharomyces cerevisiae

A particulate translation system isolated from the yeast Saccharomyces cerevisiae was shown to translate faithfully in-vitro-transcribed mRNA coding for a mating hormone precursor (prepro-alpha-factor mRNA) and to N-glycosylate the primary translation product after its translocation into the lumen of the microsomal vesicles. Glycosylation of its three potential sugar attachment sites was found to be competitively inhibited by acceptor peptides containing the consensus sequence Asn-Xaa-Thr, supporting the view that the glycan chains are N-glycosidically attached to the prepro-alpha-factor polypeptide. The accumulation in the presence of acceptor peptides of a membrane-specific, unglycosylated translation product (pp-alpha-F0) differing in molecular mass from a cytosolically located, protease-K-sensitive alpha-factor polypeptide (pp-alpha-Fcyt) by about 1.3 kDa, suggests that, in contrast to previous reports, a signal sequence is cleaved from the mating hormone precursor on/after translocation. This conclusion is supported by the observation that the multiply glycosylated alpha-factor precursor is cleaved by endoglucosaminidase H to a product with a molecular mass smaller than the primary translation product pp-alpha-Fcyt but larger than the membrane-specific pp-alpha-F0. Translation and glycosylation experiments carried out in the presence of various glycosidase inhibitors (e.g. 1-deoxynojirimycin, N-methyl-1-deoxynojirimyin and 1-deoxymannojirimycin) indicate that the N-linked oligosaccharide chains of the glycosylated prepro-alpha-factor species are extensively processed under the in vitro conditions of translation. From the specificity of the glycosidase inhibitors applied and the differences in the molecular mass of the glycosylated translation products generated in their presence, we conclude that the glycosylation-competent microsomes contain trimming enzymes, most likely glucosidase I, glucosidase II and a trimming mannosidase, which process the prepro-alpha-factor glycans down to the (Man)8(GlcNAc)2 stage. Furthermore, several arguments strongly suggest that these three enzymes, which apparently represent the full array of trimming activities in yeast, are exclusively located in the lumen of microsomal vesicles derived from endoplasmic reticulum membranes.     

Full-length prepro-alpha-factor can be translocated across the mammalian microsomal membrane only if translation has not terminated

We have previously shown that fully synthesized prepro-alpha-factor (pp alpha F), the precursor for the yeast pheromone alpha-factor, can be translocated posttranslationally across yeast rough microsomal (RM) membranes from a soluble, ribosome-free pool. We show here that this is not the case for translocation of pp alpha F across mammalian RM. Rather we found that a small amount of translocation of full-length pp alpha F is observed, but is solely due to polypeptide chains that were still ribosome bound and covalently attached to tRNA, i.e., not terminated. In addition, both signal recognition particle (SRP) and SRP receptor are required, i.e., the same targeting machinery that is normally responsible for the coupling between protein synthesis and translocation. Thus, the molecular requirements for targeting are distinct from posttranslational translocation across yeast RM. As termination is generally regarded as part of translation, the translocation of full-length pp alpha F across mammalian RM does not occur "posttranslationally," albeit independent of elongation. Most other proteins for which posttranslational translocation across mammalian RM was previously claimed fall into the same category in that ribosome attachment as peptidyl-tRNA is required. To clearly separate these two distinct processes, we suggest that the term posttranslational be reserved for those processes that occur in the complete absence of the translational machinery. We propose the term "ribosome-coupled translocation" for the events described here.     

Dissociation of complexes between 70 kDa stress proteins and presecretory proteins is facilitated by a cytosolic factor.

Members of the 70 kDa stress protein family were shown previously to facilitate the posttranslational translocation of presecretory proteins into the endoplasmic reticulum and protein precursors into mitochondria. To identify proteins that interact with 70 kDa stress proteins during the early steps of posttranslational translocation, polyclonal antibodies were raised against purified yeast cytosolic stress proteins. They were used to immunoprecipitate complexes between 70 kDa stress proteins and a radiolabeled presecretory protein, prepro-alpha-factor, that was translated in vitro. Complexes between prepro-alpha-factor and 70 kDa stress proteins were stable, but could be dissociated in the presence of ATP and crude cytosolic extracts from yeast. These results are consistent with the idea that 70 kDa stress proteins act as molecular chaperones in translocation by binding to precursor proteins before or during their passage across membranes.     

Secretion in yeast: structural features influencing the post-translational translocation of prepro-alpha-factor in vitro.

In vitro, efficient translocation and glycosylation of the precursor of yeast alpha-factor can take place post-translationally. This property of prepro-alpha-factor appears to be unique as it could not be extended to other yeast protein precursors such as preinvertase or preprocarboxypeptidase Y. In order to determine if specific domains of prepro-alpha-factor were involved in post-translational translocation, we carried out a series of experiments in which major domains were either deleted or fused onto reporter proteins. Fusion of various domains of prepro-alpha-factor onto the reporter protein alpha-globin did not allow post-translational translocation to occur in the yeast in vitro system. Prepro-alpha-factor retained its ability to be post-translationally translocated when parts or all of the pro region were deleted. Removal of the C-terminal repeats containing mature alpha-factor had the most profound influence as post-translational translocation decreased in proportion to the number of repeats deleted. Taken together, these results suggest that efficient post-translational translocation requires a signal sequence and the four C-terminal repeats. There does not however, appear to be specific information contained within the C-terminus, as their presence in fusion did not enable the post-translational translocation of reporter proteins. Lastly, the ability to post-translationally translocate radiochemically pure prepro-alpha-factor that had been isolated by immuno-affinity chromatography required the addition of a yeast lysate fraction. Moreover, post-translational translocation is a function of the microsomal membrane of yeast microsomes and not of a factor peculiar to the yeast lysate, as reticulocyte lysate supported this as well.     

Expression of acid phosphatase-beta-galactosidase hybrid proteins prevents translocation by depleting a soluble factor

We have shown that hybrid proteins composed of the yeast repressible acid phosphatase (PHO5) and bacterial beta-galactosidase (lacZ) interfere with secretion of native acid phosphatase (Wolfe, P. B. (1988) J. Biol. Chem. 263, 6908-6915). We now report that PHO5-LacZ hybrid proteins have a more general effect on secretion and prevent translocation of several secreted proteins. Translocation of both the mating pheromone alpha-factor and the vacuolar protease carboxypeptidase Y is partially blocked when PHO5-LacZ hybrids are expressed. Cell fractionation and protease sensitivity indicate that alpha-factor and carboxypeptidase Y accumulate in precursor form on the cytoplasmic surface of the endoplasmic reticulum. Indirect immunofluorescence with antibody directed against beta-galactosidase supports the localization of hybrid proteins to the endoplasmic reticulum. Analysis of the hybrid protein phenotype in vivo and in vitro suggests that the hybrid proteins deplete a soluble factor required for efficient translocation across the endoplasmic reticulum. First, a decrease in the expression of a hybrid protein in vivo decreases its effect on translocation. Second, an in vitro translation/translocation reaction, prepared from a hybrid-bearing strain, is deficient in its ability to translocate prepro-alpha-factor across yeast microsomal membranes. This deficiency is complemented by addition of cytosol prepared from wild type cells. Finally, the hybrid protein phenotype is shown to be independent of the requirement for SSA gene products.     

Prepro-carboxypeptidase Y and a truncated form of pre-invertase, but not full-length pre-invertase, can be posttranslationally translocated across microsomal vesicle membranes from Saccharomyces cerevisiae

We have determined that prepro-carboxypeptidase Y and a truncated form of pre-invertase can be translocated across the yeast microsomal membrane post-translationally in a homologous in vitro system. The yeast secretory protein prepro-alpha-factor which was previously shown to be an efficient posttranslational translocation substrate is therefore not unique in this regard, but rather the yeast ER protein translocation machinery is generally capable of accepting substrates from a ribosome-free, soluble pool. However, within our detection limits, full-length pre-invertase could not be translocated posttranslationally, but was translocated co-translationally. This indicates that not every fully synthesized pre-protein can use this pathway, presumably because normal or aberrant folding characteristics can interfere with translocation competence.     

Single-amino-acid substitutions within the signal sequence of yeast prepro-alpha-factor affect membrane translocation.

We used a genetic approach to identify point mutations in the signal sequence of a secreted eucaryotic protein, yeast alpha-factor. Signal sequence mutants were obtained by selecting for cells that partially mistargeted into mitochondria a fusion protein consisting of the alpha-factor signal sequence fused to the mature portion of an imported mitochondrial protein (Cox IV). The mutations resulted in replacement of a residue in the hydrophobic core of the signal sequence with either a hydrophilic amino acid or a proline. After reassembly into an intact alpha-factor gene, the substitutions were found to decrease up to 50-fold the rate of translocation of prepro-alpha-factor across microsomal membranes in vitro. Two of three mutants tested produced lower steady-state levels of alpha-factor in intact yeast cells, although the magnitude of the effect was less than that in the cell-free system.     

Secretion in yeast: in vitro analysis of the sec53 mutant.

Of central importance to studying protein translocation via a combined genetic and biochemical approach is the in vitro analysis of yeast conditionally-lethal secretory mutants. Analysis of sec53 presented an opportunity not only to see if mutants could be examined in recently developed yeast in vitro translocation systems, but also to characterize further the nature of this mutant originally postulated to be defective in protein translocation. Membranes from sec53 were capable of translocating and glycosylating nascent prepro-alpha-factor in vitro in both sec53 and wild-type lysates at temperatures that were non-permissive for growth of the mutant cells. These results suggested that the Sec53 protein does not function directly in the translocation and glycosylation of prepro-alpha-factor. To examine this point further, we isolated membranes from sec53 cells that had been grown at the non-permissive temperature prior to disruption. In such cases, regardless of assay temperature, membranes from sec53 cells efficiently translocated but failed to glycosylate prepro-alpha-factor in vitro. The in vitro phenotype of sec53 could be mimicked by isolating rough microsomes from wild-type cells that had been grown for 1 h in the presence of tunicamycin. Together, these results demonstrate that sec53 is not defective in translocation, rather in assembly of the dolichol-oligosaccharide substrate needed for N-linked glycosylation.     

Export of prepro-alpha-factor from Escherichia coli

Yeast prepro-alpha-factor translocates posttranslationally into yeast microsomes in vitro. This process is strongly influenced by the extreme carboxyl-terminal region of the protein. These features contrast with the properties of most eucaryotic proteins which are translocated into the endoplasmic reticulum. We have extended these studies by introducing the gene for the wild-type and several mutant forms of prepro-alpha-factor into Escherichia coli. Prepro-alpha-factor is secreted into the periplasm and processed to pro-alpha-factor. Its translocation across the plasma membrane requires the membrane potential and the secY gene product. Deletion mutant analysis showed that features of the pro-segment were essential for secretion of prepro-alpha-factor in E. coli, while the carboxyl-terminal region, which is required in yeast, is dispensible in E. coli. Neither size nor the presence of a unique topogenic sequence was sufficient to explain the requirement for the pro-segment.     

Prepro-alpha-factor has a cleavable signal sequence

MAT alpha Saccharomyces cerevisiae secrete a small peptide mating pheromone termed alpha-factor. Its precursor, prepro-alpha-factor, is translocated into the endoplasmic reticulum and glycosylated at three sites. The glycosylated form is the major product in a yeast in vitro translation/translocation system. However, there is another translocated, nonglycosylated product that contains a previously unidentified modification. Contrary to previous results suggesting that the signal sequence of prepro-alpha-factor is not cleaved, amino-terminal radiosequencing has identified this product as prepro-alpha-factor without its signal sequence, that is, pro-alpha-factor. The translocated, glycosylated proteins are also processed by signal peptidase. Moreover, we have found that both purified eukaryotic and prokaryotic signal peptidase can process prepro-alpha-factor. Experiments using a yeast secretory mutant (sec 18) blocked in transport from the endoplasmic reticulum to the Golgi indicate that the protein is also cleaved in vivo. Finally, characterization of the Asn-linked oligosaccharide present on pro-alpha-factor in the yeast in vitro system by use of specific glucosidase and mannosidase inhibitors indicates that they have had the three terminal glucoses and probably one mannose removed. Therefore they most likely consist of Man8GlcNAc2 structures, identical to those found in the endoplasmic reticulum in vivo.     

The retinal rod Na(+)/Ca(2+),K(+) exchanger contains a noncleaved signal sequence required for translocation of the N terminus

The retinal rod Na(+)/Ca(2+),K(+) exchanger (RodX) is a polytopic membrane protein found in photoreceptor outer segments where it is the principal extruder of Ca(2+) ions during light adaptation. We have examined the role of the N-terminal 65 amino acids in targeting, translocation, and integration of the RodX using an in vitro translation/translocation system. cDNAs encoding human RodX and bovine RodX through the first transmembrane domain were correctly targeted and integrated into microsomal membranes; deletion of the N-terminal 65 amino acids (aa) resulted in a translation product that was not targeted or integrated. Deletion of the first 65 aa had no effect on membrane targeting of full-length RodX, but the N-terminal hydrophilic domain no longer translocated. Chimeric constructs encoding the first 65 aa of bovine RodX fused to globin were translocated across microsomal membranes, demonstrating that the sequence could function heterologously. Studies of fresh bovine retinal extracts demonstrated that the first 65 aa are present in the native protein. These data demonstrate that the first 65 aa of RodX constitute an uncleaved signal sequence required for the efficient membrane targeting and proper membrane integration of RodX.     

Absence of a cleavable signal sequence in Sindbis virus glycoprotein PE2.

Partial NH2-terminal sequence analysis has been performed on some products that result from the translation of 26 S mRNA of Sindbis virus either in vivo or in vitro. In vivo products were obtained after pulse-labeling of virus-infected cells. In vitro products were obtained after cell-free translation either in the absence or presence of microsomal membrane vesicles from dog pancreas. The sequence data indicate that the selective translocation across the microsomal membrane required for a distinct portion of one of the integral viral envelope proteins (PE2) is not accompanied by cleavage of its putative signal sequence. Furthermore, the NH2-terminal sequence of a proteolytic derivative (PE'2) that contains the bulk of PE2 and that is generated after exposure of the microsomal vesicles to proteolytic enzymes is identical to that of intact PE2, strongly suggesting that only a COOH-terminal portion of PE2 is excluded from translocation across the microsomal membrane.