the journal of vol. 267, no. 18, issue of june 25, … · deletions in the c-terminal region of...

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1992 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 267, No. 18, Issue of June 25, pp. 12860-12667,1992 Printed in U. S. A.

Oligomycin Sensitivity-conferring Protein (OSCP) of Mitochondrial ATP Synthase THE CARBOXYL-TERMINAL REGION OF OSCP IS ESSENTIAL FOR THE RECONSTITUTION OF OLIGOMYCIN-SENSITIVE H+-ATPase*

(Received for publication, January 7, 1992)

Saroj JoshiSOV, A. A. JavedSII, and Leslie C. Gibbs** From the $Department of Cell and Molecular Biology, Boston Biomedical Research Institute, Boston, Massachusetts 02114, the §Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 021 15, the )I Department of ObstetricslGynecology, Division of Reproductive Genetics, Albert Einstein College of Medicine, Kennedy 21 1, Bronx, New York 10461, and the **Department of Biology, Trinity University, Sun Antonio, Texas 78212

Studies to establish the structureJfunction relation- ships of oligomycin sensitivity-conferring protein (OSCP) of mitochondrial ATP synthase were carried out using genetic engineering and biochemical approaches. A full-length cDNA clone encoding OSCP was isolated from a bovine heart cDNA library, and the mature form of OSCP was expressed in Escherichia coli using plasmid expression vector pKP1500. Recom- binant OSCP was found to accumulate in the cytoplasmic inclusion bodies, by virtue of which the recombinant protein could be purified to >85% purity by simple low speed centrifugation of cell lysates. Recom- binant OSCP was found to be indistinguishable from OSCP isolated from mitochondria with respect to (i) apparent molecular mass on sodium dodecyl sulfate gel electrophoresis, (ii) immunological reactivity to anti- OSCP serum, (iii) biological activity in restoring oligomycin-sensitive ATPase and Pi-ATP exchange activities to OSCP-depleted ATP synthase complexes, and (iv) insensitivity of the biological activity to sulfhydryl-directed alkylating reagents. The amino- terminal sequence of the recombinant protein revealed that the initiating methionine was not removed by E. coli, although that apparently did not affect protein folding or its biological activity.

Data on nested deletion mutations starting from the carboxyl terminus in OSCP demonstrated that, in each instance, the mutant form was expressed and the protein product was sequestered in cytoplasmic inclusion bodies, similar to the wild-type form. However, none of the variants, including the one in which only the last 10 residues were deleted, was able to restore cold- stable oligomycin-sensitive ATPase or Pi-ATP exchange activity in OSCP-depleted complexes. Taken together, these data suggest that amino acid residues 181-190 (or some of the residues in this region) in the OSCP sequence may be important for OSCP-F1 interactions.

* This work was supported by United States Public Health Service Grant GM26420 and American Heart Association Grant-in-aid 91014850. 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.

ll To whom correspondence should be addressed Dept. of Cell and Molecular Biology, Boston Biomedical Research Inst., 20 Staniford St., Boston, MA 02114. Tel.: 617-742-2010; Fax: 617-523-6649.

ATP synthase (Fo-F1, H'-ATPase) is a multisubunit membrane-bound enzyme that catalyzes the synthesis of ATP by utilizing the energy of an electrochemical gradient (AFH') that is generated during electron transport (see Refs. 1-3 for recent reviews). It consists of a hydrophilic segment (F1) that contains the catalytic sites for the hydrolysis and synthesis of ATP and a membrane-integrated segment (Fo) that constitutes a transmembrane H+ channel. As a result of investiga- tions in the last 2 decades, the knowledge concerning the structure and function of isolated F1 and Fo segments has significantly improved. However, the understanding of mechanisms underlying the coupling of AGH' to the synthesis of ATP still remains unclear. It is generally accepted that the synthesis of ATP per se takes place with little or no change in free energy. The primary function of AFH+ is to induce a conformational change that, in turn, enables F, to promote the release of newly synthesized ATP (4-8). The key questions that remain to be answered concern how is the energy of A;H' transformed into conformational change and where exactly in ATP synthase does this change occur. The answers to these questions require a knowledge of subunit interactions of the proteins present between the proton-translocating and catalytic segments of ATP synthase. In the absence of a three- dimensional structure of the enzyme, the structure/function studies of the proteins in the binding region should prove useful in yielding the desired information.

Oligomycin sensitivity-conferring protein (OSCP)' is a subunit of mitochondrial ATP synthase and is present in the stalk region between the Fo and F1 segments (9-16). Compar- ative analyses of the primary and predicted secondary structures of OSCP with corresponding structures of subunits from other ATP synthases suggest that OSCP bears partial ho- mologies to the 6 and b subunits of Escherichia coli ATP synthase as well as to the 6 subunit of chloroplast enzyme (17-20). OSCP has no intrinsic catalytic activity and is assayed by its ability to restore ATP synthesis, ATP-driven NAD' reduction by succinate, or oligomycin and dicyclohexylcarbodiimide sensitivity to the ATPase activity of OSCP- depleted F1-bound membrane preparations (21). Reconstitu- tion studies using F1- and OSCP-depleted FO proteoliposomes have demonstrated that OSCP is not needed for passive H' conductance or for blocking H' permeability (22, 23). It also appears that OSCP is not obligatory for binding of F1 to Fo

The abbreviations used are: OSCP, oligomycin sensitivity-conferring protein; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; ECF1, F, portion of E. coli ATP synthase.

12860

Deletions in the C-terminal Region of Mitochondrial OSCP 12861

provided cations are present (18,22,23). Nevertheless, OSCP is crucial for transmitting the energy of AGH+ to the catalytic sector. It has been proposed that OSCP (or OSCP in con- junction with other stalk proteins) could accomplish this by acting as a channel for tunneling protons from Fo to F1 wherein the protonic energy may be converted to conformational change. Alternatively, the transformation of ASH+ into conformation change could take place in FO itself and OSCP may simply be propagating conformational changes between the Fo and F1 segments (20, 24). The detailed molecular mechanisms underlying the function of OSCP are still enig- matic. Conventional biochemical and biophysical approaches have proved inadequate in this respect. We believe that a combination of genetic engineering techniques and a reper- toire of protein biochemical approaches will be necessary for an understanding of the role of OSCP at the molecular level. To this end, we have initiated a program to design genetically altered forms of OSCP, to express them in a heterologous expression system, and to study the mutant forms for their ability (or inability) to reconstitute partial reactions of ATP synthesis in uitro. This approach should enable us to define domains of OSCP of structural/functional importance and to establish the contribution of individual residues to OSCP functions.

In this investigation, we report on establishing a heterologous expression system in E. coli to synthesize the wild-type and mutant forms of OSCP (25). Our data demonstrate that the wild-type recombinant protein is produced in high yield and purifies readily and that the purified protein is indistinguishable from native OSCP with respect to all of the biochemical parameters investigated. Furthermore, our data on nested deletions at the carboxyl terminus of OSCP suggest that amino acid residues 181-190 (or some of the residues in this region) are important for OSCP-F1 interactions.

EXPERIMENTAL PROCEDURES AND RESULTS~

Biological Actiuity of Recombinant OSCP-It is well established that OSCP has no intrinsic catalytic activity. It may be assayed by its ability (i) to bind with defined affinities to each of the Fo, Fl, or Fo-Fl complexes that have been depleted of OSCP; (ii) to restore cold stability to F1-ATPase activity; (iii) to confer oligomycin and dicyclohexylcarbodiimide sensitivity to membrane-bound ATPase activity; or (iv) to stim- ulate energy-linked ATP-driven NAD’ reduction by succinate, NADP+ reduction by NADH, or 32Pi-ATP exchange activity in OSCP-depleted preparations. In this study, we have chosen reconstitution of oligomycin-sensitive ATPase (nonenergy-linked) and 32Pi-ATP exchange (energy-linked) activities as the methods for determining the biological activity of the recombinant protein since these are reported to be the most sensitive assays for OSCP (21).

Reconstitution of Oligomycin-sensitive ATPase-To determine the ability of the bacterially expressed protein to confer oligomycin sensitivity to membrane-bound ATPase, complexes formed of OSCP-depleted membranes and Fl-ATPase were assayed for their ability to catalyze ATP hydrolysis in the presence or absence of oligomycin. Data presented in Fig 6 (left) demonstrate that the ATPase activity of OSCP- depleted Fo-FI was only slightly ((15%) sensitive to oligomycin. However, the oligomycin sensitivity was restored to >85% in complexes that contained either the recombinant

Portions of this paper (including “Experimental Procedures,” part of “Results,” and Figs. 1-5) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal that is available from Waverly Press.

ATPase 32Pi -ATP Exchange 300 r

pg OSCP pg OSCP

FIG. 6. Wild-type form of recombinant OSCP restores oligomycin sensitivity (left) and Pi-ATP exchange (right) activities in OSCP-depleted Fo-FI complexes. For the restoration of oligomycin-sensitive ATPase, 50-pg aliquots of OSCP- and R- depleted F, were reconstituted with 12.5 pg of F,-ATPase and the indicated levels of OSCP and assayed for ATPase activity as described under “Experimental Procedures.” The ATPase activity of reconstituted Fo-FI without the inhibitor is 8.5 pmol min” mg” OSCP- and F1-depleted Fo. To reconstitute Pi-ATP exchange activity, 200-pg aliquots of OSCP- and F1-depleted submitochondrial particles were reconstituted with 50 pg of F1-ATPase and the indicated levels of OSCP and assayed for Pi-ATP exchange activity as described under “Experimental Procedures.” The exchange activity of depleted particles is 8.5 nmol min” mg” in the absence of F,-ATPase and 18.2 nmol min” mg” in its presence.

protein or native OSCP. The titration curve of the expressed protein for restoring oligomycin sensitivity matched well that obtained for authentic OSCP. Fifty percent sensitivity to oligomycin could be conferred by 0.28 pg of the recombinant protein (O), compared to 0.25 pg of OSCP isolated from mitochondria (U). Furthermore, the reconstitution of OSCP- depleted complexes with the bacterially expressed protein could be achieved under conditions similar to those employed for reconstitution with mitochondrial OSCP. These data suggest that the solubilized recombinant protein has the same conformation and presumably similar binding affinities for Fo and F1 as native OSCP.

Reconstitution of 32Pi-ATP Exchange-To determine the ability of the recombinant protein to restore Pi-ATP exchange activity to OSCP-depleted preparations, OSCP-depleted submitochondrial particles were reconstituted with Fl-ATPase and recombinant protein/mitochondrial OSCP according to previously published procedures (23, 25). Data presented in Fig. 6 (right) demonstrate that these particles had low exchange activity even upon supplementation with Fl (<20 nmol min-l mg” OSCP-depleted submitochondrial particles). The activity, however, was stimulated by a factor of at least 10 (265 nmol min” mg-’) upon addition of either the recombinant protein (0) or mitochondrial OSCP (0). The titration curves for the two proteins appeared to be again very similar, indicating similar binding affinities of the two proteins for F, and Fo segments. Furthermore, the reconstituted exchange activity using the recombinant protein was sensitive to oligomycin and uncouplers (10 p~ carbonyl cyanide p-chloro- phenylhydrazone), demonstrating that the stimulated activity is pertinent to oxidative phosphorylation (data not shown).

Role of Cys-118 in Biological Activity of Recombinant Pro- tein-It has been previously reported that the single cysteinyl residue at position 118 in the OSCP sequence can be modified by sulfhydryl-directed reagents without causing a significant decrease in the biological efficiency of the protein (21, 41). Treatment of OSCP with copper o-phenanthroline chelate, however, was reported to lead to dimerization of OSCP as well as to a loss of the ability of OSCP to bind to F1-ATPase (12). These data were interpreted to indicate that Cys-118per se is not necessary for OSCP-F, interactions or for reconstitution of energy-linked functions of OSCP-depleted prepara-

12862 Deletions in the C-terminal Region of Mitochondrial OSCP

tions; the failure of dimeric OSCP to bind to F1 is presumably due to steric hinderance at the binding site rather than to loss of the sulfhydryl group per se. To determine whether or not the same stricture applies to Cys-118 of the recombinant protein, the effect of N-ethylmaleimide and copper o-phenanthroline on the biological activity of the genetically engi- neered protein was investigated. The stimulation of Pi-ATP exchange activity provided by recombinant OSCP was unaf- fected by treatment of the protein with N-ethylmaleimide, but was reduced by 98% following incubation of the protein with the copper chelate (data not shown). It is apparent from these results that Cys-118 of the recombinant protein, similar to that of mitochondrial OSCP, is not essential for the biological activity of the protein.

Deletions in the Carboxyl-terminal End of OSCP-At present, there is no information available either on the three- dimensional structure of OSCP or on the localization of its domains or regions of functional importance. We have macro- mapped the C-terminal region of OSCP by the nested deletion approach to localize putative domains of the protein of structural/functional importance. The rationale for selecting the C-terminal segment is based on the knowledge that (i) the C- terminal part of the functionally analogous subunit of E. coli ATP synthase, namely the b subunit, is implicated in the binding of F1 to the membrane (42, 43); (ii) the C-terminal end is the second highest conserved region of the protein as revealed by comparisons of the primary structures and hydro- pathicity profiles of bovine versus yeast OSCPs; and (iii) the C-terminal end shows the highest concentration of charged residues, a feature that may be of significance.

Construction of Mutant Plasmids and Expression of Trun- cated Forms-Deletions were introduced by replacing the native codon at the site of deletion with translation stop codon TAA using the “overlap extension technique” involving the polymerase chain reaction (32). DNA sequencing of mutant plasmids indicated that the frequency of mutagenesis varied from 25 to 50%.

To induce protein expression, aliquots of mutant plasmids were used to transform bacterial cell strain KP3998. Trans- formants with the correct orientation of the OSCP gene were induced with isopropyl-1-thio-P-D-galactopyranoside to initi- ate translation as described earlier for the wild-type form. Growth curves of mutant plasmid derivatives of KP3998 suggest that the growth of host cells was not affected as a consequence of truncations in the OSCP gene. SDS-PAGE and Western blot analyses of total cell homogenates as well as subcellular fractions demonstrate that the recombinant protein was expressed in each of the variants and that the expression level was in the same range as in the wild-type form (Fig. 7, lanes 2-6). The mobility of truncated forms of OSCP on SDS-PAGE was consistent with the size predicted on the basis of the total number of amino acid residues left in the protein after the indicated deletions (Fig. 7). Further- more, the Western blot data also demonstrate that, despite a deletion of up to 40 amino acid residues out of a total of 190, the truncated forms of OSCP could still be recognized by an antiserum that was raised against an untruncated and a nondenatured form of OSCP (Fig. 7).

Data to establish the subcellular location of the mutant forms demonstrated that truncated OSCPs, similar to the wild-type form, were also sequestered in cytoplasmic inclusion bodies (Fig. 7, compare lanes 2-6 with 8-11 ). These data suggest that the deletion mutants of OSCP resemble the wild- type form in their expression characteristics as well as in immunoreactivity to anti-OSCP serum.

Biological Activity of Truncated Forms of OSCP-To test

. , . ,...., ,.>”... ”. ,.

95.5K -

55.OK - 43.OK - 36.OK - 29.OK -

18.4K -

12.4K - 1 2 3 4 5 6 7 8 9 1 0 1 1

FIG. 7. Truncated forms of OSCP are sequestered in cytoplasmic inclusion bodies. Bacterial culture suspensions containing the wild-type or truncated forms of recombinant OSCP were subjected to lysis by French press, followed by centrifugation of cell lysates a t 5000 X g for 10 min, to separate the fraction containing inclusion bodies from the remaining proteins. Aliquots of various sediment and supernatant fractions were subjected to SDS-PAGE, followed by Western analysis. Lane 1, mitochondrial OSCP; lanes 2-6, the sediment fraction after centrifugation of cell lysates of KP3998 OSCPl (wild type), KP3998 OSCP2 (mutant OSCP(181)), KP3998 OSCP3 (mutant OSCP(171)), KP3998 OSCP4 (mutant OSCP(161)), and KP3998 OSCP5 (mutant OSCP(151)), respectively; lanes 7-11, supernatant fractions after centrifugation of lysates from KP3998 OSCPl to KP3998 OSCP5, respectively. Please note that the faint band migrating with a relative mobility corresponding to 19,000 Da in lanes 3 and 7-11 is due to reactivity of a host cell protein to anti- OSCP serum.

TABLE I Truncations in the C-terminal end lead to inability of mutant OSCPs

to restore 32Pi-ATP exchange activity Submitochondrial particles that were depleted of OSCP and F,-

ATPase (AE-P) were reconstituted with F, and variant forms of OSCP and assayed for their ability to catalyze 32Pi-ATP exchange activity as described under “Experimental Procedures.”

32Pi-ATP exchange

AE-P + F1 18 AE-P + F, + mitochondrial OSCP 163 AE-P + F, + WT” OSCP(191)’ 166 AE-P + F, + mutant OCSP(181) 27 AE-P + F1 + mutant OSCP(171) 19 AE-P + F, + mutant OSCP(l61) 22 AE-P + F1 + mutant OSCP(151) 21

a WT, wild type. ’ Numbers in parentheses refer to the deletion sites in the amino acid sequence of OSCP.

the biological activity of deletion mutants, aliquots of the recombinant forms were allowed to reconstitute with mitochondrial preparations depleted of OSCP and were assayed for their ability to restore 32Pi-ATP exchange and cold-stable oligomycin-sensitive ATPase activities.

Data presented in Table I demonstrate that the wild-type form of recombinant OSCP enhanced the Pi-ATP exchange activity of OSCP-depleted submitochondrial particles to the same extent as authentic OSCP (compare line 3 with line 2). However, none of the mutant forms, including OSCP(181), in which only the last 10 amino acid residues were deleted, showed any significant stimulation in the exchange activity of depleted particles (compare lines 4-7 with line 1).

Data presented in Table I1 demonstrate that, whereas the wild-type form of OSCP restored the oligomycin sensitivity of the ATPase activity of OSCP-depleted complexes to 95%


TABLE I1 Truncations in the C-terminal end lead to inability of deletion mutants of OSCP to reconstitute oligomycin-sensitive ATPase

A membrane fraction of ATP synthase that was depleted of Fl- ATPase and OSCP (UFO) was reconstituted with F1-ATPase and the indicated form of OSCP as described under “Experimental Proce- dures.” Five-microgram aliquots of the reconstituted complexes were next assayed for their ability to catalyze ATP hydrolysis in the absence or presence of oligomycin (23, 35).

ATP hydrolyzed ~ l i ~ ~ ~ ~ ~ i ~

-Oligomycin +Oligomycin sensitivity

Gmol/min/mg UFO % UFO + FI 4.38 3.50 20 UFO + F, + WT” OSCP(191)* 4.65 0.23 95 UFo + Fl + mutant OSCP(181) 4.41 2.99 32 UFO + Fl + mutant OSCP(171) 4.35 2.91 33 UFo + F1 + mutant OSCP( 161) 4.80 3.36 30 UFO + F, + mutant OSCP(151) 4.60 3.31 28

a WT, wild type. * Numbers in parentheses refer to the deletion sites in the amino

acid sequence of OSCP.

(compare line 2 with line l), the deletion mutants showed no significant enhancement in oligomycin sensitivity over the depleted control (compare lines 3-6 with line 1). This suggests an impairment in the interactions of OSCP with F1, Fo, or Fo- F1 complexes.

To determine the site of defect in the interactions of OSCP with OSCP-depleted Fo-FI complexes, we assayed the ability of deletion mutants of OSCP to confer cold stability to F1- ATPase. Our data demonstrate that the ATPase activity of complexes reconstituted with deletion mutants of OSCP was not stable to low temperature, unlike that of the complexes reconstituted with the wild-type form. Taken together, these data suggest an impairment in the binding of OSCP to F1 due to truncations in OSCP. Since deletion of residues 181-190 itself leads to inactivation of OSCP, it is not possible to assess the role of residues 151-180 from these experiments. It is clear, however, that amino acids residues 181-190 in the OSCP sequence (or some of the residues in this segment) are crucial for OSCP-F1 interactions.

DISCUSSION

OSCP is a subunit of mitochondrial ATP synthase that is crucial for coupling the energy of AbH’ to the synthesis of ATP. We have established a heterologous expression system in E. coli to synthesize large quantities of recombinant forms of OSCP as part of our goal to elucidate structure/function correlations of OSCP. The preference for a heterologous expression system is (i) to obtain quantities of protein that would be sufficient for crystallization and structure/function analyses and (ii) to obtain protein expression in a manner that is independent of the biosynthesis of other subunits of ATP synthase. This is to ensure that the effects of the mutations will be restricted to the expression, folding, and biological activity of the mutated protein only, and not extend to the synthesis, assembly, and biological activity of other subunits of ATP synthase, thereby minimizing complications in the interpretation of data.

The second part of our investigation deals with performing nested deletions in OSCP with the intention of localizing putative structural/functional domains of this protein.

OSCP Is Expressed in E. coli in Biologically Active Form- To express the mature form of OSCP, the presequence was cleaved, and the remaining coding sequence was cloned into expression plasmid pKP1500 (Fig. 1). The rationale for choos- ing pKP1500 as the expression vector is based on the knowl-

edge that this plasmid allows the synthesis of foreign proteins in an unfused form. In addition, pKP1500 has the same temperature-dependent copy control as plasmid pUC19, a feature that could be highly useful in the event that the recombinant protein (or some of the mutant forms) turns out to be toxic to the host cells (25).

The data from SDS-PAGE and Western blot analyses of positive recombinants demonstrate that a protein, with the same apparent molecular mass and antigenicity as mitochondrial OSCP, is expressed at a high yield in an undegraded form (Fig. 3, lanes 4-8). Furthermore, the analyses of subcellular fractions of cells that show expression of OSCP revealed that the recombinant protein is sequestered in intracellular inclusion bodies (Fig. 4). This phenomenon has proven to be advantageous to us from two standpoints. First, since native OSCP is known to be sensitive to proteolytic degradation, the recombinant protein, by virtue of its accumulation in inclusion bodies, was protected from any proteases that might be present in the host cell cytoplasm. Second,, since the major purification step involving separation of recombinant OSCP from host cell proteins requires only low speed centrifugation of cell lysates, the recombinant protein could be purified faster in comparison to mitochondrial OSCP.

Our data on the biological activity of the recombinant protein using an in vitro reconstitution system demonstrated that the expressed protein is indistinguishable from mitochondrial OSCP with respect to its ability to restore to defined OSCP-depleted preparations both nonenergy as well as energy-linked functions (Fig. 6). Furthermore, the activity restored by the recombinant protein, similar to that reconstituted by mitochondrial OSCP, is insensitive to alkylating reagents, but is completely inhibited by pretreatment of the protein with a sulfhydryl oxidizing reagent (Table I). It follows from the above that the observed restoration of mitochondrial functions in OSCP-depleted preparations (or lack of restoration in the event of using oxidized protein) is due directly to the recombinant protein. The amino acid sequence analysis of the first 20 residues revealed that the sequence is indistinguishable from that of native OSCP except for having an additional methionine at the N terminus (Fig. 5). The additional methionine, however, does not seem to interfere in the folding of the protein or in its binding to F1 or Fo assemblies as judged from data to determine the biological activity of the recombinant protein.

Taken together, our data allow us to conclude that the recombinant protein expressed in E. coli represents OSCP and that the gene encoding it represents the OSCP gene. To the best of our knowledge, this constitutes the first evidence demonstrating a direct relationship among the OSCP gene, its protein product, and the involvement of the protein product in oxidative phosphorylation. In this connection, it may be mentioned that an ATP5 gene, presumably encoding yeast OSCP, has been recently cloned by Uh et al. (13). It has been proposed that the ATP5 gene product is a subunit of ATP synthase since disruption in this gene led to absence of oligomycin-sensitive ATPase. However, it is not clear from these studies whether ATPase subunits 6 and 9, which are known to be just as essential for inhibitor-sensitive ATPase as OSCP, were expressed and assembled in those mutants in which the ATP5 gene was disrupted. The claims of the authors regarding the functional relationship of the ATP5 gene product to ATP synthase should therefore be considered tentative pending a demonstration of the reconstitution of ATP synthesis in OSCP-depleted synthase by the purified ATP5 gene product under in vitro conditions.

Carboxyl-terminal Region Is Crucial for Interactions of

12864 Deletions in the C-terminal Rt

OSCP with Subunits of ATP Synthuse-Data on nested deletions to remove 10,20, 30, or 40 amino acid residues from the carboxyl terminus of recombinant OSCP demonstrated that, in each instance, the protein is expressed and the product is indistinguishable from the wild-type form with respect to yield, stability, subcellular location, and reactivity to an an- tibody that was raised against the nondenatured form of OSCP. Thus, the truncated forms of OSCP do not appear to be structurally destabilized or misfolded as judged immuno- logically or by their overall expression characteristics. Data on the determination of the biological activity of the truncated forms, however, revealed that none of the mutant forms is able to reconstitute Pi-ATP exchange, oligomycin-sensitive ATPase, or cold-stable ATPase activity to OSCP-depleted ATP synthase. A failure to confer cold stability to F1-ATPase clearly indicates an impairment in OSCP-F1 interactions. Taken together, these data suggest that the protein sequence harboring amino acid residues 181-190 (or some of the residues in this stretch) in OSCP may be important for OSCP- F1 interactions.

A similar conclusion was reached regarding the function of the carboxyl-terminal region of the b subunit of E. coli ATP synthase, which is reported to be weakly homologous to OSCP (42, 43). Incidentally, Mendel-Hartvig and Capaldi (44) have recently reported that limited trypsin treatment of ECFl leads to the proteolytic degradation of the C-terminal end of a 6 subunit, which is also considered to be weakly homologous and functionally similar to OSCP. The treated ECFl could bind to ECF1-stripped bacterial membranes, but was unable to reconstitute dicyclohexylcarbodiimide-sensitive ATPase. Apparently, the loss of ability to restore dicyclohexylcarbodiimide sensitivity to depleted membranes is due to loss of -20 amino acids from the C terminus of the 6 subunit, which would be in agreement with our present findings on the C- terminal end of OSCP.

A comparison of the primary structures of bovine and putative yeast OSCPs toward the C-terminal end reveals that out of 10 amino acid residues examined, 2 are identical and 5 are homologous to the ones in corresponding positions in the proteins from these two sources. Experiments to precisely define the contribution of each of these residues both to the binding of OSCP to Fl or Fo complexes and to energy coupling are being pursued.

Acknowledgments-We are grateful to Drs. D. Rao Sanadi, L. Kantham, and A. Phelps for critical reading of the manuscript, Mary Kenneally and Jing Shao for expert technical assistance, and Angela DiPerri for excellent secretarial help.

1.

2. 3.

4. 5. 6.

7.

8. 9.

REFERENCES Schneider, E., and Altendorf, K. (1987) Microbiol. Rev. 61, 447-

Hatefi, Y. (1985) Annu. Rev. Biochem. 64,1015-1069 Senior, A. E. (1990) Annu. Rev. Biophys. Biophys. Chem. 19,77-

Boyer, P. D. (1987) Biochemistry 26, 8503-8508 Penefsky, H. S. (1987) Biochem. SOC. Trans. 16,97-99 Penefsky, H. S. (1985) Proc. Natl. Acad. Sci. U. 5'. A. 82, 1589-

Cross, R. L., Grubmeyer, C., and Penefsky, H. S. (1982) J. Biol.

Boyer, P. D. (1989) FASEB J. 3, 2164-2178 McLennan, D. H., and Tzagoloff, A. (1968) Biochemistry 7,1603-

497

141

1593

Chem. 267, 12101-12105

1610

g w n of Mitochondrial OSCP 10. Van de Stadt, R. J., Kraaipoel, R. J., and Van Dam, K. (1972)

11. Kagawa, Y., and Racker, E. (1966) J. Biol. Chem. 241, 2461-

12. Liang, A. M., and Fisher, R. J. (1983) J. Biol. Chem. 268, 4788-

13. Uh, M., Jones, D., and Mueller, D. M. (1990) J. Bwl. Chem. 266,

14. Horak, A., Horak, H., Dunbar, B., Fothergille, J. E., and Wilson,

15. Kimura, T., Takeda, S., Asahi, T., and Nakamura, K. (1990) J.

16. Misook, L., Jones, D., and Mueller, D. M. (1988) Nucleic Acids

17. Walker, J. E., Saraste, M., and Gay, N. J. (1982) Nature 298,

18. Sandri, G., Wojtczak, L., and Ernster, L. (1985) Arch. Biochem.

19. Berzborn, R. J., Finke, W., and Otto, J. (1987) 2. Naturforsch.

20. Engelbrecht, S., and Junge, W. (1990) Biochim. Biophys. Acta

21. Senior, A. E. (1971) Bioenergetics 2, 141-150 22. Joshi, S., Pringle, M. J., and Siber, R. (1986) J. Bwl. Chem. 261,

23. Pringle, M. J., Kenneally, M. K., and Joshi, S. (1990) J. Biol.

24. Pedersen, P. L., and Carafoli, E. (1987) Trends Biochem. Sci. 12,

25. Miki, T., Yasukochi, T., Nagatani, H., Furuno, M., Orita, T., Yamada M., Imoto, T., and Horiuchi, T. (1987) Protein Eng.

26. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162,

27. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

28. Wood, W. I., Gutschier, J., Lasky, L. A., and Lanon, R. M. (1985) Proc. Natl. Acad. Sci. U. S. A. 82,1585-1588

29. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Smith, J. A., and Struhl, K. (eds) (1987) Current Protocols in Molecular Biology, Greene Publishing Associates/Wiley-Interscience,

Biochim. Biophys. Acta 267, 25-36

2466

4793

19047-19052

S. B. (1989) Biochem. J. 263, 301-304

Biol. Chem. 266,6079-6085

Res. 16, 8181

867-869

Biophys. 239, 597-602

Sect. C Biosci. 42, 1231-1238

1016,379-390

10653-10658

Chem. 266, 7632-7637

86-89

1,327-332

156-159

New York

Acad. Sci. U. S. A . 74,5463-5467 30. Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl.

31. Javed, A. A,, and Joshi, S. (1990) Biotechniques 9, 28-32 32. Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K., and Pease,

L. R. (1989) Gene (Amst.) 77, 51-59 33. Joshi, S., and Sanadi, D. R. (1979) Methods Enzymol. 66, 384-

391 34. Horstman, L. L., and Racker, E. (1970) J. Biol. Chem. 246,1336-

1344 35. Joshi, S., Hughes, J. B., Torok, K., and Sanadi, D. R. (1985)

Membr. Biochem. 6,309-325 36. Joshi, S., and Burrows, R. (1990) J. Biol. Chem. 266, 14518-

14525 37. Ovchinnikov, Y. A., Modyanov, N. N., Grinkevich, V. A., Alda-

nova, N. A., Trubetskaya, 0. E., Nazimov, I. V., Hundal, T., and Ernster, L. (1984) FEBS Lett. 166, 19-22

38. Walker, J. E., Gray, N. J., Powell, S. J., Kosting, M., and Dyer, M. R. (1987) Biochemistry 26,8613-8619

39. Xu, G.-Q., and Hitchcock-DeGregori, S. E. (1988) J. Biol. Chem. 263,13962-13969

40. Martson, F. A. 0. (1986) Biochem. J. 240, 1-12 41. Dupuis, A., Issartel, J.-P., Lunardi, J., Satre, M., and Vignais, P.

42. Takeyama, M., Noumi, T., Maeda, M., and Futai, M. (1988) J.

43. Steffens, K., Schneider, E., Deckers-Hebestreit, G., and Alten-

44. Mendel-Hartvig, J., and Capaldi R. A. (1991) Biochim. Biophys.

V. (1985) Biochemistry 24, 728-733

Biol. Chem. 263,16106-16112

dorf, K. (1987) J. Biol. Chem. 262,5866-5869

Acta 1060, 115-124


EXPERIMWTAL PROCEDURES DNA polyme- I Wenow fragmenl) and Squenaw version I were dtained fmm US Biochemicals (Cleveland. Ohia); all other DNA modifying enzymes. remidon cndonucleawr. and bacteriophages M13-mp18 and -mp19 were cbtaind from New England Biolabr (Beverly, MA) and were used according to manufacturer's protocols. [gamma -fZPldeoryATP and [alpha -%]deoxyATP were purchased from ICN Radiochemicalr. Inc. (Irvlnc. CA) and New England Nuclear (Boston, MA) respectively, d m r from P h m a c i a LKB Biolechnology Inc. (Pmataway, NI). agarose from FMC BioproduclP (Rockland. ME), and acrylamlde gel reagents from

BmRad (Richmond, CA). All other chemicals were from Sigma (St. Louis, Mo) 01 I.T. Baker Chemical Co. (Phdliprburg, NI). The cDNA library from bovine hcaR l i m e was P generous gifl of Dr. Hanmul Wohlrab

gifts of Dr. T. Miki (Faculty of Pharmaceutical Sclencea, Kyurhu University). (Boston Biomedical Rewarch Instilute) and the expression plasmid pKPl500 and hoal cell strain KP3998 were

Isoldon of RNA. Norlhem Blor Andysis, Cloning and Sequencing - Toml ecllular RNA was extracted from

bowne hurl tissue by the guanidinium imthiocyanate extractm pmcedure [261. Poly(A)+ RNA was enriched by ol~ga(d~-cel lulov chmmatognphy 127. For Nonhern blot analysis. 50 pg of total RNA or 5 pg of Paly(A)+ RNA were eleclmphoresed on 1.2% agarow formaldehyde gels and tranrferred on to Genercreen membranes by capillary diffusion according to the manufacturer'r instructions (New England Nuclear, Boslon. MA). The resulting blots were hybridized wlth %abeled mixed oligcdeorynucleottdc prober and processed ~n

accordance wilh a previously published pmcedure 1281.

For tmlatrng putalive OSCP clones, transformed cells from a plasmid pBR322 bovine hmrl cDNA library were plated at a denrrly of I , M o colonies pcr I50 mm diameler petri plate and procewd as described [27.29]. The recornbinanlr were screened by colony hybridrzarion using 32P.labeled oligcdeoxynuclmtide pmber. To ertablrrh DNA wqucncc, <DNA mwns from porltive recombinants were excised by digerrion with &HI, and rubcloned inlo bacteriophages M13~mp18 and -mp 19. DNA wqucnccr of the MI3 clones were determined by

which were inilidly "red for screening the eDNA library 1311.

Ihe dideoxy chain-lerminauon method 1301 using univerral prtmer as well as mired oligcdmrynucleolide prober

Conanrction of p K O S C P - W -AI d e w i b d in the results wclion, the 732 bp ElPRI fragment of a porillvc rcmmbmanl clone harbors 4-735 bp fragmcnl of OSCP cDNA which tncludcr full wquencc for lhe eading

Therefom, to crprenr only the malure form of OSCP the 4-735 bp EfnN fragment war first lreated wtth &I

region of OSCP precursor as well as mort of the wqucnce for L e vnlranrlald regions of OSCP gene.

Thhe larger of the two fragments. comrponding Lo 237-735 bp. was gel purified and ligaled lo P 106 bp long synthetic linker DNA m as to yield a 599 bp malure OSCP cDNA W ~ S ~ N C ~ (left pan of Fig.1) (39). The 106 bp synthetic linker contained m L e order indlcaled an ErpRl site for ease in cloning, an ATG codon for initiation of translalion. and a wquence mmrponding lo firs1 78 nucleohdes of OSCP EDNA Encoding malure OSCP, including the &I lie. Each strand of the synlhetie linker was, in turn. conrlrueted by ligating three aligcdeorynuelmlidcr ranging from 26-40 dmrynuclmtldes. The 599 bp OSCP cDNA constmct conbalnmg the coding sequence for only the mature form was cloned inLo the EfnRl sile of plasmid pKPl500 to yield OSCP expression vector pKOSCP-W. The plasmid DNA used for ligation had been piclreated with U R I (lo linearize) and calf alkaline phosphatase (Io dephosphorylate) as shown in Fig.1. The presence of OSCP cDNA inrcrl and its orientation with r e r p z l 10 promoter in the Anal plasmid ~ o n l l r u ~ t was erlabhrhed by rewct ion

analysiP using U l l l .

Conamction of nrornbimnr plnsnids pKOSCPlSIlr, pKOSCP16lr. pKOSCPI7lr and pKOSCPISlr . Recombinanl plasmids carrying delelmnr in the earbxyl lcrminai region in OSCP gene were connnrcled by rcplacmg thc mlwc codon at the sile of deletion by lrandalional slop codon TAA. Thn was accomphshed by applying Ihe 'Overlap Extensmn Technique' which inwlveS the generatlo" of two m u m l fragments from the large1 DNA. followed by exlension of their overlapping ends w n g Taq polymerase [32]. In the present

amplificalion. In rlep I. mum1 fragments were generated in lwo pardlel PCRs. each urmg I 0 0 ng DNA investigation a segment of plasmid pKOSCPWT DNA hvboring OSCP gene was largcted for mutagenerlr and

lemplalc (pKOSCP-WT). 0.1 nmol of IlanLing primer and 0.1 m o l of mutan1 primer (Fig.2). For construcllng plasmid pKOSCP18le, the m u m l pnmerr used were M3 and M3C. Each PCP. war cam& oul in a final

volume of 100 pl eontaming IO mM Tris.HCI, pH 8.3 (25'C). 50 mM KCI, 0.001% (wlv) gelatin, 1.5 mM MgC12, 200 pM dNTP, and 2.5 units Taq Polymerav m addition lo DNA templale and primers. Samples were

overlaid wth 1 0 0 pl of mineral 011 and rubjecled lo 30 cycles of amplification (denaturation - I min, 95OC;

annealing - 2 min, 37%; exlension - 3 min, 72% followed by a Anal 12 min exlension MClion at 7Z°C using a Coy TcmpCyder. The mum1 fragmenlr were gcl punfied to remove cxccs~ primers. and DNA from gel bands of appmpriatc size war elulcd using Geneclean I. Smce the mutant pnmen were designed lo be overlapping. the mutant fragments generaled by PCR also turned oul 10 be overlapping. In order 10 extend the overlapping ends of mutant DNA fragments, IO ng amounts of the latter were mixed with flanking primers pK3 and pKlC ana

fubiected lo another PCR consisting of 30 cycles of amplification and one final cycle of I2 min extension.

i 013

~~

tig.2. LXagmmmuic rkurh Io UIVsrmr rephernrN of K l B l codon by rmnskuiod sop rodon by h e

&e+ E u e # d o r Technique - R a t i o n I q u i d flanking primer pK3 5'. CTCAGGAGAGCGTKACCGACAAA-3' and mutant primer M3C S'- ACCAAGATTCAGTMCIGAGCAGA-3', while Raclion 2 required flanking primer pKlC 5 ' - GCM~T~C~ACAATTAATCATCGGCTCGTATA-~' and mum1 p r i m M3 5'. TCMjCICAGlp&TGAATCITOOT-3'. Thhe site of mulagenesir is indicated by underlining in the mum1 primers. The dcWs for PCR d o n s me described under srprimenW procedures.

12866 Deletions in the C-terminal Region of Mitochondrial OSCP

SDS-PAGE Western Blot

Deletions in the C-terminal Region of Mitochondrial OSCP

SDS-PAGE Wesiern Bloi

12867

the journal of vol. 267, no. 18, issue of june 25, … · deletions in the c-terminal region of...

Documents