Structure of the yeast mitochondrial adenosine triphosphatase ...

2 downloads 0 Views 2MB Size Report
sensitivity of subunit 3 by a factor of 2 and increased the sensitivity of a particular trypsin site (or group of sites) on subunit 1 by ?-fold. The overall degradation of.
THEJOURNAL OF BIOLOGICAL CHEMISTRK Vol. 256, No. 13, Issue of July 10, p. 6990-6994, 1981 Printed in U.S.A.

Structure of the Yeast Mitochondrial Adenosine Triphosphatase RESULTS OF' TRYPSIN DEGRADATION* (Received for publication, September 24, 1980, and in revised form, March 2, 1981)

Richard D. Todd and MichaelG. Douglas$ From theDepartment of Biochemistry, University of Texas Health Science Center, Sun Antonio, Texas 78284

MATERIALSANDMETHODS A comparison was made of the subunit sensitivities of the F1-ATPaseand the Triton-solubilized ATPase Trypsin and soybean trypsin inhibitor were purchased from Calcomplex to trypsin degradation. "he dissociation of the biochem (San Diego, CA). F1-ATPase from ATPase complex increased the trypsinF1-ATPase and ATPase complex were isolated as described previsensitivity of subunit 3 by a factor of 2 and increased ously (14)from commercially prepared yeast. F1-ATPaseand ATPase the sensitivity of a particular trypsinsite (or group of complexwere treated with varying amounts of trypsin in MTEA sites) on subunit 1 by ?-fold.The overall degradationof buffer, pH 8.0 (10%methanol, 30 mM Tris-SOa, 1 mM EDTA, 2 mM ATP, pH 8.0), for 30 min at 26 "C. After 30 min, a 4- to 40-fold excess subunits 1 and 2 appears to be the same in solubilized of soybean trypsin inhibitor was added (w/w, inhibitoritrypsin) as a ATPase complex and the F1-ATPase. Implications of 10 m g / d solution in MTEA, pH 8.0, and the samples were put on these findings for structuralmodelsof the ATPase ice. Samples were then electrophoresed and stained with Coomassie complex are discussed. blue as described previously (15).Gels were scanned with a Quickscan

The oligomycin-sensitive mitochondrial ATPase complex' of yeast consists of 10 to 12 nonidentical types of protein subunits in 1 to 3 copies (1-3). As in other systems, the complex can be dissociated into a catalytically active-water soluble portion, termed the F1-ATPase, and amembraneassociated noncatalytic portion, sometimes termed Fo (1,3,4). The Triton-solubilized ATPase complex behaves as anasymmetrical structure with a molecular weight of about 5.8 x lo5, containing at least 20 polypeptide chains (3). The dissociated F1-ATPase consists of 7 to 10 polypeptide chains (depending on isolation conditions; Refs. 1, 3, 5, and 6), has a hexagonal appearance as viewed by electron microscopy (1, 4), and has a molecular weight of about 3.5 X lo5 (1, 3, 5 ) . There have been many proposals for the structure of FIATPases from different sources (see Ref. 7 for a review). An accompanying paper presents a model for the Triton-solubilized yeastATPase complex (8). For the most part such models have been based on the results of protein cross-linking experiments (9-11) and electron microscopic images (1, 2, 12, 13). The present study is an attempt to apply a different approach, trypsin degradation, to the studyof ATPase complex structure. The results place constraints on possible FIATPase models and arecompatible with the ATPase complex model derived from subunit cross-linking studies (8).

* This investigation was supported by Grant GM25648-02 from the National Institutes of Health, by Biomedical Research Support Grant RR 05654 from the National Institutes of Health to the University of Texas Health Science Center at SanAntonio, and by Grant AQ-814 from the Robert A. Welch Foundation. 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. f To whom reprint requests should be addressed. ' The abbreviations used are: ATPase complex, the Triton-solubilized oligomycin-sensitive complex presumably representing the enzyme as it is found in mitochondria; F1-ATPase, the water-soluble catalytically active portion of the enzyme; Fo-ATPase, the membraneassociated portion of the enzyme.

(Helena Industries, Arlington, TX). The areas under the curves of the densitometric scans were proportional to the amount of protein loaded/l-cm gel slot for the following values: subunits 1, 2, and 3, between 5 and 60 pg; subunits 4, 5, and 6, between 10 and 60 pg. All of the areas were corrected for minor variations in slot width. Repeat determinations for single samples agreed within 10%. Incubation of samples with both trypsin and a 4-foldexcess of soybean trypsin inhibitor (w/w) for 60 min at 26 "C resulted in no detectable proteolytic action. Immunoprecipitation of trypsin-treated ATPase complex was as described previously (3, 14). In control experiments, trypsin sensitivities as displayed on gels were the same when ATPase samples were and were not immunoprecipitated prior to gel electrophoresis. ATPase and protein assays were performed as previously published (14). RESULTS

The present study has used the accessibility of a given subunit to the action of trypsin as a measure of its exposure to the surrounding environment. Since we know little about the inherentdigestibility of the ATPase subunitsdirect comparisons between different subunits cannot be made. However, comparisons can be made between the same subunit in different environments. Differences in trypsin accessibility between environments may reflect direct exposure of trypsin-sensitive sites by the removal of previously shielding subunits or may be secondary to conformational changes within a given subunit. The present study has made comparisons of the accessibility of ATPase subunits in the ATPase complex and the F,-ATPase. Accessibility to trypsin has been determined by gel electrophoresis, enzymatic activity, and inhibitor action for the various subunits. The trypsin degradat.ion patterns of the Triton-solubilized ATPase complex are shown in Fig. 1. In this particular experiment, the products of trypsin degradation were immunoprecipitated with ATPase complex antiserum prior to electrophoresis. Identical resultsare obtained if the immunoprecipitation is omitted. Subunits 7, 8, and 9 were not adequately resolved for quantitation by Coomassie blue staining but, as will be described later, the sensitivities of the oligomycin sensitivity conferring and endogenous ATPase inhibitor subunits (subunits 8 and 9 in yeast) were estimated by measuring ATPase activity and oligomycin inhibition of ATPase activity following trypsin treatment. As the trypsin to substrate ratio was increased, the percentage of subunits 1 to 6 remaining de-

6990

6991

ATPase Structure WEL B 100

l I

I

A B C D E

Jh'

EMIGRATION

w

FIG.1. Trypsin degradation of solubilized ATPasecomplex. ATPase complex in MTEA, pH 8.0, was degraded for 30 min at 26 "C with trypsin at the indicated relative concentrations (w/w). The reactions were terminated by cooling on iceand adding a 4- to 40-fold excess of soybean trypsin inhibitors (w/w uersus trypsin). Samples were immunoprecipitated with anti-ATPase serum, as described in Refs. 3 and 14, and 50-pg aliquots were electrophoresed on a 15-cm 10 to 17%acrylamide, 0.1%sodium dodecyl sulfate gradient slab gel. A, densitometric scans of the electrophoretic patterns of the photograph in B. B, photograph of the gel electrophoretic separation of the

trypsin-degraded ATPase subunits, stained with Ccomassie blue. I to 6, subunit numbers. la is the major immunoprecipitable trypsin product of subunit 1. A to E , trypsin to substrate ratios of 0, 0.0315, 0.0625,0.315, and 0.630, respectively (w/w). Thegraph is a plot of the immunoprecipitable subunits remaining after trypsin degradation at subunit 2; the above relative concentrations: 0, subunits 1 + la; 0, A, subunit 3; 0, subunit 4; subunit 5; A, subunit 6. - - -, extrapolations to complete degradation of a particular subunit from sensitivities at low trypsin to substrate ratios.

creased. The degree of subunit degradation at a given trypsin to subunit ratio varied with each subunit but fell into three general classes withsubunit 4 >subunits 3,5,and 6 >subunits 1 and 2 (relative sensitivities, 7:3:1, respectively, for the three classes). As discussed previously,no conclusions canbe made from the relative sensitivities of the subunits since the inherent trypsin digestability of the individual subunits is not known. Trypsin treatment of the ATPase complex also results in the production of an immunoprecipitableproduct of subunit 1 (best seen in A, Fig. 1, labeled l a ) which exhibits a slightly greater mobility on the gel. In determining the percentage of subunit 1 remaining, the sum of 1 + l a was taken. As will be shown below,the production of l a is greatly different between ATPase complex and FI-ATPase. It can be determined from Fig. 1B that theratio of the percentages of subunits remaining for subunits 1 and 2 (subunit 1 + la/subunit 2) is approximately 1 at all trypsin concentrations tested. Fig. 2 compares the trypsin sensitivities of subunits 1,2, and 3 in the FI-ATPase (data points)and the ATPase complex (solid lines, from data of Fig. 1).At low trypsin to substrate ratios, subunits 1 and 2 are degraded slightly faster in the F1ATPase while subunit 3 is degraded substantially faster. At ratios of0.2 or greater, the subunit 1 and 2 curves are the same for the FI-ATPase and the ATPase complex. The increased sensitivities of subunits 1 and 2 in the FI-ATPase at low trypsin to substrate levels isprobably due to theabsence of the more trypsin sensitive lower molecular weight subunits of the ATPase complex. In the ATPase complex subunits 4, 5, and 6 are 3 to 7 times more sensitive to trypsin than subunits 1and 2. The increased sensitivities of subunits 1 and 2 in the F,-ATPase, then, probably do not reflect structural differences betweenthe F1-ATPase and the ATPase complex, per se, but represent a change in the average trypsin substrate sensitivity. This idea is supported by comparing the subunit 1 to subunit 2 ratios at different trypsin concentrations between the FI-ATPase and the ATPase complex. Fig. 3 demonstrates that the subunit 1 to subunit 2 ratio is the same for

the FI-ATPase and the ATPase complex at any given trypsin to substrate ratio: the data pointsare for the FI-ATPase and the solid line is for the ATPase complex (data from Fig. 1). It is concluded that, with respect to the overall trypsin degradation patterns, subunits 1 and2 are in similar environments in the FI-ATPaseand the ATPase complex. Subunit 3 is much moretrypsin-sensitive in the F1-ATPase and is completely degraded at a trypsin to substrate ratio where approximately 60% ofthe ATPase complex subunit 3 is still remaining (Fig. 2). This increase in trypsin sensitivity cannot be explained by a change in the trypsin substrate pool and represents a structural or conformational difference between the F,-ATPase and the ATPase complex. The dissociation of the ATPase complex to form the FI-ATPaseincreases the exposure of subunit 3 to theaction to trypsin. Table I summarizes the relative susceptibilities of subunits 1, 2, and 3 to trypsin in both the FI-ATPaseand the ATPase complexes. Linear extrapolation of the decrease in a given subunit with increasing trypsin at low trypsin to substrate ratios gives the indicated sensitivities. Subunits 1 and 2 are approximately 3-fold more sensitive in the FI-ATPase while subunit 3 is about 6 times more sensitive. If, as discussed above, the increased sensitivities of subunits 1 and 2 do not reflect real differences betweenthe FI-ATPase and the ATPase complex, then the sensitivity difference of subunit 3 is reduced to about 2-fold. It is concluded that subunit 3 is at least twice as exposed to trypsin in the F1-ATPaseas in the ATPase complex. It has previously been reported that limited protease degradation of the FI-ATPase from the bacteria Streptococcus faecalis (16),Escherichia coli (17, 18),Mycobacterium phlei (19), and Micrococcus lysodeikticus (20) results in the conversion of subunit 1 (termed a in another nomenclature) to a slightly more rapidly migrating form on gel electrophoresis. We observe an analogous situation for the yeast ATPase complex and have shown that the product is still immune precipitable (Fig. 1). Mild proteolysis of beef heart FI-ATPase

.,

6992

ATPase Structure Subunit I l 80 ook

6oi*\

TABLE I Relative sensitiuities of R-ATPase and ATPase complex subunits to trypsin degradation

Sensitivity to Trypsin“

Subunit

ATPase

F1-ATPase complex/FIATPaseb

comolex

1

2 3

5

100

Subunit 2

40 0

Ps O0.0 2

100

h

0.2

0.4

0.6

Subunit 3

l+la conversiond

ATPase

0.125 0.165 0.0360 0.0480021.3

0.040

0.055 0.0055 0.00225

N

3-1 3.0 6.5

~

~

1.03 1.oo 2.17 7.10

“Sensitivity of a subunit to trypsin evaluated by extrapolating percentage of subunit remaining after degradation at low trypsin ratios to the trypsin to substrate ratio at which no subunit remains. Data from Fig. 1for subunits 1,2, and 3. hRatios of trypsin sensitivities of given subunit in the ATPase complex and in the F1-ATPase. ‘Ratios of trypsin sensitivities of given subunit in the ATPase complex and in the Fl-ATPase normalized to the subunit 2 ratio. Ratios determined as in footnote a except extrapolation is to 100% conversion of subunit I to subunit la. Data from Fig. 5.

80

la

40

Trypsin to Substrate Ratio

I

(Pto?ein/Proiein) FIG.2. Trypsin degradation of subunits 1,2,and 3 of the FIATPase, Ff-ATPase in MTEA, pH 8.0, was degraded for 30 min a t 26 “C with trypsin at the indicated relative concentrations (w/w).

L

b

ZT

The reactions were terminated by cooling on ice and adding a 5-fold or greater excess of soybean trypsin inhibitor (w/wversus trypsin). $0-pg samples were electrophoresed on 15-cm 10 to 17%acrylamide, 0.1% sodium dodecyl sulfate slab gels, as described under “Materials and Methods,” stained with Coomassie Blue, and scanned. Data points, for Ff-ATPase;-, for the ATPase complex (Fig. 1).

5 0

3 0

CD

N c

L?

v)

2’ol t.5

0

0.) 0 2 0.3 04

05 0.6 0.7

Trypsin lo SubstrateRotio (Protein /Protein) FIG.

3. Comparison of the susceptibilities of subunits 1 and

2 to trypsin degradation in the F1-ATPase and the ATPase complex. The ratios of the amounts of subunits (1 + la) and 2

remaining are plotted uersus relative trypsin concentration. Thirty-

and the F1-ATPase (0)were p g samples of the ATPase complex (0) digested at the indicated trypsin to substrate ratios and electrophoresed on 10% acrylamide, 0.1% sodium dodecyl sulfate slab gels the better to resolve subunit 2 from subunit 1 and its major tryptic fragment, subunit la. TheCoomassie blue-stained gels were scanned and the areas under the curves were measured to determine the relative amounts of subunit (1 + la) tosubunit 2.

to removebound nucleotides (21) presumably results in a similar conversion of the mammalian enzyme; however, the gels usedin this studywould not have detected such a mobility shift. Fig. 4 presents densitometric scans showing that the subunit 1degradation product (termed subunitla) also occurs in yeast B1-ATPase.In contrast to the yeast ATPase complex,

Migrotion FIG. 4. Densitometric scans of the subunit 1 to subunit l a conversionin the FpAWase. Thirty-pg samples were prepared as described in legend of Fig. 1with trypsinto substrate ratiosof: A, 0.0; B, 0.001; and C, 0.01 (wt/wt). Thirty-pg aliquots were electrophoresed on a 15-cm 10%acrylamide, 0.1%sodium dodecyl sulfate slabgel until proteins smaller than subunit 2 had migrated off the end of the gel. The gel was processed as described under “Materials and Methods.”

the subunit 1 to laconversion is complete at very low trypsin tosubstrate ratios. Fig. 5 shows that the subunit 1 to l a conversion is complete for the F,-ATPase (Fig. 4B)at trypsin to substrate ratios where less than 10% of the conversion has occurred in the ATPase complex (Fig. 4A). As shown in Table I, the subunit 1to laconversion is approximately 21-fold more sensitive in the F,-ATPase. Correcting for the %fold increase in subunit 1and 2 sensitivity (see above), the locus for subunit 1 to l a conversion is about 7 times more accessible in the FI-

~

~

~

6993

ATPase Structure

0’

I

I

I

02

03

~

c?l

-

0

0

0.1

I

I

I

0.2

0.3

0.4

0

01

04

TrypsintoSubstrateRatio(Protein/Protein) FIG. 6. Effect of trypsin degradation on the adenosine triphosphatase activity of the FrATPase and the ATPase complex. FI-ATPase and ATPase complex samples wereprepared for

040

0.005

,

0.010

1

4

0.015

0.02

TrypsintoSubstrateRatio(Protein/Protein) FIG. 5. Conversion of subunit 1 to subunit l a by trypsin. Samples were prepared and analyzed as described in the legend to Fig. 1. A; ATP-ase complex;B, F1-ATPase.Note difference in trypsin to substrate ratio scales. AU samples were electrophoresed on 15-cm 10% acrylamide, 0.1%sodium dodecyl sulfate slab gels as in Fig. 4. The subunit 1 and la curves were manually resolved by fitting two normal curvesto the observed 1+ l a patterns. Percentage conversion values determined by two differentmethods agreed within 10%.

ATPase. Whether this represents the exposure of a unique trypsin-sensitive site or a group of sites has not been demonstrated. The subunit 1to la conversion patterns and the subunits 1, 2, and 3 decreases are the same when the digested samples are chromatographed on Sephadex G-150 prior to gel electrophoresis (data not shown). Thus, the subunit 1 to l a conversion takes place on intact F1-ATPase and ATPase complex, not on dissociated subunits free in solution. The increased sensitivity of the subunit 1 to la site(s) therefore cannot be ascribed to dissociation of the F1-ATPase during trypsin degradation. It has been reported previously that degradation of submitochondrial particles (22) or FI-ATPase (21,23)frombeef heart mitochondria with low levels of trypsin enhanced ATPase activity. The effect waspresumably due to removal of the ATPase inhibitor (Ref. 23; subunit 8 of the present study). It was of interest, therefore, to determine the enzymatic activity of yeast FI-ATPase and ATPase complex following trypsin degradation. Since the Ft-ATPase preparations used in this study (3, 14) lack the ATPase inhibitor (subunit 8), there should be an increase of activity only with the ATPase complex. In addition, the degree of oligomycin inhibition of the ATPase activity should serve as a measure of the intactness of the oligomycin binding protein, subunit 9 (24), and its associated subunits. Fig. 6 presents the ATPase activity profies of the FI-ATPase and the ATPasecomplex followingtrypsin degradation at the indicated trypsin to substrate ratios. As predicted, there is an increase in ATPase activity only for the ATPase complex. In addition, there is a striking increase in oligomycin-insensitive ATPase activity for the ATPase complex (Fig. 6, dashed line, open circles). This indicates that the ATPase inhibitor

ATPase activity determinations as described under “Materials and Methods” and in the legend of Fig. 1 except samples were cooled for less than 5 min, trypsin inhibitor was added, and samples were putat room temperature until assayed (15 to 30 min). W,activity in the absence of oligomycin. 0-- -0, activity in the presence of 20 pg of oligomycin/assay.

(subunit 8) and the oligomycin sensitivity conferring protein (subunit 9) possess similar trypsin sensitivities in the ATPase complex whichare about&fold greater than subunits 1 and 2. We are assuming here that the inhibitory effects of subunits 8 and 9 are not related to one another. No simple explanation for the shape of the ATPase activity profile of the F,-ATPase is presently available. The initial sharp decrease and then gradual decrease in activity for the F1-ATPase does not correspond to any known tryptic attack site. The activity curves probably reflect a combination of activation and inhibition events, some of which have not been identified. DISCUSSION

The principal new findingsof the present study arethat the dissociation of F,-ATPase from the solubilized yeast oligomycin-sensitive mitochondrial ATPase complex increases the sensitivities of subunit 3 and a site or sites on subunit 1 to trypsin degradation. In addition, earlier findings of the trypsin activation of the ATPase complex (22, 23) have been confirmed and extended, and the trypsin sensitivities of all of the ATPase complex subunits except subunit 7 have been estimated. Increased trypsin sensitivity is most easily visualized as an increased exposure of a subunit or part of a subunit to the action of trypsin. However, possible conformational changes of exposed areas on the ATPase complex during the preparation of the F1-ATPase have not been ruled out. Inspection of the data of Fig. 3 suggests this is not the case, at least for subunits 1 and 2. The overall sensitivity of subunits 1 and 2is the same for the FI-ATPase and the ATPase complex. This overall sensitivity reflects the average accessibility of trypsin to more than 30 potential trypsin sites on each subunit.* It seems unlikely that conformational changes would fortuitously compensate to maintain the same average trypsin sensitivity for subunits 1 and 2. If subunits 1 and 2 are in the same conformations in soluM. Douglas, unpublisheddata.

6994

ATPase Structure

A

strate strong subunit 1-Fo-ATPaseassociations and subunit 13-8 associations (8). Recently, more refined studies on the digestion products of subunit 1 in E. coli (18) have shown that an NHz-terminal sequence of 15 to 19 amino acids is removed by mild proteolysis and that active enzyme missing this fragment can no longer bind added 6 subunit or rebind to F,-depleted membranes. The data presented here and an accompanying paper (8) support the association of subunit 1 with the S subunit (subunit 5 in yeast) and favor model C in Fig. 7 for the orientation of subunit 1 with the membrane-associated portion of the complex. The results of trypsin degradation of both the F1-ATPase and the ATPase complex are consistent with the proposed model. Based on the model, it can be predicted that subunits 1, 2, 3, and 8 will remain exposedin inside-out submitochondrial particles and that the other subunits will be less accessible to trypsin in any membrane-bound form. Preliminary studies indicate that these predictions are correct. REFERENCES

FIG. 7. ATPase complexmodels demonstrating different interpretations of changes in subunit sensitivity to trypsin deg-

radation. Models are not drawn to scale and only certain subunits are identified (by number) in each model. Components of the FI portion are indicated by a bracket. Subunit stoichiometries, where indicated, are from Ref. 3.

bilized ATPase complex and F,-ATPase, it seems reasonable to expect that theconformation of subunit 3 is also preserved. In this case, the changes in trypsin sensitivity of subunit 3 between solubilized ATPase complex and F1-ATPase reflect an increased exposure of subunit 3 to trypsin. This increased exposure is compatible with two types of structural models. First, subunit 3 may be at the boundary of the Fl and membrane-associated components of the ATPase complex (Fig. 7 , model A ) . Dissociation of the F1 and membrane-associated portions would, therefore, increase the exposure of subunit 3. Second, subunit 3 could be complexed with subunits 5 and/or 8 (missing in the present F1-ATPasepreparations; see Ref. 8). Formation of the Fl-ATPase would break apart this complex thus exposing further subunit 3 (Fig. 7, model B ) . Chemical cross-linking studies of the ATPase complex demonstrates a strong subunit 3-subunit 8 association (8). It should be noted that these models use the subunit stoichiometries previously determined for the yeast ATPase complex (3).In addition, the stoichiometries of subunits 1, 2, and 3 are conserved in the dissociation of the ATPase compIex to the Fl-ATPase (3). The increased sensitivity of the subunit 1 to subunit la conversion site(s) represents a specific change in accessibility to trypsin. Whether this represents a single site or a small group of trypsin sensitive sites is not known. Sequence studies are underway to determine if one or both ends of subunit 1 are attacked in this conversion. Whether one orboth ends are involved, the linear increase in per cent conversionwith increasing trypsin (Fig. 5) indicates that all subunit 1 molecules in a given F1-ATPase arein the same conformation, with respect to trypsin degradation. Again, two general types of modeIs are compatible with the above data. First, assuming one end of subunit 1is involved, the subunit1 to la conversion site 1 is close to theFI-Foboundary (Fig. 7, model C ) . Second, the loss of subunits 4 and/or 8 could expose the conversion site (Fig. 7, model D ) . Chemical cross-linking studies demon-

1. Tzagoloff, A., and Meager, P. (1971) J. BioL Chem. 246, 73287336 2. Ryrie, I. J., and Gallagher, A. (1979) Biochim. Biophys. Acta 545, 1-14 3. Todd, R. D., Griesenbeck, T. A., and Douglas, M. G . (1980) J. Biol. Chem. 255,5461-5467 4. Schatz, G., Penefsky, H. S., and Racker, E. (1967) J. Biol. Chem. 242,2552-2560 5. Takeshige, R., Hess, B., Bohr, M., and Zmmermann-Telschow, H. (1976) Hoppe-Seyler's 2. Physiol. Chem. 357, 1605-1622 6. Douglas, M. G., Koh, Y., Dockter, M. E., and Schatz, G. (1977) J. Bioi. Chem. 252,8333-8335 7. Kagawa, Y . ,Sone, N., Hirata, H., Yoshida, M. (1979) J. Bioenerg. Biomembr. 11.39-78 8. Todd, R. D., and Douglas, M. G. (1981)J.Biol. Chem. 256,69846989 9. Bragg, P. D., and Hou, C. (1975) Arch. Biochem. Biophys. 167, 311-321 10, Baird, B. A., and Hammes, G. G . (1976) J. Biol. Chem. 251,69536962 11. Enns, R.,and Criddle, R. S. (1977) Arch. Biochem. Biophys. 183, 742-752 12. Munoz, E., Freer, J. H., Ellar, D. I., and Salton, M. R. (1968) Biochem. Biophys. Acta 150, 531-533 13. Wakabayashi, T., Kubuta, M., Yoshida, M., and Kagawa, Y. (1977) J. Mol. Biol. 117,515-519 14. Todd, R. D., McAda, P. C., and Douglas, M. G. (1979) J. Biol. Chern. 254, 11134-11141 15. Douglas, M. G., Koh, Y., Ebner, E., Agsteribbe, E., and Schatz,G . (1979) J . Bioi. Chem. 254,1335-1339 16. Abrams, A., Morris, D., and Jenson, C. (1976) Biochemistry 15, 5560-5566 17. Bragg, P. D., and Hou, C. (1978) Can. J.Biochem. 56,559-564 18. Dunn, S. D., Heppel, L. A,, and Fullmer, C. S. (1980) J. Bioi. Chem. 255,6891-6896 19. Ritz, C. J., and Brodie, A. F. (1977) Biochem. Biophys. Res. Commun. 75,933-939 20. Mollinedo, F., Larraga, V., Coll, F. J., and Murioz, E. (1980) Biochem. J. 186,713-723 21. Leimgruber, R. M., and Senior, A. E. (1976) J.Bid. Chem. 251, 7103-7109 22. Leimgruber, R. M., and Senior, A. E. (1976) J . Biol. Chem. 251, 7110-7113 23. Horstman, I. L., and Racker, E. (1970) J . Biol. Chem. 245, 13361544 24. Criddle, R. S.,Arulanandan, C., Edwards,T., Johnston, R., Scharf, S., and Ems, R. (1976) in Genetics and Biogenesis of Chloroplusts a n d Mitochondria (Bucher, T., Neupert, W., Sebald, W., and Werner, S., eds) pp. 151-157, North-Holland, Amsterdam