H., and Greenwood, C. (1963) Biochern. J. 86,. 541-554) and confirmed by ourselves. We conclude that reconstitution of monomeric (subunit 111-less) en-.
Vol. 262, No. 21, Issue of July 25, pp. 10077-10079,1987 Printed in U.S.A.
THE JOURNAL OF BIOLOGICAL CHEMISTRY Q 1987 by The American Society of Biological Chemists, Inc
Reconstitution of Monomeric Cytochromec Oxidase into Phospholipid aa3 Units* Vesicles Yields Functionally Interacting Cytochrome (Received for publication, September 29,1986)
Giovanni AntoniniS, Maurizio Brunoritll, Francesco Malatesta$, Paolo Sartit, and Michael T.Wilson11 From the $Departmentof Experimental Medicineand Biochemical Sciences, University of Rome “Tor Vergata,” Rome, Italy, the §Departmentof Biochemical Sciences and Consiglio Naziomle ddle Richerche Center of Molecular Biology, Universityof Rome “La Sapienza, ” Rome, Italy, and the (1 Department of Chemistry, University of Essex, Colchester, Essex, United Kingdom When the carbon monoxide complex of fully reduced cytochrome c oxidase, reconstituted into liposomes, is mixed with oxygen-containingbuffer, complex kinetic progress curvesare observed. This pattern is seen irrespective of whether the oxidase used in reconstitution is the dimeric ormonomeric (subunit 111-depleted) enzyme. These findings are interpreted in the light of similar experiments on the detergent-solubilized enzyme reported by Gibson and Greenwood (Gibson, &. H., and Greenwood, C. (1963) Biochern. J. 86, 541-554) and confirmed by ourselves. We conclude that reconstitution of monomeric (subunit 111-less) enzyme yields, preferentially, vesicles containing more than one functional unit, possibly associated as dimers. This result is of significance to our understanding of the relationships between aggregation state and proton pumping capacity of cytochrome oxidase.
(initially at the CO off rate) it may accept electrons from cytochrome a centers in other units inwhich cytochrome a3 is stillreduced and in combination with CO. According to this interpretation, the autocatalytic oxidation of cytochrome a is not a property of the isolated, monomeric, basic functional unit of the oxidase, but of assemblies of one ormore of these, either brought together by collision or as parts of stable dimers or higher order aggregates. Accordingly, autocatalytic, relatively rapid, oxidationof cytochrome a would be diagnostic of intermonomer electron transfer. We have taken advantage of this approach to ascertain whether phospholipidvesicles, reconstituted with monomeric cytochrome c oxidase, contain single monomers (functional units) or multiples of these. This question is of central importance in deciding whether cytochrome oxidase monomers, obtained either by protein modification (e.g. removal of subunit 111) or by preparation from other species (e.g. shark heart),arestill capable of pumpingprotons.Ourresults indicate that whether the reconstitution is carried out using either a monomeric or a dimeric enzyme, oxidation of cytochrome a is autocatalytic and complete before full oxidation of cytochrome a3. We conclude from this result and parallel data obtained on solubilized monomeric cytochrome oxidase that reconstitution starting with monomeric enzyme yields vesicles containing more than one monomer,possibly assembled as stable dimers. This conclusion is in agreement with recent data obtained by different methods from other laboratories (3, 4) and leaves open the question of whether the monomer is competent in pumping protons.
Gibson and Greenwood (1, 2) described experiments designed tomeasurethecarbon monoxide dissociationrate constant from fully reduced cytochrome c oxidase. These experiments, performed by mixing the CO complex of the enzyme with oxygen-containing buffer in a stopped-flow apparatus, brought to light an unexpected phenomenon. The time course of the reaction was expected to be monophasic and rate limited at the CO dissociation rate constant from cytochrome a3, since the other metal sites with spectroscopic signatures, i.e. cytochrome a and Cu,, are oxidized by oxygen via cytochrome a3. However, the progress curves were complex, consisting of a slow exponential and a faster autocataEXPERIMENTAL PROCEDURES lytic phase, the amplitudes of which depended on the obserCytochrome c oxidasefrom ox heart waspurifiedaccording to vation wavelength. The interpretationof these processes was Yonetani ( 5 ) ;the final suspending buffer, 0.1 M phosphate, pH 7.3, as follows: (i) the slow rate reports the oxidation of cyto- contained 0.2%K+/cholate. Hammerhead shark cytochrome oxidase, chrome a3, rate limited at the CO “off” rate; aconclusion prepared following Wilsonet al. (6), from the heart of Sphyrna lewini independently verifiedby parallelexperimentswithnitric was a kind gift of Dr. D. Bickar. Subunit 111-depleted enzyme was to Puettner et al. (7); ultracenoxide which displaces CO but does not oxidize the enzyme; prepared and characterized according trifugation revealed the preparation to be monodispersed, with an S and (ii) the faster autocatalytic phase reports the oxidation of 6-8 S, similar to monomeric shark oxidase (6, 8), whereas of cytochrome a, as seen for example a t 450 nm. This auto- value the native bovine enzyme had an S value of 12-13,characteristic of catalytic time course for oxidation of cytochrome a has been the dimer. The protein concentration was determined spectrophotointerpreted in terms of electron redistribution between func- metrically, Ac 605 (redox)= 11 mM” cm” total heme ( 5 ) .Cytochrome tional units of the enzyme, such that once a unit is oxidized c (type VI) was purchased from Sigma. Soybean phospholipids(L-W phopshatidylcholineType 11-S, Sigma), employed for enzyme recon* This work was partially supported by Stimulation Action Grant stitution, were repurified before use by acetone-ether refractionation ST1-086-J-C(CD)from the E. E. C. and grants from the Minister0 cycles. Reconstitution procedures for both native and subunit IIIdella Pubblica Istruzioneof Italy (to M. B.) and the S. E. R. C. of the depleted enzyme were carried out as previously reported (9).COV’ United Kingdom (to M. T. W.). The costs of publication of this article (cytochrome c oxidase vesicles) preparations were characterized in and, unless were defrayed in part by the payment of page charges. This article terms of respiratory control ratio and sidedness (9, 10) must therefore be hereby marked “aduertisement”in accordance with otherwise stated, were suspended in 10 mM Hepes buffer containing 18 U.S.C. Section 1734 solely to indicate this fact. ll Fogarty Scholar-in-Residence,FIC,National Institutes of * The abbreviationsused are:COV, cytochrome c oxidase vesicle(s); Health, Bethesda, MD 20205. Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid.
10077
10078
Kinetics of Reconstituted Monomeric Cytochrome Oxidase
39.6 mMKC1 and 40.6 mM sucrose, pH 7.3. Stopped-flow experiments were carried out using a thermostated Durmm-Gibson stopped-flow apparatus equipped with a 2-cm light path observation chamber. The kinetics of CO displacement by molecular oxygen from dithionite-reduced cytochrome oxidase were followed both in the visible and in the Soret region. All solutions were carefully degassed and nitrogen equilibrated for several cycles. Typically, to a suspension of degassed and N2-equilibrated COV, containing 5 p M oxidase, dithionite (0.5 mM final), and co (0.1 mM final) were added. The resulting suspension, containing the carbon monoxide derivative of the fully reduced enzyme, was mixed with oxygensaturated (1.34 mM) buffer in the stopped-flow apparatus. When necessary, the COV preparation was diluted anaerobically adding a known volume of degassed buffer containing 0.1 mM CO and 0.5 mM dithionite. RESULTS
Wehave reconstitutedbothsubunit 111-depleted cytochrome oxidase (subunit 111-less COV) and the enzyme containing the full set of subunits into phospholipid vesicles (COV) (9)and have studied the kinetics of displacement by dioxygen of the carbon monoxide complex of the fully reduced enzyme. The results of a typical experiment, shown in Fig. 1 at 430 and 450 nm, indicatethat themeasured time course is similar for subunit 111-less COV and for COV. At 430 nm, two kinetic components in the opposite directions are observed, as reported by Gibson & Greenwood (1)for the enzyme in detergent and confirmed by us under various experimental
0.5
I
0
0.4 h
7
””
0.3
v)
v
4
B
Y
0.2 0.1
0.0
0
-1
+1
0
-1
+l
log[Cytochrome OxidaseIlcM FIG. 2. Effect of oxidase concentrationon the oxidation of cytochromea by Osin COV. All conditions are thesame as in Fig. 1, except that the monitoring wavelength was either 600 or 450 nm. COV weresequentially diluted with Oz-freebuffers as described under “Experimental Procedures.” The rateconstants for cytochrome a oxidation were determined from semilogarithmic plots of absorbance change uersus time and correspond to maximal rates. Cytochrome oxidase concentrations are given in terms of total heme. The arrow denotes the value of the overall CO off rate. Panel A , 0, COV; panel E , 0 subunit 111-lessCOV. The additional points (*) in panel B indicate the behavior of subunit 111-lesscytochrome oxidase in detergent. Experimental points were averaged for clarity. 2.0
.-g
1.8
U
0
m
1.6
8
1.4
-
1.2
E
CI)
0
1.o
20
40
60
80
time ( e )
I
I = -
m
n L 0
100
v)
a
a
a
75
FIG. 3. Reaction with oxygen of the CO derivative of reSphyrna lewini duced cytochrome oxidase from shark heart in detergent. The CO derivative of shark cytochrome oxidase at three different concentrations (8p ~ 1,p ~ and , 0.06 p ~ was ) mixed in the stopped-flow apparatus with oxygen (1.34 mM). The time course of the reaction at 450 nm (open symbok),which is autocatalytic at high concentration, slows down, and becomes exponential at the lowest concentration. The kinetics ofCO dissociation, followed at 585 nm, are reported for comparison (0).Cytochrome oxidase concentrations in the figure are after mixing. Conditions: 25 m M K+/ phosphate buffer, 0.5%Tween 80,pH 7.3, temperature 20 “C.
conditions.’ The slower kinetic component represents the dissociation rate constantfor CO ( k = 0.02sec” for COV and for subunit 111-less COV), whereas the faster process has the 25 spectral properties of cytochrome a oxidation. At 450 nm, where the absence of a slow phase indicates that only the oxidation of cytochrome a is monitored, the reaction time course is clearly autocatalytic with an initial rate which is t i m e (s) close to that characteristic of CO dissociation from reduced FIG. 1. Displacement of carbon monoxide by dioxygen fromcytochrome q . In the case of COV and subunit 111-less COV, reduced COV. Cytochrome oxidase reconstituted into liposomes, the kinetics of the reaction were wavelength-dependent at reduced with minimal dithionite in the presence of 0.1 mM CO, was every concentration of cytochrome oxidase examined, indimixed with oxygen-saturated buffer (10 mM Hepes, 39.6 mM KCl, cating thatin all cases completion of cytochrome Ua-CO 40.5 mM sucrose, pH 7.3). The reaction was followedat 430 nm (fast oxidation lags behind that of cytochrome a. The autocatalytic phase, cytochrome a oxidation; slow phase, CO displacement and time course of cytochrome a oxidation and its maximal rate cytochrome a3 oxidation) and at 450 nm (only cytochrome a oxida50
tion). Panel A, h = 430 nm; 0,subunit 111-less COV0.3 pM; 0, c o v , 4.4 pM. Panel B, X = 450 nm; 0,subunit 111-lesscov 0.3 p M ;0,cov 0.3 pM. Temperature 20 ‘c.
M. T. Wilson, G . Antonini, F. Malatesta, P. Sarti, and M. Brunori, manuscript in preparation.
Kinetics of Reconstituted Monomeric Cytochrome Oxidase (which is 5-10 times greaterthan theCO off rate) were totally independent of protein concentration for both COV and subunit 111-less COV, as shown in Fig. 2, over a concentration range from approximately 5 to 0.05 NM. The same reaction has been extensively reinvestigated also for the enzyme in solution and complete description and analysis will be published elsewhere. It is relevant, however, at this point to emphasize that the monomeric subunit IIIless enzyme in detergent exhibits a concentration-dependent rate of oxidation of cytochrome a which at very low concen) the autocatalytic character and tends trations (
