Mitochondrial Precursor Protein - jbc. · PDF filemitochondrial matrix-targeting signal of rat preorni- thine carbamyl transferase to murine cytosolic dihy- drofolate ... least two

THE JOURNAL OP BIOLOGICAL CHEMISTRY 0 1990 by The American Society for Biochemistry and Molecular Biology, Inc.

Mitochondrial Precursor Protein

Vol. 265, No. 19, hue of July 5. pp. 11069-11076,199O Printed in U.S.A.

EFFECTS OF 70-KILODALTON HEAT SHOCK PROTEIN ON POLYPEPTIDE FOLDING, AGGREGATION, AND IMPORT COMPETENCE*

(Received for publication, February 21, 1990)

William P. Sheffield*, Gordon C. Shore, and Stephen K. Randall From the Department of Biochemistry, McGill University, Montreal, Quebec H3G 1 Y6, Canada

A hybrid precursor protein constructed by fusing the mitochondrial matrix-targeting signal of rat preorni- thine carbamyl transferase to murine cytosolic dihy- drofolate reductase (designated pO-DHFR) was ex- pressed in Escherichia coli. Following purification un- der denaturing conditions, pO-DHFR was capable of membrane translocation when diluted directly into im- port medium containing purified mitochondria but lacking cytosolic extracts. This import competence was lost with time, however, when the precursor was di- luted and preincubated in medium lacking mitochon- dria, unless cytosolic proteins (provided by rabbit re- ticulocyte lysate) were present. Identical results were obtained for purified precursor made by in vitro trans- lation. The ability of the cytosolic proteins to maintain the purified precursor in an import-competent state was sensitive to protease, N-ethylmaleimide (NEM), and was heat labile. Further, this activity appeared to be signal sequence dependent. ATP was not required for the maintenance of pO-DHFR competence, nor did purified 70-kDa heat shock protein (the constititive form of Hsp70) substitute for this activity. Interest- ingly, however, purified Hsp70 prevented aggregation of the precursor in an ATP-dependent manner and, as well, retarded the apparent rate and extent of PO- DHFR folding. Partial purification of reticulocyte ly- sate proteins indicated that competence activity resides within a large mass protein fraction (200-250 kDa) that contains Hsp70. Sucrose density gradient analysis revealed that pO-DHFR reversibly interacts with com- ponents of this fraction. Pretreatment of the fraction with NEM, however, significantly stabilized the sub- sequent formation of a complex with the precursor. The results indicate that Hsp70 can retard precursor polypeptide folding and prevent precursor aggrega- tion; however, by itself, Hsp70 cannot confer import competence to pO-DHFR. Maintenance of import com- petence correlates with interactions between the pre- cursor and an NEM-sensitive cytosolic protein frac- tion. Efficient dissociation of the precursor from this complex appears to require a reactive thiol moiety on the cytosolic protein(s).

The fact that protein import into mitochondria may proceed

* This study was financed by operating grants (to G. C. S.) from the Medical Research Council and National Cancer Institute of Canada. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ Recipient of a studentship from Fonds de la Recherche en Sante du QuBbec. Present address: Dept. of Pathology, McMaster Univer- sity, Hamilton, Ontario L8N 325, Canada.

via a post-translational sorting pathway dictates that the precursor protein, following its release from the ribosome, maintains a conformation that will allow productive interac- tions with the proteinaceous sites that mediate membrane recognition and translocation (Attardi and Schatz, 1988; Hart1 et al., 1989). The potential role of cytosolic factors in actively promoting such an import-competent state remains unclear despite the fact that several groups have observed stimulation of polypeptide uptake into mitochondria by rabbit (Argan et al., 1983; Miura et al., 1983; Ohta and Schatz, 1984; Pfanner and Neupert, 1987; Randall and Shore, 1989; Shef- field et al., 1986; Zimmerman et al., 1988) or yeast (Murakami et al., 1988; Ohta and Schatz, 1984) cytosolic extracts. Recent genetic (Deshaies et al., 1988) and biochemical (Murakami et al., 1988) evidence implicates cytosolic 70-kDa heat shock- related protein (Hsp70)l as playing a role in the import of at least two yeast mitochondrial precursor proteins. In addition, an NEM-sensitive activity appears to play a critical role (Murakami et al., 1988; Randall and Shore, 1989).

Here, we describe a purified hybrid precursor protein whose import competence is transient unless cytosolic factors are present. Utilization of this precursor has allowed us to delin- eate activities necessary for progress along the import path- way. We have obtained a reticulocyte cytosol fraction enriched in the activity necessary for sustained import competence. The hybrid precursor interacts in a reversible fashion with components of this enriched fraction. The ability of the precursor protein to dissociate from these components corre- lates with the ability of the precursor to be imported. Further, examination of the interaction of purified Hsp70 with the hybrid precursor protein indicates that Hsp70 reduces the aggregation of the precursor (under import conditions) and apparently retards the rate of folding of the precursor, thus providing direct evidence for the interaction of Hsp70 with the mitochondrial precursor protein. Although Hsp70 alone cannot confer import competence to the precursor protein, it is probably significant that it is a constituent of the fraction that confers import competence.

EXPERIMENTAL PROCEDURES

General-Earlier studies describe the isolation of rat heart mito- chondria (Argan and Shore, 1985), determination of mitochondrial electrochemical potential (Gillespie et al., 1985), in vitro import assays (Argan and Shore, 1985), transcription of SP64 templates and trans-

1 The abbreviations used are: Hsp70, 70 kDa heat shock protein; DHFR, dihydrofolate reductase (5,6,7,8-tetrahydrofolate:NADP+-ox- idoreductase, EC 1.5.1.3); NEM, N-ethylmaleimide: SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophor&is; PO- DHFR, fusion of targeting seauence of DOCT and murine DHFR: pOCT, precursor to -orni<hine carbamyi transferase; Hepes, 4-(2: hydroxyethyl)-l-piperazineethanesulfonic acid; DTT, dithiothreitol; UCP, uncoupling protein.

11069

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

11070 Mitochondrial Precursor Proteins

lation of resultant mRNA in messenger-dependent rabbit reticulocyte lysates (Nguyen et al., 1986), SDS-PAGE, and fluorography (Shore et al., 1979). Additional details are provided in the figure legends. Restriction enzyme digestions were carried out according to the manufacturer’s specifications.

Oligonucleotides-Two synthetic oligonucleotides, 5’ CATGGT- GTCTAATTTGAG 3’ and 5’ GATCCTGAAATTAGACAC 3’, were synthesized on an ABI 380A DNA synthesizer (Applied Biosystems, Forest City, CA) and annealed by snap cooling as described (Maniatis et al., 1982) to form a NcoI-BamHI adaptor. This adaptor was employed as described below to construct the expression plasmid pODX.

Construction of Plasmids-An SP64 (Green et al., 1983) derivative was constructed, employing standard DNA manipulations (Maniatis et al., 1982), in order to serve as a template for transcription/ translation of pO-DHFR cDNA. A 650-base pair TaqI-BglII partial restriction fragment from pSV,DHFR (Subramani et al., 1981) was inserted between the AccI and BamHI sites of pSP64. PstI lineariza- tion of this plasmid at a site just 5’ to the AccI insertion site followed by T, polymerase-mediated blunt ending, and restriction at the unique SphI site yielded a major fragment, which, when ligated to a 548-base pair SphI-PuuII fragment of pSPO19 (Nguyen et al., 1986), formed pSP(pO-DHFR).

The expression plasmid pODX was constructed by introducing the NcoI-BaAHI adaptor described above, which proiided the first 6 amino acids of the DO-DHFR reading frame. and a 729-base uair BamHI-SstI partial iestriction fragment of pS$pO-DHFR), probid- ing the remaining 215 codons, between the iVco1 and SstI sites of the expression vector pOTS/Nco (Shatzman and Rosenberg, 1986). The latter was the generous gift of Dr. A. Shatzman, Smith Kline Beckman (King of Prussia, PA). pODX therefore contains an open reading frame of 221 amino acids under the control of the heat-inducible leftward promoter of bacteriophage X and consisting of the 32-amino acid pOCT signal sequence, 4 amino acids of mature ornithine car- bamoyl transferase followed by a glycine residue introduced in the manipulations, and amino acids 3-187 of mouse dihydrofolate reduc- tase. The conservative substitution Leu to Val at position 2 was also introduced in the pOCT signal sequence during the construction.

Induction and Purification of BacterialpO-DBFR-Escherichiu coli N5151 cells (Shatzman and Rosenberg, 1986) were transformed with pODX, grown overnight at 30 “C in LB b&h containing 50 fig/ml sodium ampicillin, diluted 20-fold with fresh medium, and allowed to grow at 30 ‘C to OD, 0.7-0.8. pO-DHFR synthesis was induced by the addition of an equal volume of 56 “C medium to give a final temperature of 42 “C. Induced cells were shaken for 2 h at 42 “C and then harvested by centrifugation, washed once in phosphate-buffered saline, and stored frozen at -70 “C until use. Cells obtained from 1.5 liters of culture were resuspended in 40 ml of 8.0 M urea and 10 mM sodium phosphate, pH 6.8, and lysed by two passages through a French pressure cell. Protamine sulfate was added to a 0.2% final concentration and protamine-nucleic acid complexes removed by centrifugation. The supernatant was subjected to ion-exchange chro- matography on CM-Sepharose (Pharmacia LKB Biotechnology Inc.); the 0.5 M NaCl eluate was mixed with 1 volume of 4.2 M (NH&Sod, and the precipitate recovered by microcentrifugation was washed briefly with water and resuspended in 7.0 M urea. This fraction was subjected to preparative isoelectric focusing in a flatbed of Sephadex G-200 (Pharmacia Fine Chemicals, 1982) employing Pharmalyte 8- 10.5 (Pharmacia) to generate the pH gradient. After focusing at constant power (30 watts, 5 h), protein location was determined by contact print staining (Pharmacia Fine Chemicals, 1982) and the region of interest excised from the bed, eluted with 7.0 M urea, concentrated by (NH&SO1 precipitation (as described above), di- alyzed into UHD buffer (5.0 M urea, 10 mM Hepes, pH 7.5, 1.0 mM DTT), aliquoted, and flash frozen at -70 “C. pO-DHFR concentra- tions were determined by protein assay to be approximately 150 FM; 4 liters of induced cells processed in this way yielded 1.5-2.0 mg of purified protein.

Purification of Bacterial DHFR’-Cells harboring pODX were grown, induced, and stored as described; they were resuspended in 25 mM potassium phosphate, pH 7.0, lysed in the French press, treated with-protamine sulfate, centrifuged, and the resulting supernatant combined with 3 volumes of (NH,),SO,. Following centrifugation, the pellet was resuspended in 50 mM sodium phosphate, pH 5T6, and 1.0 mM DTT and subjected to methotrexate-agarose affinity chromatog- raphy (Kaufman and Pierce, 1971) according to the manufacturer’s instructions (Sigma). DHFR’ eluted in 40 mM potassium phosphate, pH 8.5, and 1.0 mM dihydrofolate was combined with Pharmalyte

(15:1, by volume) and passed through Sephadex G-200 to mimic the conditions of PO-DHFR isolation. (NH&SO, was added to a final concentration-O.6 g/ml, and the resuiting$eci$tate was resuspended in 7.0 M urea and dialyzed overnight at 4 “C into UHD buffer. DHFR’ concentration was adjusted to 150 pM and the preparation stored as described for pO-DHFR. N-terminal sequencing indicated that DHFR’ lacks amino acids l-23 of the signal sequence.

Radiolabeling of Bacterial pO-DHFR-Bacterial cells harbouring pODX were grown overnight at 30 “C and repeatedly washed with M9 salts to reduce endogenous methionine. Cells were then adjusted to an ODsm of 1.0 in 42 “C labeling medium (0.5 X Difco methionine assay medium (Difco), 1 X M9 salts, 0.2% glucose, 1.0 mM MgSOo, 0.1 mM CaC&, 1.0 mM thiamine hydrochloride, 50 rg/ml sodium ampicillin) containing 100 &i/ml Tran3’S-Label (a mixture of [35S] methionine and 135Slcvsteine: ICN. Irvine. CA) and incubated at this temperature for i h.- fiacterial cell’harvesting,’ storage, and purifica- tion of radiolabeled pO-DHFR were as described for the nonradioac- tive preparation except that lysis was by sonication and protein detection, on the isoelectric focusing flatbed was by scintillation counting of bed samples. [36S]pO-DHFR had a specific activity of 5 X 10’ cpm/mg.

In Vitro Import of Bacterial PO-DHFR-60 ng of purified radioac- tive pO-DHFR (30,000 cpm) in UHD buffer was dil&ed 50-fold into ice-cold medium containing 40 mM KCl, 1.0 mM M~AQ, 5.0 mM Hepes, pH 7.5, 125 mM sucrose, 0.5 mM DTT, 0.5 mM ‘kTF, 2.5 mM sodium succinate, 0.04 mM ADP, and 1.0 mM potassium phosphate, pH 7.5. A %s volume of concentrated mitochondria (10 mg of protein/ ml in this same medium) was added (to give a final concentration of 0.5 mg of mitochondrial protein/ml) and the mixture incubated for 30 min at 30 “C. In assays in which reticulocyte lysate was present, the concentrations of KCl, MgAcp, DTT, and ATP in the import medium were adjusted to compensate for the introduction of these compounds (Argan and Shore, 1985) from the lysate. Unless otherwise indicated, import reactions were subsequently digested with 100 pg/ ml trypsin (Sigma) for 30 min on ice. Trypsin was inactivated by addition of a 5-fold weight excess of soybean trypsin inhibitor (Sigma), and mitochondria were recovered and analyzed by SDS- PAGE and fluorography.

Affinity Purification of in Vitro Translation Products-Reticulocyte lysate (50 ~1) translation reactions programmed with mRNA tran- scribed from linearized pSPO19 (Nguyen et al., 1986) or pSP(pO- DHFR) were diluted 20-fold in phosphate-buffered saline containing phenylmethylsulfonyl fluoride, combined with 20 ~1 of antisignal peptide antiserum (Nguyen et al., 1988), and incubated at 4 “C over- night. A 10% (w/v) suspension of protein A-Sepharose in phosphate- buffered saline containing 0.02% sodium azide (80 ~1) was then added and reactions rotated fo; 1 h at room temperature. After repeated washings. first with nhosphate-buffered saline and then with double- distilled water, elution df the precursor was performed with 10 M urea. Eluates (100 ~1) were adjusted to 5 M urea and either immedi- ately employed in import assays or flash-frozen in liquid nitrogen and stored at -70 ‘C until use. Efficiency of import appeared to decrease slowly over time when precursor was stored under these conditions; nevertheless, immunopurified precursor behaved in all manner as the purified bacterial pO-DHFR.

Fractionation of Reticulocyte Lysate-Reticulocyte lysates were ob- tained from New Zealand White rabbits according to Jackson and Hunt (1983). A high speed supematant of reticulocyte lysate was prepared by centrifugation at 180,000 x g for 60 min and proteins precipitated by addition of (NH&SO, to 30% saturation. The super- natant obtained by centrifugation (SS34 rotor, 20 min, 10,000 rpm) was brought to 50% saturation in (NH&SO,, recentrifuged, and the resulting pellet resuspended in 15 mM KPi, pH 7.5, 10% glycerol, 1 mM DTT, and chromatographed on a column of Sepharose CL-6B (equilibrated in the same buffer). Fractions were collected, flash frozen in liquid nitrogen, and stored at -70 “C until use.

Velocity Sedimentation Analysis-After incubation of [3”S]pO- DHFR under various conditions for 2 h at 30 ‘C, 0.5-ml samples were layered on top of 5.0-ml20-60% (w/w) sucrose gradients containing 15 mM KPi, pH 7.5,2 mM MgAc2, and where present, 1 mM ATP and centrifuged in a VTi-80 rotor for 2 h at 60,000 rpm. Fractions were collected and sedimentation profiles of [35S]pO-DHFR determined by scintillation counting. The remaining portion of the fraction was concentrated on Centricon 10 units (Amicon Corp.) and washed with import buffer. Imports were then carried out following the addition of mitochondria.

Zmmunoblot Analysis-Immunoblotting was performed as de- scribed (Towbin et al., 1979) except that alkaline phosphatase-con-

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Mitochondrial Precursor Proteins 11071

jugated second antibodies were used (Protoblot system; Promega Biotec, Madison, WI). Antibodies directed against the first 27 amino acids of rat pOCT (anti-SP; Nguyen et al., 1988) were affinity purified employing pO(l-27) peptide (Gillespie et al., 1985) immobilized on CnBr-activated Sepharose (Pharmacia) by standard procedures. Anti- serum to mouse DHFR was the generous gift of Dr. Wayne Flintoff, University of Western Ontario.

Densitometry-Appropriately exposed autoradiograms and stained gels were scanned on a LKB Ultrascan XL enhanced laser densitom- eter. Regression analysis was employed to generate lines of best fit of results quantified by densitometry.

RESULTS

Expression of pO-DHFR in bacteria was achieved by in- serting the recombinant cDNA into pOTS/Nco, a derivative of the PAS-1 vector (Shatzman and Rosenberg, 1986), in which the cDNA comes under the transcriptional control of the heat-inducible X leftward promoter (Fig. 1A). Following transformation of E. coli N5151 cells and induction at 42 “C, a significant amount of pO-DHFR was made, accounting for roughly 3% of total bacterial protein (Fig. 1B). The hybrid protein was purified to near homogeneity by extraction of protein in 7 M urea followed by sequential fractionation of the soluble extract by CM-Sepharose chromatography and preparative isoelectric focusing in a Sephadex G-200 bed (Fig.

A

B

DI 23 4 COO-WFI

‘“8 -p - -m

C STAIN BLOT

STAIN ANTI-SP - - -

ANTI -OR ANTI -SP

a b ab

ml 2 1 2 1 2

FIG. 1. Bacterial expression, purification, and character- ization of pO-DHFR. Panel A shows relevant portions of the expression plasmid pODX, with promoter (XL) and transcription termination elements (to), relevant restriction sites, and protein coding regions (wide bones). Shaded region, 32-amino acid pOCT signal sequence. Panel B shows the SDS-PAGE profiles of E. coli N5151 cells transformed with pODX, grown either at 30 “C (lane a) or after 2 additional h of induction at 42 “C (lane b) and analyzed either by Coomassie Blue staining (stain) of a 12% SDS gel or by immunoblotting employing antisignal peptide antibody (anti-SP). Panel C shows the SDS-PAGE profiles of 4.0 rg of pO-DHFR (lane I) or 4.0 Fg of its proteolytic derivative, DHFR’ (lane 2), purified as described under “Experimental Procedures,” analyzed either by Coo- massie Blue staining of a 12% SDS gel (stain) or by immunoblotting of the gel employing anti-DHFR antiserum (anti-DE) or antisignal peptide antibody (anti-SP). m (markers): phosphorylase b (94 kDa), albumin (67 kDa), ovalbumin (43 kDa), carbonic anhydrase (30 kDa), soybean trypsin inhibitor (20.1 kDa), and a-lactalbumin (14.4 kDa). Panel D shows the in vitro imuort of radioactive DO-DHFR. Lane 1. 6 ng of pO-DHFR, representing 10% of input precursor, no mite: chondria added, lane 2, mitochondrial pellet after import incubation; lane 3, mitochondrial pellet after import and trypsin treatment; lane 4, mitochondrial pellet after import in the presence of 1.0 PM carbonyl cyanide p-chlorophenylhydrazone followed by trypsin treatment. An autoradiogram of the dried and fluorographed gel is shown; the positions of precursor (p) and processed, mature (m) products are indicated.

1C). Upon rapid dilution of the purified denatured product to reduce the concentration of urea to 100 mM and to introduce the protein into a buffer medium that supports protein import into mitochondria in uitro, pO-DHFR was found to acquire enzymatic activity rapidly (see Fig. 9) as well as the ability to be imported and processed by purified and well washed mi- tochondria in the absence of additional cytosolic components (Fig. 1D). Thus, this relatively small monomeric reporter protein fused to the matrix-targeting signal sequence of pre- cursor to ornithine carbamoyl transferase rapidly acquires a functional form following dilution out of urea.

If high concentrations of urea were not present during the early steps of protein extraction from E. co& the recovery of a proteolytic fragment of pO-DHFR resulted. The breakdown product, designated DHFR’, was enzymatically active (not shown), indicating that it largely retained its native DHFR structure, but it was missing the pOCT signal sequence (Fig. 1C) and consequently was not imported by mitochondria in vitro (not shown). DHFR’ was purified in this study to provide a control product that lacks a mitochondrial targeting signal but, like pO-DHFR, is a recombinant expression product of bacteria and has undergone isolation conditions that are very similar to those used for functional, full-length pO-DHFR (see “Experimental Procedures”).

pO-DHFR but not DHFR’ effectively competed for import into mitochondria with precursor proteins destined either for the inner membrane (uncoupling protein) or for the matrix (pOCT) (Fig. 2). These findings demonstrate that competition for import is dependent upon the presence of a functional signal sequence on DHFR (Fig. 1D) and that the bacterial expression product of pO-DHFR at relatively low concentra- tions can saturate one or more rate-limiting steps of the import pathway. The profile of pO-DHFR competition for pOCT import was very similar to that obtained for an intrinsic protein of the mitochondrial inner membrane, the uncoupling protein (UCP) of brown adipose tissue (Fig. 2B). Uncoupling protein is made without a cleavable signal sequence (Ridley et al., 1986) and is targeted to mitochondria by internal signals that functionally may resemble typical matrix-targeting se- quences (Liu et al., 1988). Fifty percent inhibition of import of both pOCT and UCP occurred at a concentration of com- petitor pO-DHFR of 0.2-0.3 pM (Fig. 2). Interestingly, this concentration of pO-DHFR required to inhibit import of precursors by 50% corresponded to approximately %o the concentration of a synthetic pOCT signal peptide that was required to bring about a comparable level of import inhibition (Gillespie et al., 1985). Taken together with the observation that pO-DHFR itself is imported to the matrix compartment of mitochondria (Fig. 10) and that neither excess pO-DHFR nor excess DHFR’ had any deleterious effect on the mito- chondrial electrochemical potential (not shown), the results suggest strongly that the bacterial expression product of PO- DHFR is employing the standard import pathway for protein translocation into the organelle.

Transient Import Competence of Purified pO-DHFR-Al- though purified pO-DHFR is capable of being imported by isolated mitochondria in the absence of additional factors provided by a cell cytosol, its ability to do so is relatively short lived following the dilution of pO-DHFR out of concentrated urea (Fig. 3). Import of bacterial pO-DHFR was reduced by 80% following preincubation for 1 h at room temperature prior to the addition of mitochondria. Addition of reticulocyte lysate had only a very modest stimulatory effect on the import of bacterial pO-DHFR immediately following dilution of PO- DHFR out of concentrated urea (Fig. 3, zero time point), but lysate had a dramatic effect on preventing the loss of import

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

11072

A

100 .

t 60 P z 60 ?

fI.fs

F40 l 4 .

: 20

0 0.0 0.2 0.4 0.6 0.6

COMPETITOR CONC. (uhf)

B 100

t 60 0

8 z60 .

Y 20 i;--l-

Y i=40 s , .

0 0.0 0.2 0.4 0.6 0.6

COMPETITOR CONC. (14)

Mitochondrial Precursor Proteins

pOCT

123

UCP

rr, -P

i&.4 123

FIG. 2. Inhibition of pOCT and UCP import by excess PO- DHFR but not by DHFR’. Unlabeled bacterial expression products, pO-DHFR or DHFR’, in 5.0 M urea were diluted into ice-cold import assay medium lacking mitochondria but containing radioactive in uitro translation products (either pOCT (pOC’2’) or UCP) and a final concentration of 5.0 mg/ml with respect to total reticulocyte lysate protein. Concentrated mitochondria were added to start the 20-min import incubation; the reaction was terminated by the addition of carbonyl cyanide p-chlorophenylhydrazone and mitochondria and then trypsinized. Products were resolved by SDS-PAGE and fluorog- raphy, and the relative amount of imported product was quantified by laser densitometry. Results were plotted uersus concentration (PM) of competing amounts of either pO-DHFR (closed circles) or DHFR’ (open circles). Fluorographs of import assays from the same experi- ment but with no added competitor are also shown. Lone 1, 10% of input precursor; lane 2, mitochondrial pellet after import and subse- quent trypsin treatment; lane 3, as in lane 2 but import was carried out in the presence of carbonyl cyanide p-chlorophenylhydrazone. Panel A, pOCT; panel B, UCP.

100 . l .

+LYSATE ; 80 P E 60

F F 40 s L 20

I

0 L -LYSATE

kd”LiAT:ON (/OURS;

FIG. 3. Effect of preincubation on import competence of bacterial pO-DHFR diluted out of urea in the presence or absence of reticulocyte proteins. Import of radiolabeled PO- DHFR was as described in Fig. 1 except that diluted precursor in import medium (see “Experimental Procedures”) was preincubated at 20 “C for the indicated times in the presence (closed circles) or absence (open circles) of 50 mg/ml reticulocyte proteins. Products were resolved by SDS-PAGE and fluorography and the processed, trypsin-protected product quantified by laser densitometry. Results were plotted uersus time of preincubation.

competence of pO-DHFR (Fig. 3) even for periods as long as 16 h (not shown). Reticulocyte lysate showed little or no capacity, however, to restore competence to otherwise incom- petent precursor (see Fig. 6), suggesting that lysate factor primarily functions to stabilize rather than induce the import competent conformation of pO-DHFR.

The loss of import competence of pO-DHFR did not cor- respond to a breakdown or to a substantive covalent modifi- cation, at least insofar as these modifications could be de- tected by an altered mobility following SDS-PAGE. Moreover, the transient import competence did not arise as a conse- quence of bacterial expression of the precursor since PO- DHFR purified from reticulocyte lysate translation mixture by affinity binding and elution in urea behaved in the same fashion as the precursor purified from E. coli (see Fig. 6). Further, it is important to note here that the bona fide mitochondrial matrix precursor, pOCT, following urea dena- turation, similarly showed a dependence upon lysate for effi- cient import (not shown).

Signal Sequence Dependence of the Competence Factor-A dose-response curve for the concentration of reticulocyte ly- sate that is required to maintain pO-DHFR competent for translocation into mitochondria showed that at an input level of pO-DHFR of 1.2 pg/ml, a 10% volume of reticulocyte lysate (10 mg of protein/ml) resulted in maximal import (Fig. 4A). To test whether there was a requirement for the signal se- quence on pO-DHFR for the competence activity, the effect of a 20-fold molar excess of DHFR’ relative to pO-DHFR was determined when a subsaturating amount of reticulocyte ly- sate protein (5 mg/ml) was preincubated with pO-DHFR. This amount of lysate corresponds to roughly 80% of that required to have a maximal effect on input pO-DHFR (see arrow in Fig. 4A). Inclusion of excess DHFR’ in the preincu- bation period did not interfere with the ability of lysate to maintain the import competence of pO-DHFR (Fig. 4B). This lack of competition suggests that it is the presence of the pOCT signal sequence on DHFR which is required for recog- nition by the competence factor. As expected, when excess unlabeled pO-DHFR was included in the competence assay, radioactive pO-DHFR was not subsequently imported (not shown), but of course the presence of excess pO-DHFR would be expected to inhibit the import of radioactive precursors at a number of steps on the import pathway (see Fig. 2).

Fractionation of Reticulocyte Lysate-In an earlier study, we identified an NEM-sensitive activity that stimulated im- port of pOCT in vitro (Randall and Shore, 1989). When lysate was fractionated on Sepharose CL-GB, stimulatory activity eluted in a fraction corresponding to an apparent mass of 200-250 kDa (Fig. 5). This fraction also corresponded to an activity capable of conferring import competence to bacterial

A

*hC 1 2 3

ob7i+Tii LYSATE CONC. (ma/ml)

FIG. 4. Dose-response relationship of bacterial pO-DHFR to reticulocyte import competence factor and insensitivity of factor to excess DHFR’. Panel A. nurified radioactive DO-DHFR (5 ng) was preincubated in the presence of varying amounts of reticulocyte lysate for 3 h at 20 “C prior to subsequent import. Import was quantified by laser densitometry and the results plotted uersus lvsate protein concentration. Arrowhead denotes concentration of reticulocyte lysate protein required to retain 80% import competence of pO-DHFR. Panel B, 5 ng of radioactive pO-DHFR was combined with a 20-fold molar excess of unlabeled DHFR’ and diluted into preincubation medium containing 5 mg/ml reticulocyte lysate protein for 3 h at 20 “C: subseauent imnort was analvzed bv SDS-PAGE and fluorography (lane 3). iane 1, iO% input; l&e 2, fmport pellet; lane 3, as in lane 2 except with excess DHFR’ in the preincubation.

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Mitochondrial Precursor Proteins

-5

I 30 50 70 90

FRACTION

FIG. 5. Resolution of reticulocyte cytosol proteins and im- port stimulatory activity by gel filtration. Reticulocyte lysate proteins (4000 mg) were ammonium sulfate fractionated, separated on a Sepharose CL-6B column (450 ml), and fractions of 5.0 ml collected as described under “Experimental Procedures.” Protein (closed circles) was determined according to Bradford (1976). Stimu- lation of pOCT import was measured by utilizing the dilution of native precursor (i.e. nondenatured) method described previously (Randall and Shore, 1989). The average amount of imports was obtained from each fraction by laser densitometric analysis of two separate SDS-PAGE, and import was expressed (open circles) as fold- stimulation relative to a control containing no additional lysate proteins (control activity = 1.0). In this representative experiment, control import represented 2.5% of total input precursor whereas the CL-6B peak of stimulation had an import of approximately 10.5% of total input precursor. An estimate of molecular mass of the activity peak (inset, open square) was obtained by passing the comparing elution volumes of standards on the same column in separate runs, cytochrome c (12 kDa), bovine albumin (68 kDa), aldolase (158 kDa), and blue dextran 2000.

pO-DHFR and based on this activity, represented a purifica- tion of approximately 500-fold relative to whole lysate (based on the amount of protein needed to give 50% of maximal competence activity). We have utilized this enriched fraction (designated 6B) to characterize the requirements for import competence and the interactions of this activity with precur- sor proteins. For convenience, pO-DHFR translation product purified from reticulocyte lysate (see “Experimental Proce- dures”) was used for routine assays in this regard. Although this product in all instances yielded the same response to competence factor as did bacterial pO-DHFR, the efficiency of its import was somewhat reduced possibly because the removal of contaminating lysate proteins was carried out in the absence of urea. When the urea-denatured translation product, pO-DHFR, was preincubated in the presence of 6B proteins, import competence was maintained (Fig. 6, fifth lane from left), so the 6B fraction thus substitutes for complete lysate in this regard (see also Fig. 7). Although the 6B fraction was sufficient (shown in Fig. 6, at saturating concentrations) to confer import competence, it was also observed that whole lysate had a moderately stimulatory effect when included in the subsequent import incubation (Fig. 6, sixth lane from left; and also see Fig. 7), suggesting that steps subsequent to import competence may as well be influenced by cytosol.

The ability of the 6B fraction to confer competence was not dependent on ATP (Fig. 6, first lane from right) or other nucleotide triphosphates. NEM pretreatment of the 6B frac- tion, however, prevented subsequent import of the precursor (Fig. 6, seventh lane from left), but it should be noted that we cannot determine whether the competence activity itself is NEM sensitive or if the NEM-treated 6B fraction blocked a

PRE- INCUBATION: - - - 6B 6B NEM :;c66 6B

6B 66 I-ATPI

ASSAY 1 IO%

- lys cccp - lys - - cccp -

FIG. 6. Fractionated reticulocyte lysate maintains import competence of purified pO-DHFR in an ATP-independent NEM-sensitive manner. A reticulocyte lysate fraction, designated 6B peak and prepared as described in Fig. 5 and under “Experimental Procedures,” was tested for the ability to maintain immunopurified pO-DHFR (from in uitro translations) competent for import. Follow- ing a preincubation (2 h at 30 “C), the import assay was initiated by the described additions immediately followed by the addition of rat heart mitochondria. The additions to the preincubation or the sub- sequent import assay (1 h at 30 “C) were saturating amounts of lysate (10 mg/ml) or 6B peak fraction (38 pg/ml). The final concentration of carbonyl cyanide p-chlorophenylhydrazone (cccp), where added, was 1 pM. All preincubations contained 1 mM ATP except where noted. Preincubation and import buffers were as described under “Experimental Procedures.” To inactivate 6B with NEM, the fraction was treated for 30 min at 30 “C with 3 mM NEM, and then excess NEM was inactivated with equal molar DTT. Mock treatment uti- lized NEM previously inactivated with equal molar DTT. A fluoro- graph of a 12% SDS gel is shown. Arrow shows position of fully processed import product. IO%, %o of precursor employed in the assay.

PRE- I NCUBATION : - - hsp hsp lys 1~s lys

ASSAY : - lys - lys - hsp lys IO%

FIG. 7. Hsp70 alone cannot confer import competence. Fol- lowing a preincubation in the presence of either saturating amounts of reticulocyte lysate (lys, 10 mg/ml; see Fig. 4) or purified Hsp70 (hsp, approximately 5 Fg/ml), the pO-DHFR import assay was initi- ated by the addition of either additional lysate or Hsp70, and mito- chondria as in Fig. 6.

later step, e.g. binding of precursor to a receptor. A trivial explanation for the observed NEM effect (i.e. that NEM- treated proteins were intrinsically inhibitory to import) was ruled out by the observation that pO-DHFR directly diluted from urea into NEM-treated lysate in the presence of mito- chondria was still capable of import (not shown).

Members of the Hsp70 family of proteins have been sug- gested to have a role in the translocation of both endoplasmic reticulum and mitochondrial precursor proteins (Chirico et al., 1988, Deshaies et al., 1988). We have thus examined Hsp70 purified from reticulocyte lysate (Fig. 8) for its effect on import competence (Randall and Shore, 1989 and Fig. 7). At a concentration approximately equivalent to the content of Hsp70 in the reticulocyte lysate employed in the assays de- scribed here (Figs. 3 and 4) as well as at 20-fold higher concentrations Hsp70 had no effect on precursor competence (i.e. when present in the preincubation; Fig. 7). This same preparation of Hsp70 was found to be active in moderately stimulating import of another precursor (nondenatured pOCT; Randall and Shore, 1989) and promoted ATP-depend- ent prevention of aggregation of pO-DHFR (see Fig. 10); nevertheless, it was without apparent effect on pO-DHFR import (Fig. 7). Although Hsp70 did not confer import com-

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

11074 Mitochondrial Precursor Proteins

A, 1234 B

60 hsp70 -- CnBr 77 --

--+I@- - 70 - -- IIC

relative activity 1 7.5 ND

FIG. 8. Comparison of purified Hsp70 and CL-6B peak frac- tion by SDS-PAGE. A, lane 2, reticulocyte lysate high speed super- natant (1200 pg), lane 3, 6B fraction (19 rg), and lane 4, Hsp70 (0.25 pg) were obtained as described under “Experimental Procedures,” siparated on 12% polyacrylamide gels, and-stained with Coomassie Brilliant Blue R-250. The relative amounts of competence activity for each fraction were established by titrating the amount of fraction required to give 50% of maximal competence activity. ND, no com- petence activity detected. B, the 70-kDa polypeptides (arrow) were excised from the 6B fraction or Hsp70 (resolved on gel in A), subjected to CnBr cleavage (or mock treated, minus CnBr), and then separated on 15% SDS-PAGE and silver stained. Molecular mass markers lane I were 94,67,43, X0,20 (A and B), and 14 kDa (B) as shown.

petence, it was apparently contained within the 6B complex (confirmed by CnBr cleavage fingerprinting, Fig. 8).

Recovery of Enzymatic Activity following Dilution of PO- DHFR Out of Urea-Upon dilution of bacterial pO-DHFR out of urea, renaturation, as measured by the acquisition of dihydrofolate reductase activity, was rapid. Activity could be detected within 10 s, and a maximal, stable rate was attained within 5 min (Fig. 9). This rate was then constant up to at least 30 min. The presence of Hsp70 clearly reduced both the initial activity rates (particularly between 0 and 1 min) and the final stabilized rate (after 5 min). Not surprisingly (since Hsp70 is present in 6B), the partially purified 6B fraction containing competence activity also reduced recovery of en- zymatic activity. The data (Fig. 9) indicate that approximately 50 ng of Hsp70 and 4.0 Fg of 6B in the assay resulted in approximately 50% inhibition of the final stabilized reductase activity associated with approximately 7 pg of bacterial PO- DHFR. Maximal inhibition of approximately 85% could be obtained at the highest concentrations of 6B or Hsp70 tested, and this inhibition of refolding was nucleotide triphosphate dependent (not shown). When Hsp70 or 6B was presented to renatured pO-DHFR (i.e. 5 min after dilution out of urea), no difference in enzymatic activity was detected (not shown), indicating that there was no inhibitory effect after pO-DHFR had refolded. No significant effect on the recovery of either DHFR’ or chicken DHFR activity following renaturation out of urea was found by either Hsp70 or 6B (not shown), indi- cating that the presence of the signal sequence on this protein was responsible, whether directly or indirectly, for recognition by Hsp70 or 6B.

Sucrose Density Gradient Analysis of pO-DHFR-Sedimen- tation analyses were undertaken to examine the potential interaction between pO-DHFR and components of the 6B fraction. The precursor alone (monomeric mass = 26 kDa), following dilution from urea and incubation for 2 h to render

hsp 70

A340 2 0.1

2min

\ \

\ .9dC%O” FIG. 9. Recovery of pO-DHFR activity during renaturation

is prevented by purified Hsp70 or 6B fraction. Dihydrofolate reductase activity was determined by monitoring NADPH oxidation spectrophotometrically essentially as described previously (Osborn and Huennekens, 1958). The complete assay mixture (0.75 ml) con- tained 66 mM KPi, pH 7.5, 13 mM o-mercaptoethanol, 66 /.LM dihy- drofolate, 66 pM NADPH, 1 mM ATP. There was no detectable NADPH oxidation in the absence of pO-DHFR (i.e. neither Hsp70 nor the 6B fraction catalyzed NADPH oxidation). Complete assay mixtures (minus pO-DHFR) were preincubated with either control buffer, buffer containing Hsp70 (0.05 pg), or 6B (3.8 fig). The assay was initiated by the addition of 5.0 ~1 of purified bacterial expressed pO-DHFR (7 pg) in 7.0 M urea (at arrow) into the assay and activity measured at 22 “C.

it import incompetent, sedimented as an aggregate of apparent molecular mass of 150 kDa, with a minor species of apparent molecular mass of 400 kDa (Fig. 10B). In contrast, when PO- DHFR was incubated with the 6B fraction following dilution from urea, it reproducibly sedimented with an apparent mass of 230 kDa (Fig. lOA), i.e. slightly larger than that of PO- DHFR alone (Fig. 10B). Fig. 1OA also shows the effect of pretreating the 6B fraction with NEM prior to analyzing the interaction between 6B and pO-DHFR. Under these condi- tions, a stable and even larger complex with pO-DHFR was formed, corresponding to a mass of approximately 350 kDa. When the peak fractions of pO-DHFR.6B complexes (or those of pO-DHFR alone) in Fig. 10 were recovered and analyzed for import of the constituent pO-DHFR into mito- chondria, only that associated with the 6B complex that was not treated with NEM was found to be imported with high efficiency. Neither pO-DHFR associated with the NEM- treated complex nor pO-DHFR incubated in the absence of 6B was import competent. In both cases and in all of the sucrose gradients described in this study, pO-DHFR was recovered as the full-sized precursor polypeptide, as judged by SDS-PAGE.

Preincubation of precursor pO-DHFR with either Hsp70 (Fig. 11) or 6B (Fig. 12) in the presence of ATP resulted in a significant conversion of the higher mass species of the pre- cursor to a species of precursor that was apparently mono- meric in size (recovered at the top of the gradient). This may represent either deaggregation or the prevention of aggrega- tion (this assay does not distinguish between the two, as Hsp70 or 6B were present during pO-DHFR renaturation). It is likely that for 6B, the activity can be ascribed to the presence of Hsp70 contained within that fraction. However,

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Mitochondrial Precursor Proteins 11075

Volume. ml

FIG. 10. NEM pretreatment of the 6B fraction results in the formation of a stable pO-DHFR. 6B complex. A, the 6B fraction was treated with 3 mM NEM (open. circles) for 30 min at 30 ‘C, and then excess NEM was inactivated with equal molar DTT. In the mock treatment (closed circles), NEM was first inactivated with DTT before adding to the 6B fraction. After a 2-h incubation with immu- nopurified pO-DHFR (20,000 cpm) the mixture was resolved on sucrose gradients as described under “Experimental Procedures.” The amount of 6B (76 fig, 0.19 fig/ml) in these assays was approximately 5-fold excess over that required for saturation in the import assays in order to favor pO-DHFR in the complexed form. B, PO-DHFR was preincubated in the absence of 6B and then resolved on sucrose gradient as in A. Mass markers were ferritin (Fe, 440 kDa), catalase (Co, 220 kDa), and hemoglobin (Hb, 60 kDa).

when peak fractions were recovered from the gradients and tested for ability to be imported, only precursor preincubated with the 6B fraction was competent for import.

DISCUSSION

The initial observations on the requirements for members of the Hsp70 family in the post-translational translocation of proteins into the endoplasmic reticulum and mitochondria revealed that Hsp70 plays an important, although not exclu- sive, role in the cytoplasm (Chirico et al., 1988, Deshaies et al., 1988; Murakami et al., 1988); an additional protein(s), sensitive to NEM, is also required (Chirico et al., 1988; Mu- rakami et al., 1988; Sanz and Meyer, 1989; Zimmerman et al., 1988). Suggested roles for Hsp70 have included an ability to maintain precursor proteins in a nonaggregated form (Pel- ham, 1986); an ability to retard the folding of nascent chains during polypeptide synthesis, which in turn aids the formation of a correct (in this case, translocation-competent) confor- mation (Rothman, 1989); and an ability to function as an ATP-dependent “unfoldase” (Rothman and Kornberg, 1986). The last possibility is consistent with the requirement for polypeptide unfolding either before or during translocation across membranes (Eilers and Schatz, 1986). Consistent with the suggestion that Hsp70 might maintain precursors in a nonaggregated form is the observation that proteins com-

I I I I I 1 2 3 4 5

Volume, ml FIG. 11. Hsp’70 reduces aggregation of pO-DHFR. pO-DHFR

was preincubated with Hsp70 (1 pg, 2.5 rg/ml final concentration), approximately 50% of that utilized in import assays (Fig. 7), in the presence (open circles) or absence (closed circles) of 2 mM ATP and then separated on sucrose gradients (ATP either present or absent in the gradients as indicated) as described in Fig. 10). Markers were as described in Fig. 10. P, precursor.

Volume, ml FIG. 12. pO-DHFR can stably interact with the 6B fraction

in the absence of ATP. pO-DHFR and the 6B fraction were preincubated for 2 h at 30 “C (as in Fig. 10) in the absence (closed circles) or presence (open circles) of 2 mM ATP and were then resolved on sucrose gradients (either containing ATP or not), as described under “Experimental Procedures.” The position of precursor (in the absence of any additional proteins) after sedimentation is indicated by P (and the profile is shown in Fig. 10).

prised of multimeric subunits are probably dissociated by an ATP-dependent mechanism prior to or during import into mitochondria (Chen and Douglas, 1988).

Here, we present evidence that Hsp70, purified from retic- ulocyte lysate, can maintain a mitochondrial precursor protein (PO-DHFR) in a nonaggregated form by a mechanism that

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

11076 Mitochondrial Precursor Proteins

requires ATP. In addition, Hsp70 can retard the folding of the precursor protein following dilution of the latter out of concentrated urea. It may be that urea-denatured polypep- tides mimic the local structure of polypeptides as they emerge from the ribosome during synthesis (Sanz and Meyer, 1989) so that this finding may be consistent with the suggestions (Flynn et al., 1989; Rothman, 1989) that Hsp70 can interact with multiple sites on a nascent chain to prevent misfolding and therefore promote correct folding. Nevertheless, we pres- ent evidence that these effects of Hsp70 on pO-DHFR are not sufficient to maintain the precursor competent for import into mitochondria but rather that competence requires an additional NEM-sensitive cytosolic activity.

Our observations on the cytosol-dependent maintenance of import competence of urea-denatured pO-DHFR are reminis- cent of the bacterial plasma membrane (Collier et al., 1988; Crooke et al., 1988) and yeast endoplasmic reticulum (Sanz and Meyer, 1989) translocation systems. Whether the require- ment for cytosolic competence activity provided by reticulo- cyte lysate reflects a requirement to maintain the signal sequence of pO-DHFR available for recognition by its cognate receptor or to prevent the mature region of the molecule from acquiring a structure that is incompatible with the mitochon- drial unfolding machinery, or both, is not presently known. Nevertheless, from competition experiments we conclude that maintenance of import competence by the reticulocyte lysate factor, whether directly or indirectly, requires the presence of a signal sequence on pO-DHFR. This could result from a direct recognition of the signal sequence or a recognition of a mature portion of the precursor which may be unmasked by the influence of the signal.

The cytosolic competence activity in reticulocyte lysate resides within an approximately ZOO-kDa protein fraction that is sensitive to the alkylating reagent, NEM. Components of the fraction interact reversibly with pO-DHFR to maintain import competence, and dissociation of the precursor from the complex appears to require a reactive thiol on the factor. The maintenance of import competence by the factor is independent of nucleoside triphosphates, but of course extra- mitochondrial ATP is required for subsequent import steps (Chen and Douglas, 1988; Ohta and Schatz, 1984; Pfanner and Neupert, 1987). Interestingly, Hsp70 is a constituent of the 200-kDa protein fraction. Despite the fact that Hsp70 alone cannot confer import competence to pO-DHFR, its ability to retard polypeptide folding and prevent aggregation raises the possibility that Hsp70 might play a role in “holding” unfolded precursors, thus allowing a secondary factor to in- teract and confer import competence upon the precursor, analogous to the suggestion proposed by Rothman (1989).

Acknowledgments-We would particularly like to thank Douglas Millar for his excellent technical assistance in some of these experi- ments and Dr. Allan R. Shatzman and Dr. Martin Rosenberg of Smith Kline & Beckman Laboratories for their gracious provision of the E. coli expression vectors utilized in this paper.

REFERENCES

Argan, C. A., and Shore, G. C. (1985) Biochem. Biophys. Res. Com- mm. 131,289-298

Argan, C., Lusty, C. J., and Shore, G. C. (1983) J. Biol. Chem. 258, 6667-6670

Attardi, G., and Schatz. G. (1988) Annu. Reu. Cell Biol. 4, 289-333 Bradford, %I. M. (1976) Anal. Biochem. 72, 248-254 Chen. W.-J.. and Doualas. M. G. (1988) J. Biol. Chem. 263. 4997-

5060 I

Chirico, W. J., Waters, M. G., and Blobel, G. (1988) Nature 332, 805-810

Collier, D. N., Bankaitis, V. A., Weiss, J. B., and Bassford, P. J., Jr. (1988) Cell 53, 273-283

Crooke, E., Brundage, L., Rice, M., and Wickner, W. (1988) EMBO J. 7, 1831-1835

Deshaies, R. J., Koch, B. D., Werner-Washburne, M., Craig, E. A., and Schekman, R. (1988) Nature 332,800-805

Eilers, M., and Schatz, G. (1986) Nature 322,228-232 Flynn, G. C., Chappell, T. G., and Rothman, J. E. (1989) Science

245,385-390 Gillespie, L. L., Argan, C. A., Taneja, A. T., Hodges, R. S., Freeman,

K. B., and Shore, G. C. (1985) J. Biol. Chem. 260,16045-16048 Green, M. R., Maniatis, T., and Melton, D. A. (1983) Cell 32, 681-

684 Hartl, F.-U.. Pfanner. N., Nicholson, D. W., and Neupert, W. (1989)

Biochim. kophys. Act; 988, l-45’ _

Jackson. R. J.. and Hunt, T. (1983) Methods Enzvmol. 96.50-74 Kaufman, B. ‘T., and Pierce; J. t. (1971) Biochem. Biobhys. Res.

Commun. 44,608-613 Liu, X., Bell, A. W., Freeman, K. B., and Shore, G. C. (1988) J. Cell

Biol. 107,503-509 Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular

Cloning: A Laboratory Manual, pp. l-545, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Miura, S., Mori, M., and Tatibana, M. (1983) J. Biol. Chem. 258, 6671-6674

Murakami, H., Pain, D., and Blobel, G. (1988) J. Cell Biol. 107, 2051-2057

Nguyen, M., Argan, C., Lusty, C. J., and Shore, G. C. (1986) J. Biol. Chem.26 1,800-805

Nguyen, M., Bell, A. W., and Shore, G. C. (1988) J. Cell Biol. 106, 1499-1505

Ohta, S., and Schatz, G. (1984) EMBO J. 3,651-657 Osborn, M. J., and Huennekens, F. M. (1958) 233,969-974 Pelham, H. R. B. (1986) Cell 46,959-961 Pfanner, N., and Neupert, W. (1987) J. Biol. Chem. 262,7528-7536 Pharmacia Fine Chemicals, (1982) Isoelectric Focusing, pp. 77-106,

Ljungforetagen AB, Orbro, Sweden Randall, S. K., and Shore, G. C. (1989) FEBS Lett. 250, 561-564 Ridley, R. G., Patel, M. V., Gerber, G. E., Morton, R. C., and Freeman,

K. B. (1986) Nucleic Acids Res. 14,4025-4035 Rothman, J. E. (1989) Cell 59,591-601 Rothman, J. E., and Kornberg, R. D. (1986) Nature 322,209-210 Sanz, P., and Meyer, D. I. (1989) J. Cell Biol. 108,2101-2106 Shatzman, A. R., and Rosenberg, M. (1986) Ann. N. Y. Acad. Sci.

478,233-248 Sheffield, W. P., Nguyen, M., and Shore, G. C. (1986) Biochem.

Biophys. Res. Commun. 134,21-28 Shore, G. C., Carignan, P., and Raymond, Y. (1979) J. Biol. Chem.

254,3141-3144 Subramani, S., Mullighan, R., and Berg, P. (1981) Mol. Cell. Biol. 1,

854-864 Towbin, H., Staehlin, T., and Gordon, J. (1979) Proc. Notl. Acad. Sci.

U. S. A. 76, 4350-4354 Zimmerman, R., Sagstetter, M., Lewis, M. J., and Pelham, H. R. B.

(1988) EMBO J. 7, 2875-2880

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from

W P Sheffield, G C Shore and S K Randallpolypeptide folding, aggregation, and import competence.

Mitochondrial precursor protein. Effects of 70-kilodalton heat shock protein on

1990, 265:11069-11076.J. Biol. Chem. 

  http://www.jbc.org/content/265/19/11069Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/265/19/11069.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on May 21, 2018

http://ww

w.jbc.org/

Dow

nloaded from