Targeting of Aminopeptidase I to the Yeast Vacuole Is Mediated by Ssa1p, a Cytosolic Member of the 70-kDa Stress Protein Family*

The two cytosolic members of the highly conserved 70-kDa stress protein family, Ssa1p and Ssa2p, were specifically retained by the prepro-NH2 extension of the vacuolar aminopeptidase I precursor (pAPI) conjugated to agarose (Sulfolink). A temperature-sensitive mutant strain a1tsa234 (ssa1 ts ssa2 ssa3 ssa4), when incubated at the restrictive temperature, was able to assemble the API precursor into dodecamers, but failed to pack pAPI into vesicles and to convert it into mature API (mAPI), a process that occurs in the vacuole. Altogether these results indicate that Ssa1p mediates the targeting of pAPI to the vacuole.

The yeast vacuolar protein aminopeptidase I (API) 1 is synthesized as a cytosolic precursor and transported to the vacuole by a cytoplasm to vacuole targeting pathway (1,2). Genetic studies indicate that this pathway uses many of the molecular components of the degradative autophagy pathway (3,4). API transport is made specific and saturable by an unknown receptor that appears to recognize specific transport motifs in the prepro-NH 2 extension of the pAPI (5)(6)(7).
There is evidence that, soon after its synthesis, pAPI is assembled into homododecamers (8), which become associated with spherical particles (termed Cvt complex) and are delivered to the vacuole by a vesicle-mediated mechanism (9,10). Depending on the environmental conditions, the cell uses differ-ent membrane bound structures to sequester the Cvt complex and deliver pAPI to the vacuole. It has been proposed that, during vegetative growth, the Cvt complex is selectively wrapped by a double membrane sac (10) or translocated to a prevacuolar compartment (5), whereas in cells subjected to starvation conditions the Cvt complex is taken up by an autophagosome (10).
While searching for proteins interacting with the prepro-NH 2 extension of pAPI, the segment that mediates the transport of the protein precursor to the vacuole (5,11), we have observed its specific in vitro interaction with Ssa1p and Ssa2p, two highly homologous members of the cytosolic hsp70 family that are constitutively expressed during logarithmic growth and have been involved in protein folding and translocation (12)(13)(14). To study the role of Ssa1p in the targeting of pAPI to the vacuole, we analyzed this process in a mutant strain carrying a temperature-sensitive allele of SSA1 and inactivated alleles of the SSA2, SSA3, and SSA4 genes. We show here that targeting of the pAPI to the vacuole is mediated by Ssa1p under both growing and nitrogen starvation conditions.

EXPERIMENTAL PROCEDURES
Yeast Strains and Media-To generate the yeast strains used in this work we introduced mutations in each of the four SSA genes in the W303-1b genetic background. Gene disruptions of SSA2, SSA3, and SSA4 were performed by the short flanking homology technique (15). ssa2::LEU2 and ssa3::TRP1 disruption cassettes were amplified by PCR using genomic DNA of strain a1-45 (12) as template. SSA4 was replaced by the dominant-resistant module KanMX, which confers resistance to Geneticin (16). The PCR fragments contained in their 3Ј and 5Ј ends a minimum of 45 base pairs homologous to the flanking regions of the target gene, thus allowing for homologous recombination at their genomic loci. Disruption cassettes were consecutively transformed into W303-1b cells (17), and disruption of each gene was verified by analytical PCR using specific oligonucleotides for the 5Ј and 3Ј regions of the corresponding gene. The resultant strain, deleted for SSA2, SSA3, and SSA4, was named ESY170 (see Table I). To obtain strain ESY216 (ssa1 ts ssa2ssa3ssa4) a 1.7-kilobase pair KpnI-SphI fragment of the ssa1 temperature-sensitive allele in strain a1-45, containing mutation Pro 417 to Leu (12), was amplified and cloned in the integrative vector pRS306 (URA3) (18), digested at the unique ClaI site internal to the cloned fragment, and transformed into strain ESY170. Transformants being both Ura ϩ -and temperature-sensitive for growth were counterselected in fluorotic acid-containing plates and analyzed again for temperature sensitivity. Correct integration and presence of the Pro 417 3 Leu mutation were confirmed by PCR and sequencing. Strain ESY228 was constructed by disrupting the API gene using the HIS3 marker as described previously (5). ESY228 was used to overexpress the API gene carried in a 2-plasmid under the control of the GAL1-10 promoter (5). The genotypes of all the strains used in these studies are listed in Table  I. The pGAL-API has been described previously (5,11).
Yeast cells were grown in synthetic minimal medium: 0.67% yeast nitrogen base, 2% glucose (SD), 2% raffinose (SRaf), or 2% galactose * This work was supported by a grant from the European Community (EC) (FMRX-CT96-0058) (to I. V. S. and M. J. M.) and by grants from the Comisión Interministerial de Ciencia y Tecnología (94-0035) (to I. V. S.) and the Fondo de Investigaciones Sanitarias (98/279) (to M. J. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
(SGal) and the appropriate auxotrophic requirements. Synthetic minimal medium containing 2% glucose and 0.17% yeast nitrogen base without amino acids and ammonium sulfate (SD-N) was used for nitrogen starvation experiments.
Standard techniques for yeast strain propagation and genetic manipulation were used as described (19). Yeast transformation was performed by the method of Ito et al. (20) and DNA manipulations as described (21).
Purification of Ssa1p/Ssa2p by Affinity Chromatography-60 ml of SD medium containing the appropriate requirements were inoculated with W303-1b cells and grown overnight at 30°C. Cells were harvested in the logarithmic phase by centrifugation, resuspended in 2 ml of yeast nitrogen base without amino acids and ammonium sulfate prepared to 0.17%, 0.4% NH 4 Cl, 0.3% glucose, incubated for 30 min at 30°C and labeled for 1 h at 30°C with 1 mCi of [ 35 S]methionine/cysteine (specific activity Ͼ 1000 Ci/ml; Promix, Amersham Pharmacia Biotech). The cells were then washed with cold phosphate-buffered saline, 0.15 M NaCl (PNC buffer), disrupted with glass beads in 1 ml of the same buffer, phenylmethylsulfonyl fluoride added to 1 mM, and cell debris removed by centrifugation two times for 10 min at 2000 ϫ g. The resulting extract was centrifuged for 30 min at 150,000 ϫ g in a TL100.3 rotor (Beckman) and the supernatant (S150) made 2% in Triton X-100, whereas the pellet (P150) was resuspended in 1 ml of PNC buffer, 2% Triton X-100 incubated for 30 min at 4°C and then centrifuged for 30 min at 150,000 ϫ g.
The 43-amino acid-long MEEQREILEQLKKTLQMLTVEPSKNN-QIANEEKEKKENENSWC-COOH and MEEQREGLEQLKKTLQM-LTVEPSKNNQIANEEKEKKENENSWC-COOH peptides from the prepro-NH 2 extension of the pAPI (prepro-peptide) were coupled to agarose (Sulfolink) (Pierce) at a ratio of 5 mg/ml gel, according to the manufacturer's instructions (Pierce). The S150 and P150 fractions were mixed with 1 volume of PNC buffer and incubated overnight with the prepro-peptide conjugated to agarose (Sulfolink). The resin was then washed three times for 15 min with PNC buffer, 0.5% Triton X-100, once for 15 min with phosphate-buffered saline, 1 M NaCl, 0.5% Triton X-100, and then quickly with water. Aliquots from the yeast crude extract, the S150 and P150 fractions, the fourth wash, and the washed resin were boiled in Laemmli buffer, resolved by SDS-12% PAGE, and the electrophoretograms analyzed for protein both by Coomassie Blue staining and autoradiography.
The 70-kDa polypeptide retained by the affinity resin was in-geldigested with trypsin and the generated peptides separated by RP-HPLC, concentrated on Poros R2-beads, and transferred to the MALDI target as described previously (22). For protein identification, peptide ions were subjected to post-source decay (PSD) analysis and the Sequest algorithm (23,24) used to search in a nonredundant protein data base containing over 290,000 entries with the experimental fragmentation pattern.
Studies on the Processing of the pAPI into mAPI-A1a234 and a1 ts a234 cells grown overnight at 24°C in SD medium to 1 OD 660, and, when required, starved for 2 h in SD-N medium, were concentrated to 25 OD 660 and incubated in fresh medium at 24°C for 30 min. The cell suspension was divided in two, then incubated at 24 or 37°C for 2 min, and the cells labeled for 10 min with 50 Ci of [ 35 S]methionine/cysteine (Promix, Amersham Pharmacia Biotech) per OD 660 unit . The labeled protein was chased by resuspension of the cells in SD or SD-N medium containing 4 mM cysteine and 8 mM methionine. Chasing was terminated by addition of cold trichloroacetic acid to a final concentration of 5%. The protein precipitates were left on ice for 15 min, collected by centrifugation at low speed, washed with cold acetone, dried , and upon their freezing in liquid nitrogen, resuspended in 50 mM Tris-HCl pH 7.5, 5 mM EDTA (buffer S) containing 1% SDS by vigorous vortexing using glass beads. Samples were then boiled for 4 min, diluted 5-fold in 2.5 volumes with buffer S, and then made 0.25% Tween 20 and 0.1 M NaCl. Equal amounts of protein were incubated for 2 h at 4°C with rabbit anti-API serum (5) and then for 1 h with protein G-Sepharose. The immunoprecipitates were washed five times with buffer S, 0.15 M NaCl, boiled in Laemmli buffer, resolved by SDS-10% PAGE, and the fixed gels treated with 1 M sodium salicylate, dried, and autoradiographed.
Protease Protection Experiments-a1 ts a234 cells were grown at 24°C to 1 OD 660 in SD medium, spheroplasted, and metabolically labeled and chased at 24 and 37°C. Spheroplasts were adjusted to 15 OD 660 /ml with SL medium (1 M sorbitol, 1% glucose, 1% proline, 0.17% yeast nitrogen base without amino acids and ammonium sulfate, with the appropriate auxotrophic requirements) (2), and 1 ml was preincubated for 20 min and metabolically labeled for 10 min with 100 Ci of [ 35 S]methionine/ cysteine per OD 660 at the temperature of choice. Labeled spheroplasts were then diluted 10-fold in SL medium containing 8 mM methionine, 4 mM cysteine, and 0.2% yeast extract and chased for 2 min or 4 h. Chases were stopped by incubation for 2 min with 10 mM NaN 3 and lysates from spheroplasts prepared by osmotic lysis in 200 mM sorbitol, 5 mM MgCl 2, 20 mM Pipes, pH 6.8. Unlysed spheroplasts were removed by centrifugation for 2 min at 500 ϫ g. Lysates were treated at 4°C for 30 min with 50 g/ml proteinase K (Life Technologies, Inc.) in the absence and presence of 0.2% Triton X-100. Digestion by the protease was stopped with a mixture of protease inhibitors (5 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and pepstatin, aprotinin, and leupeptin at 1 g/ml each) and the protein precipitated with trichloroacetic acid, using 100 g/ml bovine serum albumin as carrier, and API immunoprecipitated as described above and analyzed by SDS-10% PAGE and autoradiography.
Analysis of API Dodecamers by Sedimentation Velocity Centrifugation-a1 ts a234⌬ape1 cells carrying the 2-pGAL-API were grown in SRaf medium overnight to 1 OD 660 and then shifted to SGal medium for 4 h at 24 or 37°C. The cells were incubated the last 15 min in SGal medium with or without 20 mM NaN 3 , harvested by low speed at the corresponding temperatures, and immediately frozen in liquid nitrogen. Crude cell extracts prepared with glass beads in cold 50 mM Tris-HCl, pH 8, 5 mM EDTA, and protease inhibitors were fractionated by centrifugation on glycerol (20 -50%) (8) and the gradient fractions scrutinized for API by Western blot using the rabbit polyclonal anti-API antibody.

RESULTS
The Heat Shock Proteins Ssa1p and Ssa2p Are Specifically Retained by Prepro(API)-Peptidyl-Agarose Affinity Resin-Incubation of [ 35 S]methionine/cysteine-labeled S150 and P150 fractions with prepro(API)-peptidyl-agarose (Sulfolink) resulted in specific retention of a 70-kDa polypeptide as shown both by Coomassie Blue staining and autoradiography (Fig.  1A). The mass spectrometric screening of HPLC fractions from a trypsin digest of the 70-kDa polypeptide showed that 16 out of 18 collected fractions contained peptide material. The peptide ion with a m/z value of 1424.63 in fraction 7 (Fig. 1B, left) was subjected to PSD analysis (Fig. 1B, right). After manually denoting all the ␥ ions displayed in the PSD spectrum (23), a sequence tag (303.14) SEQ/KEDE (1020.06) was obtained. Ad- This study ditional information regarding the presence of phenylalanine and serine residues in the peptide was deduced from the detection of specific ammonium ions of both amino acids. The search of the peptide in a nonredundant protein data base using the Sequest algorithm (24) led to the NH 2 -FKEEDEKESQR-COOH sequence present in the Ssa1p and Ssa2p proteins of yeast (Fig. 1C). This finding was confirmed by PSD analysis of two other peptide ions present in both the Ssa1p and Ssa2p proteins (Fig. 1C). Moreover, 15 peptide masses shared by both homologue proteins could be identified in the different HPLC fractions. One peptide mass was found to be unique for Ssa1p and three for Ssa2p. The presence of both proteins Ssa1p and Ssa2p in the polypeptide band was confirmed by PSD analysis of two of these unique peptides (Fig. 1C).
To further investigate if the interaction of the Ssa1p/Ssa2p pair with the prepro-NH 2 extension of the pAPI required a structural motif contained in the primary or secondary structure of the peptide, the Ile residue in position 7 was substituted by a Gly. This substitution has been shown to disrupt the ␣Ϫhelix conformation of the peptide into a random coil and to inhibit the transport of the pAPI to the vacuole (7). We observed that the introduction of Gly in position 7 did not suppress the retention of the Ssa1p and Ssa2p by prepro(API)-peptidyl affinity columns (data not shown), thus indicating that the two chaperones recognized a motif contained in the primary structure of the peptide (see "Discussion").
Targeting of the pAPI to the Vacuole by the Cvt Pathway Is Mediated by Ssa1p-To examine whether Ssa1p was actually involved in the targeting of the pAPI to the vacuole, we studied the processing of the pAPI by a1 ts a234 cells. These cells were developed by introducing the disrupted SSA2, SSA3, and SSA4 genes as well as a ssa1 ts allele from a1-45 cells (12) in the genetic background of W303-1b. After their shift to the nonpermissive temperature, the a1-45 cells have been shown to stop growing immediately as well as to undergo defects in the translocation and processing of several ER and mitochondrial protein precursors (12).
Processing of the pAPI into the mature form was first studied in a1 ts a234 cells under vegetative conditions (Fig. 2). The cells were pulse-labeled for 10 min with [ 35 S]methionine/cysteine at either 24 or 37°C and then chased for periods between 0 and 120 min. Processing was analyzed by autoradiography of the precursor and mAPI species immunoprecipitated and separated by SDS-PAGE. It was observed that whereas the precursor was converted into the mature form at 24°C, the permissive temperature for Ssa1p activity, the conversion was completely inhibited at the restrictive temperature. These results strongly suggest that cytosolic Ssa1p is involved in the transport of the pAPI to the vacuole where its conversion into the mature form occurs. Furthermore, the absence of mAPI species after the 120-min chase at the restrictive temperature (Fig. 2B, lane 3) strongly suggests that in the absence of Ssa1p the mechanism of API transport is blocked rather than slowed down.
To study whether Ssa1p mediates the transport of pAPI to the vacuole in starved cells, a1 ts a234 cells grown in SD medium at 24°C were subjected to nitrogen starvation for 2.5 h. Transport of the pAPI to the vacuole was studied in a pulse-chase experiment as described above. As observed in vegetative cells the starved cells were able to process the pAPI into the mature form at 24°C but not at 37°C (Fig. 2). This result indicates that Ssa1p is also essential for the transport of the pAPI to the vacuole in nitrogen-starved cells.
Previous studies have shown that cells carrying the ssa1 ts allele recovered their ability to grow upon shifting the incubation temperature from 37 to 24°C (12). We, therefore, investigated whether the recovery of the Ssa1p activity at the permissive temperature rescued the processing of newly synthesized pAPI molecules. For this purpose, first, a1 ts a234 cells preincubated at 37°C for 30 min were labeled at 24°C for 10 min with [ 35 S]methionine/cysteine, and after their chase at the same temperature for 0, 30, 60, and 120 min the conversion of pAPI into mAPI analyzed by immunoprecipitation and autoradiography. The study revealed a progressive conversion of pAPI into mAPI in the 60 -120-min period (Fig. 3A), thus indicating that the inhibition of the pAPI transport to the vacuole was reversed. Furthermore, when the experiment was repeated by performing the labeling at 37°C, the result was comparable (Fig. 3B). This result indicated that the pAPI synthesized at 37°C was rescued at the permissive temperature and transported and processed in the vacuole (see "Discussion").
Assembly of the pAPI into Dodecamers Does Not Require Ssa1p-The observation that newly synthesized pAPI molecules are assembled into homododecamers in the cytoplasm and transported in this form into the vacuole (8) prompted us to study if Ssa1p was required for their assembly. For this purpose, a1 ts a234⌬ape1 cells carrying the pGAL-API plasmid were induced with galactose for 4 h at both 24 and 37°C and crude extracts fractionated by sedimentation velocity centrifugation in glycerol gradients as described (8). Western analysis of the gradient fractions using an anti-API specific antibody showed that at both temperatures pAPI migrated with the mobility expected for dodecamers (Fig. 4), thus excluding a role for Ssa1p and the other family members in their assembly.
pAPI Remains Unprotected in the Cytoplasm of a1 ts a234 Cells at the Restrictive Temperature-To further define the role of Ssa1p/Ssa2p in the targeting of pAPI to the vacuole, we investigated whether pAPI molecules were retained or not in the cytoplasm at the permissive and restrictive temperature. This is important since Ssa1p/Ssa2p are localized in the cytoplasm and it has been reported in a recent study that all the pAPI in the cell is within vesicles in a ⌬ssa1 mutant (25). For this purpose, the sensitivity of newly synthesized pAPI molecules to proteinase K was studied in a1 ts a234 spheroplasts pulsed for 10 min with [ 35 S]methionine/cysteine at 24 and 37°C and then chased 2 min or 4 h at the corresponding temperature. The long chase period was adjusted to the slow processing of pAPI in spheroplasts at 24°C, which amounted to 60 -70% of the newly synthesized pAPI during a 4-h chase (not shown). The protection of labeled pAPI from proteinase K digestion in the absence or presence of detergent was analyzed in osmotically lysed spheroplasts by autoradiography after immunoprecipitation (see "Experimental Procedures" and the legend to Fig. 5). When lysates from a1 ts a234 spheroplasts were incubated at the permissive temperature, it was observed that between 2 min and 4 h after its synthesis nearly 85% of pAPI was converted into iAPI (i.e. intermediate API) and mAPI. Moreover, in the absence of detergent, the majority of pAPI and iAPI was digested by proteinase K into the comparatively protease-resistant mAPI, whereas in its presence digestion was complete (Fig. 5A). It should be noted that no significant vacuolar disruption, that would have interfered with the protease protection assay, occurred during lysis of spheroplasts, as shown by the absence of mAPI in the immunoprecipiates obtained from spheroplasts chased for 2 min at the permissive temperature. When the experiment was performed with lysates from a1 ts a234 spheroplasts incubated at the restrictive temperature, we observed that 4 h after its synthesis the majority of pAPI remain unprocessed, the small amount of mAPI detected probably being generated by minor vacuolar breakage during the disruption of the spheroplasts by osmotic lysis. Moreover, all the pAPI was digested by the protease in the absence of detergent (Fig. 5A).
Analysis of the protease protection of CPY in a parallel experiment run under the conditions described for pAPI showed that, as expected, after 2 min of chase the CPY precursor (pCPY) transported through the secretory pathway was FIG. 2. Processing of pAPI into mAPI is inhibited in growing and nitrogen starved a1 ts a234 cells at the restrictive temperature. A1a234 and a1 ts a234 cells grown in SD medium (i.e. vegetative conditions) and incubated further for 2.5 h in SD-N medium (i.e. starvation conditions) (see "Experimental Procedures") were incubated at 24 or 37°C for 2 min and then pulse-labeled for 10 min with [ 35 S]methionine/cysteine and chased for the indicated times at the corresponding temperature. Processing of cell extracts and analysis of the API species were performed by immunoprecipitation and autoradiography as described under "Experimental Procedures." FIG. 3. Shift from 37 to 24°C restores processing of pAPI in a1 ts a234 cells. a1 ts a234 cells grown in SD medium at 24°C were incubated at 37°C for 20 min and then labeled with [ 35 S]methionine/ cysteine for 10 min at the same temperature (B) or were preincubated first at 37°C for 30 min and then at 24°C for 3 min before their labeling at 24°C for 10 min (A). In both experiments labeled pAPI was chased for the indicated times at the permissive temperature. Analysis of the API species was performed as described in the legend to Fig. 2.   FIG. 4. Ssa1p is not involved in the assembly of pAPI into dodecamers. Extracts from a1 ts a234⌬ape1 cells expressing API under the control of the GAL1-10 promoter and incubated for 4 h with galactose were fractionated by sedimentation velocity centrifugation using 20 -50% glycerol gradients and the distribution of the API species through the gradients studied by Western blot using an anti-API polyclonal antibody. Molecular mass standards were ovalbumin (Ov, 45 kDa), catalase (Ct, 240 kDa), apoferritin (Af, 450 kDa), and thyroglobulin (Tg, 669 kDa).
protected from proteinase K in the absence of detergent, whereas it was digested into the protease-resistant mature form (mCPY) when detergent was added. Furthermore, after 4 h of chase, CPY was observed to migrate with the size corresponding to mCPY and to be resistant to the protease both in the absence or presence of detergent (Fig. 5B). These results were consistent with the entrance of pCPY in the secretory pathway concomitant with its synthesis and reassessed the validity of the protease protection assay as a valid technique to study the retention of pAPI in the cytoplasm.
Altogether these results indicated that a significant amount of the pAPI synthesized and chased at the permissive temperature remained unprotected in the cytoplasm long after its synthesis and that at the restrictive temperature in the absence of a functional Ssa1p protein, all the precursor retained in the cytoplasm was unprotected.

DISCUSSION
The results of studies on the transport of API to the vacuole are compatible with two models of transport that are not mutually exclusive. In the first model, API assembled into dodecamers is selectively wrapped by a double membrane sac that eventually fuses with the vacuole to unload the protein packed in the inner vesicles into the vacuolar lumen (9). The second model proposes that the precursor is translocated through the membrane of a transport intermediate or the vacuolar membrane (5).
We show here that Ssa1p and Ssa2p, the two constitutive chaperones involved in the folding, oligomerization, and targeting of proteins from the cytosol to such diverse organelles as ER (14,26,27), mitochondria, (28) and nucleus (29), interact specifically in vitro with the peptide that mimics the prepro-NH 2 extension of pAPI, an extension that is necessary and sufficient for the targeting of pAPI to the vacuole (11).
The predominance of the 70-kDa peptides among the few proteins retained by the prepro(API)-peptidyl affinity column stresses the specificity of the interaction between the Ssa1p/ Ssa2p and the prepro-NH 2 extension of pAPI. Nevertheless, when considering that chaperones often bind to exposed hydrophobic patches on incompletely folded proteins and that most of the peptide mimicking the pre-part and the pre-(Gly 7 ) peptide are unfolded, the possibility that their in vitro interaction could be unspecific should be considered. With regard to this, it is important to note that the mean hydrophobicity per residue of nonpolar face in the pre-part (MEEQREGLEQLLKKTLQ) is only 0.28 (7) and that the pro-part (MLTVEPSKNN-QIANEEKEKKENENSW) is even less hydrophobic. These observations make it unlikely that the interaction of the prepro-NH 2 extension with Ssa1p/Ssa2p is an artifact provoked by the physicochemical properties of the peptide.
We have shown that, in aqueous solution, only 24% of the peptide mimicking the pre-part and 1.7-2.4% of the pre-(Gly 7 ) peptide fold into an ␣Ϫhelix (7). However, Ssa1p/Ssa2p are retained with similar efficiency by the prepro-and pre-(Gly 7 ) pro-affinity columns, suggesting that either the motifs recognized by the chaperones are contained in the primary structure of the pre-part or in the pro-part. In addition, the recovery of the pAPI extracted from a1 ts a234 cells grown at 24 and 37°C as a dodecamer upon centrifugation in glycerol gradients indicates that Ssa1p is not involved in its oligomerization.
The inhibition of the conversion of pAPI into mAPI in a1 ts a234 cells incubated at the restrictive temperature under either vegetative or nitrogen starvation conditions indicates that Ssa1p mediates the transport of pAPI to the vacuole under both conditions. This result is in contrast with recent observations by Satyanarayana et al. (25) who have reported that Ssa1p is not involved in the transport of pAPI to the vacuole under nitrogen starvation conditions favoring autophagocytosis.
The possibility that Ssa1p may play a role in engaging the properly folded precursor with the capture/transport machineries that operate in its transport to the vacuole should be considered. Wrapping of pAPI complexes by double membrane sacs might only occur after their interaction with the cytoplasmic chaperone, whereas if pAPI is translocated through the membranes of transport vesicles/vacuole the chaperone could interact with a specific membrane receptor. The demonstration that a significant amount of newly synthesized pAPI remains unprotected in the cytoplasm 4 h after its synthesis, at the permissive temperature, as shown by the protease protection assay, agrees with the results of previous studies (2) and seems to discard the existence of membrane barriers between pAPI and Ssa1p that could prevent their interaction, as indicated in a recent study (25). Furthermore, the complete digestion of the pAPI synthesized and chased for 4 h at the restrictive temperature, by proteinase K in the absence of detergent, indicates that the inactivation of Ssa1p inhibits the pAPI protection. This observation strongly suggests that the chaperone mediates the capture/downloading of pAPI into Cvt vesicles. FIG. 5. pAPI remains unprotected in the cytoplasm of a1 ts a234 cells at the restrictive temperature. a1 ts a234 spheroplasts were metabolically labeled at 24 and 37°C for 10 min with [ 35 S]methionine/ cysteine and then chased for either 2 min or 4 h at the corresponding temperature. Osmotically produced lysates were then incubated at 4°C for 30 min without or with 50 g/ml proteinase K, in the absence or presence of 0.2% Triton X-100. Note the absence of mAPI after the 2-min chase, thus, indicating that no significant vacuolar disruption was produced during the spheroplasts lysis. API (A) and CPY (B) proteins were immunoprecipitated using specific antibodies, resolved by 10% SDS-PAGE, and analyzed by autoradiography. pCPY, precursor CPY; mCPY, mature CPY; iAPI, intermediate API.