Portal de Periódicos da CAPES

Mitochondrial Hsp70 Ssc1: Role in Protein Folding

2001; Elsevier BV; Volume: 276; Issue: 9 Linguagem: Inglês

10.1074/jbc.m009519200

1083-351X

Qinglian Liu, Joanna Krzewska, Krzysztof Liberek, Elizabeth A. Craig,

thermodynamics and calorimetric analyses

Ssc1, the major Hsp70 of the mitochondrial matrix, is involved in the translocation of proteins from the cytosol into the matrix and their subsequent folding. To better understand the physiological mechanism of action of this Hsp70, we have undertaken a biochemical analysis of Ssc1 and two mutant proteins, Ssc1–2 and Ssc1–201. ssc1–2 is a temperature-sensitive mutant defective in both translocation and folding; ssc1–201contains a second mutation in this ssc1 gene that suppresses the temperature-sensitive growth defect ofssc1–2, correcting the translocation but not the folding defect. We found that although Ssc1 was competent to facilitate the refolding of denatured luciferase in vitro, both Ssc1–2 and Ssc1–201 showed significant defects, consistent with the data obtained with isolated mitochondria. Purified Ssc1–2 had a lowered affinity for a peptide substrate compared with wild-type Ssc1 but only in the ADP-bound state. This peptide binding defect was reversed in the suppressor protein Ssc1–201. However, a defect in the ability of Hsp40 to stimulate the ATPase activity of Ssc1–2 was not corrected in Ssc1–201. Thus, the inability of these two mutant proteins to efficiently facilitate luciferase refolding correlates with their defect in stimulation of ATPase activity by Hsp40s, indicating that this interaction is critical for protein folding in mitochondria. Ssc1, the major Hsp70 of the mitochondrial matrix, is involved in the translocation of proteins from the cytosol into the matrix and their subsequent folding. To better understand the physiological mechanism of action of this Hsp70, we have undertaken a biochemical analysis of Ssc1 and two mutant proteins, Ssc1–2 and Ssc1–201. ssc1–2 is a temperature-sensitive mutant defective in both translocation and folding; ssc1–201contains a second mutation in this ssc1 gene that suppresses the temperature-sensitive growth defect ofssc1–2, correcting the translocation but not the folding defect. We found that although Ssc1 was competent to facilitate the refolding of denatured luciferase in vitro, both Ssc1–2 and Ssc1–201 showed significant defects, consistent with the data obtained with isolated mitochondria. Purified Ssc1–2 had a lowered affinity for a peptide substrate compared with wild-type Ssc1 but only in the ADP-bound state. This peptide binding defect was reversed in the suppressor protein Ssc1–201. However, a defect in the ability of Hsp40 to stimulate the ATPase activity of Ssc1–2 was not corrected in Ssc1–201. Thus, the inability of these two mutant proteins to efficiently facilitate luciferase refolding correlates with their defect in stimulation of ATPase activity by Hsp40s, indicating that this interaction is critical for protein folding in mitochondria. dihydrofolate reductase wild type Molecular chaperones such as members of the 70-kDa class (Hsp70s) bind to nonnative conformations of proteins, thus facilitating their folding and translocation across membranes (1Craig E.A. Science. 1993; 260: 1902-1903Crossref PubMed Scopus (162) Google Scholar, 2Hartl F.U. Nature. 1996; 381: 571-580Crossref PubMed Scopus (3137) Google Scholar). The C-terminal 28-kDa region of Hsp70s binds unfolded polypeptides, whereas the highly conserved N-terminal 44-kDa domain regulates that binding through its interaction with adenine nucleotides (3Bukau B. Horwich A.L. Cell. 1998; 92: 351-366Abstract Full Text Full Text PDF PubMed Scopus (2443) Google Scholar). It is thought that Hsp70 proteins, like many GTPases, have a two-state conformation. When an ADP molecule is bound in the nucleotide-binding site, the Hsp70 exhibits relatively stable polypeptide substrate binding; when ATP is bound, binding of substrate is relatively unstable. The 44-kDa domain has a low intrinsic ATPase activity; therefore, ATP hydrolysis converts Hsp70 to a form having a relatively stable interaction with unfolded proteins. Exchange of ADP for ATP results in destabilization of the interaction. It is thought that a polypeptide first interacts with a Hsp70 in the ATP-bound state, and then hydrolysis of ATP to ADP stabilizes this interaction. This cycle of interaction is facilitated by cochaperones. Procaryotes and mitochondria contain Hsp40-type cochaperones as well as nucleotide release factors such as GrpE of Escherichia coli and Mge1 ofSaccharomyces cerevisiae (4Packschies L. Theyssen H. Buchberger A. Bukau B. Goody R.S. Reinstein J. Biochemistry. 1997; 36: 3417-3422Crossref PubMed Scopus (156) Google Scholar, 5Sakuragi S. Liu Q. Craig E.A. J. Biol. Chem. 1999; 274: 11275-11282Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar). In the simplest scenario, nucleotide release factors are thought to destabilize the interaction of unfolded proteins with Hsp70, as release of ADP from a DnaK·ADP complex can be increased up to 5000-fold by GrpE action (4Packschies L. Theyssen H. Buchberger A. Bukau B. Goody R.S. Reinstein J. Biochemistry. 1997; 36: 3417-3422Crossref PubMed Scopus (156) Google Scholar). However, the action of Hsp40s such as DnaJ of E. coli is less well understood. Hsp40s stimulate the ATPase activity of Hsp70s, which is thought to facilitate their binding to unfolded polypeptide substrates (6Kelley W.L. Curr. Biol. 1999; 9: R305-R308Abstract Full Text Full Text PDF PubMed Scopus (102) Google Scholar). Hsp40s contain a canonical J domain that interacts with the ATPase domain of Hsp70s (7Greene M.K. Maskos K. Landry S.J. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 6108-6113Crossref PubMed Scopus (249) Google Scholar). There may also be a site of interaction of Hsp40s with the C-terminal domain of Hsp70s, since mutant Hsp70s that show a defect in interaction with peptide substrates also show a defect in interaction with Hsp40s (8Suh W.-C. Burkholder W. Lu C.Z. Zhao X. Gottesman M. Gross C. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15223-15228Crossref PubMed Scopus (230) Google Scholar, 9Davis J.E. Voisine C. Craig E.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9269-9276Crossref PubMed Scopus (67) Google Scholar). In addition to interacting with Hsp70s, many Hsp40s, including E. coli DnaJ, bind unfolded or partially folded polypeptides, preventing their aggregation (10Langer T. Lu C. Echols H. Flanagan J. Hayer M.K. Hartl F.U. Nature. 1992; 356: 683-689Crossref PubMed Scopus (795) Google Scholar). Based on a variety of in vitro analyses, a model of how Hsp40s and Hsp70s cooperate in protein folding has evolved (2Hartl F.U. Nature. 1996; 381: 571-580Crossref PubMed Scopus (3137) Google Scholar, 3Bukau B. Horwich A.L. Cell. 1998; 92: 351-366Abstract Full Text Full Text PDF PubMed Scopus (2443) Google Scholar). Polypeptide substrates first bind Hsp40. Then, the ATPase domain of Hsp70 interacts with Hsp40 via its J domain. This interaction not only brings the substrate in close proximity to Hsp70, but it also stimulates hydrolysis of ATP, trapping the substrate. Subsequent release of nucleotide and rebinding of ATP destabilize the Hsp70-substrate complex, resulting in its release. In eucaryotes, all major cellular compartments contain at least one Hsp70 and one Hsp40. Ssc1, the major Hsp70 of the mitochondrial matrix, is involved in the translocation of proteins from the cytosol across the mitochondrial inner membrane to the matrix and their subsequent folding. Two temperature-sensitive mutants, ssc1–2 andssc1–3, have been used extensively in the analysis of the physiological roles of Ssc1 (11Kang P.J. Ostermann J. Shilling J. Neupert W. Craig E.A. Pfanner N. Nature. 1990; 348: 137-143Crossref PubMed Scopus (547) Google Scholar, 12Gambill B.D. Voos W. Kang P.J. Miao B. Langer T. Craig E.A. Pfanner N. J. Cell Biol. 1993; 123: 109-117Crossref PubMed Scopus (222) Google Scholar, 13Schneider H.C. Westermann B. Neupert W. Brunner M. EMBO J. 1996; 15: 5796-5803Crossref PubMed Scopus (100) Google Scholar, 14Voos W. von Ahsen O. Muller H. Guiard B. Rassow J. Pfanner N. EMBO J. 1996; 15: 2668-2677Crossref PubMed Scopus (100) Google Scholar, 15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). The ssc1–3 mutation causes an amino acid substitution in the ATPase domain that results in severe defects in translocation. The ssc1–2 mutation, which changes a single amino acid in the peptide binding domain (P442S) (12Gambill B.D. Voos W. Kang P.J. Miao B. Langer T. Craig E.A. Pfanner N. J. Cell Biol. 1993; 123: 109-117Crossref PubMed Scopus (222) Google Scholar), has less severe effects and is therefore more useful in dissecting the roles of Ssc1 in both translocation and folding. Intragenic suppressors of ssc1–2, which cause an additional single amino acid alteration in the peptide binding domain of Ssc1–2, have been isolated. These suppressor mutations allow robust growth at the intermediate temperature of 34 °C but do not completely suppress the growth defect of ssc1–2 above that temperature. Two of these, Ssc1–201 and Ssc1–202 (D519E and V524I, respectively) have been analyzed (15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). Both suppress the defects in protein translocation as well as the interaction with the membrane tether Tim44. However, initial results suggested that the folding defect is not suppressed (15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). To better understand the role of Ssc1 in protein folding within the mitochondrial matrix, we have undertaken a characterization of the biochemical properties of Ssc1, Ssc1–2, and Ssc1–201. We found that neither the ATPase activity of Ssc1–2 nor of the suppressor protein Ssc1–201 was stimulated by Hsp40s. In addition, both show significant defects in their ability to facilitate the refolding of luciferasein vitro. These results make evident the importance of Hsp40s in facilitating protein folding in the mitochondrial matrixin vivo and provide insight into the complex interactions between Hsp40s and Hsp70s. pRS314-SSC1(His tag) was constructed from pRS314-SSC1 (16Miao B. Davis J.E. Craig E.A. J. Mol. Biol. 1997; 265: 541-552Crossref PubMed Scopus (82) Google Scholar) by inserting six codons encoding histidine at the 3′ end of theSSC1 gene using polymerase chain reaction. The segments ofssc1–2, ssc1–201, and ssc1–202 encoding mutations were subcloned into this plasmid. An expression plasmid (pGEM-Su9-DHFR*)1 that contained a mutant version of DHFR (C7S, S42C, and D49C), DHFR*, for use in import and folding assays was constructed. The Su9 and DHFR* portions were polymerase chain reaction-amplified from pGEM-Su9-DHFR (15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar) and DHFR* (17Hajek P. Koh J.Y. Jones L. Bedwell D.M. Mol. Cell. Biol. 1997; 17: 7169-7177Crossref PubMed Google Scholar), respectively, using Pfu DNA polymerase (Stratagene, La Jolla, CA). A PstI/BstXI fragment of the final polymerase chain reaction product was inserted into pGEM-Su9-DHFR to replace the wild-type (WT) portion of DHFR. The plasmid was confirmed by DNA sequencing and restriction enzyme digests. The yeast strain used for Ssc1 purification was obtained by crossingWY11 (his3–11, 15 leu2–3, 112 ura3–52 trp1-Δ1 pep4::HIS3) withPJ53–52C (trp1–1 ura3–1 leu2–3, 112 his3–11, 15 ade2–1 can1–100 GAL2+ met2-Δ1 lys2-Δ2 ssc1ΔClaI::LEU2 pRS316-SSC1) followed by sporulation and dissection to generate a strain of the genotypepep4::HIS3 trp1–1 ssc1ΔClaI::LEU2 pRS316-SSC1 (QL1). After transforming QL1 with pRS314-SSC1(His tag) containing mutant or WT versions ofSSC1, cells lacking pRS316-SSC1 were selected on 5-fluoroorotic acid plates. The resulting yeast strains QL2(ssc1–2), QL3 (ssc1–201),QL4 (ssc1–202), and QL5 (WTSSC1) have a His-tag version of SSC1 as the only copy of SSC1 and were used to purify Ssc1 wild-type and mutant proteins. Mitochondria were isolated from wild-type and ssc1mutant (ssc1–2 and ssc1–201) yeast strains grown at 25 °C in YPGlycerol media (1% yeast extract, 2% peptone, 3% glycerol, and 2% ethanol) as described previously (12Gambill B.D. Voos W. Kang P.J. Miao B. Langer T. Craig E.A. Pfanner N. J. Cell Biol. 1993; 123: 109-117Crossref PubMed Scopus (222) Google Scholar, 15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). Precursor proteins Su9-DHFR and Su9-DHFR* were synthesized in rabbit reticulocyte lysate in the presence of [35S]methionine. Before import, the precursor proteins were denatured by 7 murea, and mitochondria were incubated at 37 °C for 15 min to induce the temperature-sensitive phenotype. The import reactions and subsequent immunoprecipitations using Ssc1 antibodies were carried out as described (15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar). Briefly, after import for 2 min at 25 °C, the import reaction was stopped by the addition of valinomycin. Mitochondria were further incubated at 25 °C for various times then lysed with Triton X-100 buffer (100 mm NaCl, 10 mm Tris-HCl, pH 7.5, 5 mm EDTA, 0.1% Triton X-100). After a clarifying spin, the extract was incubated with Ssc1 antibody cross-linked to protein A-Sepharose beads at 4 °C for 1 h. The imported proteins bound to the protein A-Sepharose beads were analyzed by SDS-polyacrylamide gel electrophoresis and digital autoradiography. To assess folding of imported proteins, a 2-min import reaction was carried out. Mitochondria were then disrupted with 0.6% Triton X-100 and treated with proteinase K (15 μg/ml) (Roche Molecular Biochemicals) for 15 min on ice. After the digestion was stopped by the addition of phenylmethylsulfonyl fluoride, proteins were precipitated with 5% trichloroacetic acid and analyzed by SDS-polyacrylamide gel electrophoresis and digital autoradiography. The yeast strains QL2, QL3,QL4, and QL5 were grown at 25 °C in YPGlycerol media. Cells were harvested at A600 = 1.0–1.5, and mitochondria were isolated as described previously with some modifications (12Gambill B.D. Voos W. Kang P.J. Miao B. Langer T. Craig E.A. Pfanner N. J. Cell Biol. 1993; 123: 109-117Crossref PubMed Scopus (222) Google Scholar). Briefly, after treatment with zymolyase (0.45 mg/ml) (ICN Biomedicals, Costa Mesa, CA), the cells were homogenized in homogenization buffer (0.6 m sorbitol, 10 mmTris-HCl, pH 7.5, and 1 mm phenylmethylsulfonyl fluoride). Mitochondria were separated from unbroken cells and nuclei by 2,300 × g centrifugation for 5 min and pelleted from other cytosolic components by 20,000 × gcentrifugation for 12 min. The mitochondria pellet was resuspended in 0.5% Triton X-100 in IMAC buffer (20 mm Hepes-KOH, pH 7.4, 150 mm KCl, 2.5 mm magnesium acetate, 20 mm imidazole, 10% glycerol) with 1 mmphenylmethylsulfonyl fluoride. The extract was loaded onto a nickel column (Novagen, Madison, WI) equilibrated with IMAC buffer. After washing with IMAC buffer including 1 mm ATP, 1m KCl, the column was eluted with an imidazole gradient (20–240 mm in IMAC buffer). The fractions containing Ssc1 protein were pooled and loaded onto a Q-Sepharose column (Amersham Pharmacia Biotech) equilibrated with low salt buffer (20 mmHepes-KOH, pH 7.4, 50 mm KCl, 10% glycerol). After washing with low salt buffer, Ssc1 was eluted with high salt buffer (20 mm Hepes-KOH, pH 7.4, 240 mm KCl, 10% glycerol). The fractions containing Ssc1 protein were pooled, frozen in liquid nitrogen, and stored at −75 °C. Mge1, Mdj1, and DnaJ were purified as described before (9Davis J.E. Voisine C. Craig E.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9269-9276Crossref PubMed Scopus (67) Google Scholar, 16Miao B. Davis J.E. Craig E.A. J. Mol. Biol. 1997; 265: 541-552Crossref PubMed Scopus (82) Google Scholar, 18Zylicz M. Yamamoto T. McKittrick N. Sell S. Georgopolous C. J. Biol. Chem. 1985; 260: 7591-7598Abstract Full Text PDF PubMed Google Scholar, 19Karzai A.W. McMacken R. J. Biol. Chem. 1996; 271: 11236-11246Abstract Full Text Full Text PDF PubMed Scopus (202) Google Scholar, 20Horst M. Oppliger W. Rospert S. Schonfeld H.J. Schatz G. Azem A. EMBO J. 1997; 16: 1842-1849Crossref PubMed Scopus (118) Google Scholar). Peptide P5 (CALLLSAPRR) was labeled with fluorescein to generate F-P5 as described (21Montgomery D.L. Morimoto R.I. Gierasch L.M. J. Mol. Biol. 1999; 286: 915-932Crossref PubMed Scopus (129) Google Scholar). Various concentrations of Ssc1 proteins were incubated with F-P5 (10 nm) at 25 °C in buffer A (25 mm Hepes-KOH, pH 7.4, 100 mm KCl, 11 mm magnesium acetate, 10% glycerol). After binding reached equilibrium, anisotropy measurements were made with the Beacon 2000 fluorescence polarization system (Panvera, Madison, WI) at 25 °C with excitation at 490 nm and emission at 535 nm. The data were fitted to a quadratic single site binding equation by using Microsoft EXCEL to calculate theKd. In experiments utilizing ADP, Ssc1 was incubated with 500 μm nucleotide for 15 min in buffer A. Similar results were obtained with Ssc1 proteins with and without a 37 °C preincubation and in the absence of ADP. For the binding in the presence of ATP, Ssc1 proteins were incubated with 2 mm ATP at 25 °C for 30 s in buffer A, then F-P5 was added. By 2 min, before significant hydrolysis occurred, binding equilibrium was reached, and anisotropy readings were taken. F-P5 (10 nm) was incubated with Ssc1 proteins at a concentration of ∼5 μm. At this Ssc1 concentration, >80% F-P5 was bound to wild-type and mutant Ssc1 proteins at equilibrium. After the addition of a 1000-fold excess of unlabeled P5 to the reaction, anisotropy measurements were taken every 10 s to monitor the release of F-P5. The release rate koff was calculated by fitting the release curve to one phase exponential decay using Prism 2.0 (GraphPad, San Diego, CA). Ssc1·ATP complexes were isolated as described (16Miao B. Davis J.E. Craig E.A. J. Mol. Biol. 1997; 265: 541-552Crossref PubMed Scopus (82) Google Scholar), with several modifications. 25 μg of Ssc1 protein was incubated with 100 μCi of [α-32P]ATP (DuPont, 3000 Ci/mmol) in buffer A containing 25 μm ATP in a 100-μl final volume on ice for 4 min. The complex was isolated immediately on a NICK column (Amersham Pharmacia Biotech). Single-turnover ATPase assays were performed as described (16Miao B. Davis J.E. Craig E.A. J. Mol. Biol. 1997; 265: 541-552Crossref PubMed Scopus (82) Google Scholar) at 25 °C in buffer A in the presence of different proteins (Mge1, Mdj1, or DnaJ) and peptide P5 at various concentrations. At the indicated times, the reaction was stopped, and the percent conversion of ATP to ADP was determined. The rate of ATP hydrolysis was calculated by fitting the data to a first-order rate equation by nonlinear regression analysis using Prism 2.0. Firefly luciferase (4 μm) was denatured for 3 h at 30 °C in buffer B (40 mm Tris-HCl, pH 7.4, 50 mm KCl, 1 mm dithiothreitol, 15 mmmagnesium acetate) containing 6 m urea. For refolding, diluted luciferase (50 nm) was incubated at 25 °C for 2 h in buffer B (40 μl) supplemented with ATP (5 mm), an ATP regenerating system (10 mmphosphocreatine, 100 μg/ml phosphocreatine kinase), 0.15 mg/ml bovine serum albumin, and chaperone proteins as indicated in the figure legends. The luciferase activity was determined in a Beckman scintillation counter using the luciferase assay system E1500 (Promega, Madison, WI). To study protein folding within the mitochondrial matrix, DHFR was utilized as a test substrate. A protein, Su9-DHFR, containing a presequence that directs the protein synthesized in the cytosol into the mitochondrial matrix fused to the mouse DHFR was synthesized in reticulocyte lysate in the presence of [35S]methionine. The synthesized protein was denatured with urea and imported into mitochondria isolated from wild-type,ssc1–2, and ssc1–201cells. In agreement with previous results (15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar), this unfolded protein was imported intossc1–2 as efficiently as into mitochondria from wild-type or ssc1–201 cells, as very similar amounts of radiolabeled protein were imported into the 3 mitochondria preparations (Fig.1 A). The folding state of the imported DHFR was assessed by testing its susceptibility to digestion by protease. After import, the isolated mitochondria were disrupted by Triton X-100, and the extracts were treated with proteinase K. 49% of imported DHFR was proteinase K-resistant in wild-type mitochondria, whereas only 17% was resistant in ssc1–2 mitochondria. 23% of imported DHFR was resistant in ssc1–201mitochondria. Therefore, both ssc1–2 andssc1–201 mitochondria are defective in the folding of this test substrate. However, DHFR imported into ssc1–201mitochondria consistently showed slightly more proteinase K resistance than that in ssc1–2 mitochondria. Since we observed a defect in folding in ssc1–2 andssc1–201 mitochondria, we wanted to test the ability of the mutant proteins to function in the refolding of denatured proteinsin vitro. First we developed a purification scheme for wild-type and mutant Ssc1 proteins. We utilized an SSC1 gene encoding a C-terminal polyhistidine extension. These constructs rescued the lethality of the Δssc1 mutation, and strains carrying His-tagged mutant versions maintained the same temperature-sensitive growth phenotype as those carrying untagged genes (data not shown). Under the conditions used, refolding of denatured luciferase was dependent upon the presence of Ssc1, the mitochondrial Hsp40 Mdj1, and the nucleotide exchange factor Mge1 as well as Hsp78, a member of the Hsp100 family of chaperones that facilitates disaggregation of protein aggregates (22Glover J.R. Lindquist S. Cell. 1998; 94: 73-82Abstract Full Text Full Text PDF PubMed Scopus (1117) Google Scholar, 23Diamant S. Ben-Zvi A. Bukau B. Goloubinoff P. J. Biol. Chem. 2000; 275: 21107-21113Abstract Full Text Full Text PDF PubMed Scopus (203) Google Scholar). In the absence of Ssc1, less than 1% of luciferase activity was recovered, whereas in the presence of Ssc1, ∼60% of luciferase activity was found as compared with the native luciferase (Fig. 2 A). However, Ssc1–2 was not able to significantly facilitate the refolding of luciferase under any condition tested. In addition, no activity of Ssc1–201 was observed at low concentrations of Ssc1, where the refolding activity of wild-type Ssc1 is nearly maximal (1 μm). However, at increasing concentrations, some refolding activity was observed, reaching about 45% maximal wild-type activity at 6 μm. Therefore, although the suppressor protein has some refolding activity, it is severely compromised compared with wild-type protein. This result is consistent with the slightly increased level of resistance of imported DHFR to protease inssc1–201 mitochondria compared with ssc1–2mitochondria, even though the level of Ssc1 protein is similar in all three types of mitochondria (data not shown). Ssc1 interaction with imported proteins can be monitored in mitochondria by testing for the ability of Ssc1-specific antibodies to coimmunoprecipitate radiolabeled imported protein. In wild-type mitochondria, this association is transient. Consistent with previously published results (15Voisine C. Craig E.A. Zufall N. von Ahsen O. Pfanner N. Voos W. Cell. 1999; 97: 565-574Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar, 24Manning-Krieg U.C. Scherer P. Schatz G. EMBO J. 1991; 10: 3273-3280Crossref PubMed Scopus (152) Google Scholar), only a low percentage of imported protein, about 5%, was immunoprecipitated immediately after import (Fig.3 A). With time, that amount decreased, to 1–2% by 15–35 min after import. Significantly larger amounts of DHFR were coimmunoprecipitated in ssc1–2 andssc1–201 mitochondria. Initially 12–15% was precipitated; by 35 min after import, 7–9% was precipitated. Thus, ssc1–2 and ssc1–201 mitochondria showed a defect in folding of DHFR and a prolonged association with imported protein. Two possible explanations for the prolonged association came to mind. First, Ssc1–2 and Ssc1–201 might have an increased affinity for unfolded proteins. Perhaps the substrates are released more slowly once they have bound to the mutant Hsp70, causing this prolonged association. Alternatively, the mutant Hsp70s might have normal interaction with unfolded proteins but be defective in other interactions, resulting in a defect in folding. Proteins would thus remain in a partially unfolded state for a long period of time and remain substrates for Ssc1 binding. In this case each DHFR molecule would undergo many more cycles of interaction with mutant Ssc1 proteins, causing an apparent prolonged interaction. We proceeded to test these ideas. To examine substrate binding properties, wild-type, Ssc1–2, and Ssc1–201 proteins were tested in a fluorescence anisotropy-based peptide binding assay. This assay has been previously used to assess the interaction of peptide substrates with DnaK (9Davis J.E. Voisine C. Craig E.A. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 9269-9276Crossref PubMed Scopus (67) Google Scholar, 21Montgomery D.L. Morimoto R.I. Gierasch L.M. J. Mol. Biol. 1999; 286: 915-932Crossref PubMed Scopus (129) Google Scholar). For studies of Ssc1, a model peptide P5 (CALLLSAPRR), having a portion of the mitochondrial-targeting sequence of aspartate aminotransferase from chicken, was selected. P5 was fluorescently labeled on its N-terminal cysteine with fluorescein (F-P5). The anisotropy assay follows the relative rotational diffusion of the fluorescein after excitation with polarized light. Because of its small size, F-P5 should rotate in solution rather rapidly and, thus, have a low anisotropy value. When F-P5 is bound to Ssc1, it should rotate more slowly and, thus, display a significantly higher anisotropy value. Binding assays were performed using increasing concentrations of Ssc1 protein, and the increase in anisotropy was fitted to a single-site binding model. Analysis of wild-type Ssc1 binding experiments, carried out in the presence of ADP, yielded a dissociation constant (Kd) of 0.22 (±0.036) μm (Fig.4, A and B). Ssc1–2 showed about a 5-fold lower affinity for peptide, having aKd of 1.1 (±0.095) μm. However, the affinity of Ssc1–201 for peptide was only slightly lower than that of wild-type, having a Kd of 0.28 (±0.031) μm. Therefore, the suppressor mutation appears to have significantly reversed the defect of Ssc1–2 in binding peptide substrate. To look at the defect of Ssc1–2 in more detail, we monitored the displacement of the prebound F-P5 after the addition of a large excess of unlabeled P5 by measuring the decrease in anisotropy. Upon the addition of P5 to a wild-type Ssc1·F-P5 complex, the anisotropy change showed a single phase with a rate constant of 0.0024 (±0.000022) s−1 (Fig. 4 C). Experiments with Ssc1–2 revealed a very similarkoff (0.0028 (±0.000059) s−1). This small difference inkoff observed using ADP-bound Ssc1–2 cannot account for the 5-fold difference in Kd observed. Therefore, we conclude that this difference in Kd is due to a difference in the on-rate of peptide. Similar values were obtained when the experiments were performed in the absence of nucleotide. We also tested the suppressor protein Ssc1–201 to determine the basis of the restoration of the Kd to wild-type levels. The koff was decreased about 2.2-fold compared with wild-type protein, 0.0011 (±0.0000062) s−1 compared with 0.0024 s−1. Therefore, to obtain aKd similar to wild-type, thekon of Ssc1–201 would need to be 2.8-fold slower than that of wild type. We proposed that both thekon and koff are affected by the suppressor mutation, resulting in a nearly normal affinity. The affinity of Hsp70 for peptide substrates is much lower in the ATP-bound form than in the ADP-bound form (25Schmid D. Baici A. Gehring H. Christen P. Science. 1994; 263: 971-973Crossref PubMed Scopus (424) Google Scholar, 26Greene L.E. Zinner R. Naficy S. Eisenberg E. J. Biol. Chem. 1995; 270: 2967-2973Abstract Full Text Full Text PDF PubMed Scopus (93) Google Scholar). As expected, upon the addition of ATP, the anisotropy readings rapidly decreased, indicating peptide release. Because of the rapidity of this reduction, we were unable to compare the rate of release of peptide upon ATP addition among the different proteins. To look more carefully at the interaction of Ssc1 with peptide when it is in the ATP-bound state, we carried out anisotropy assays in the presence of ATP. As expected, in the presence of ATP, the affinity for peptide of both wild-type Ssc1 and Ssc1–2 was dramatically reduced (Fig. 4 A). Because of this reduction we were unable to approach saturation of binding. However, the affinity of wild-type Ssc1 and Ssc1–2 appears similar, based on the increase in anisotropy observed at high concentrations of protein. We estimate that theKd of both wild-type and Ssc1–2 proteins in the ATP-bound form is on the order of 20 μm, if at saturation the anisotropy reading is similar to that of the ADP-bound form. Therefore, there is about a 90-fold difference between the affinity of the ATP and ADP form of wild-type protein for P5. This difference is similar to the difference in affinity of ATP and ADP-bound DnaK for peptides (25Schmid D. Baici A. Gehring H. Christen P. Science. 1994; 263: 971-973Crossref PubMed Scopus (424) Google Scholar, 27Mayer M.P. Schroder H. Rudiger S. Paal K. Laufen T. Bukau B. Nat. Struct. Biol. 2000; 7: 586-593Crossref PubMed Scopus (311) Google Scholar, 28Pellecchia M. Montgomery D.L.