The influence of temperature and osmolyte on the catalytic cycle of cytochrome c oxidase Jack A. Kornblatt 1 , Bruce C. Hill 2 and Michael C. Marden 3 1 Enzyme Research Group, Concordia University, Montreal, Quebec, Canada; 2 Department of Biochemistry, Queen’s University, Kingston, Ontario, Canada; 3 INSERM U473, Le Kremlin-Bice ˆ tre Cedex, France The influence of temperature on cytochrome c oxidase (CCO) catalytic activity was studied in the temperature range 240–308 K. Temperatures below 273 K required the inclusion of the osmolyte ethylene glycol. For steady-state activity between 278 and 308 K the activation energy was 12 kcalÆmol )1 ; the molecular activity or turnover number was 12 s )1 at 280 K in the absence of ethylene glycol. CCO activity was studied between 240 and 277 K in the presence of ethylene glycol. The activation energy was 30 kcalÆmol )1 ; the molecular activity was 1 s )1 at 280 K. Ethylene glycol inhibits CCO by lowering the activity of water. The rate limitation in electron transfer (ET) was not associated with ET into the CCO as cytochrome a was predominantly reduced in the aerobic steady state. The activity of CCO in flash-induced oxidation experiments was studied in the low temperaturerangeinthepresenceofethyleneglycol.Flash photolysis of the reduced CO complex in the presence of oxygen resulted in three discernable processes. At 273 K the rate constants were 1500 s )1 ,150s )1 and 30 s )1 and these dropped to 220 s )1 ,27s )1 and 3 s )1 at 240 K. The acti- vation energies were 5 kcalÆmol )1 ,7kcalÆmol )1 ,and 8kcalÆmol )1 , respectively. The fastest rate we ascribe to the oxidation of cytochrome a 3 , the intermediate rate to cyto- chrome a oxidation and the slowest rate to the re-reduction of cytochrome a followed by its oxidation. There are two comparisons that are important: (a) with vs. without ethy- lene glycol and (b) steady state vs. flash-induced oxidation. When one makes these two comparisons it is clear that the CCO only senses the presence of osmolyte during the reductive portion of the catalytic cycle. In the present work that would mean after a flash-induced oxidation and the start of the next reduction/oxidation cycle. Keywords: cytochrome coxidase; osmolytes; rate limitations. Cytochrome c oxidase (CCO) is the terminal electron transfer (ET) enzyme of the mitochondrial electron trans- port chain and a site of energy transduction. During the catalytic cycle, the enzyme accepts electrons one at a time from cytochrome c; it stocks electrons in four metal centers (Cu A , cytochrome a, cytochrome a 3 and Cu B ) and finally transfers four electrons to oxygen. The reduced oxygen combines with four protons to form two molecules of water. At the same time, protons are pumped from one side of the protein to the other [1]. As the CCO is normally inserted in the mitochondrial membrane, this pumping, added to the consumption of protons in the mitochondrion, results in the formation and maintenance of a transmembrane gradient of protons. It, in conjunction with the membrane potential, powers the synthesis of ATP as well as other energy requiring functions [2]. The CCO from bovine heart contains 13 different protein subunits, two hemes in the form of cytochromes a and a 3 , two coppers in the form of Cu A with a third as Cu B ;italso contains Mg, Zn and some tightly bound phospholipid or detergent. The three-dimensional structures of the bovine heart CCO [3,4] and the Paracoccus denitrificans CCO [5] are available from the protein data bank. The cytochrome c binding site is on the side of the oxidase that faces the cytosolic compartment of the cell. Electrons enter the oxidase one at a time from cytochrome c; they enter via Cu A [6–8] which is contained within subunit II of the protein. The groundwork for establishing the ET sequence was performed by Chance et al. [9,10] and by Gibson and Greenwood [11]. The sequence of electron transfers into, out of, and within CCO is now more or less defined but there is not universal agreement on the oxidation pathway (see [8,12–14]). The initial ET is from cytochrome c into Cu A [7] which rapidly equilibrates with cytochrome a [15]. The second electron rereduces Cu A thereby forming the initial two-electron reduced oxidase. In the absence of oxygen or the presence of CO, there is probably a two- electron reduction of the cytochrome a 3 /Cu B site and this is followed by rereduction of the Cu A /cytochrome a couple. Our scheme for working with the oxidative pathway is based on Hill’s analysis [8] but it is not critical for the data reported here. The majority view for the oxidation pathway is summarized in a recent paper by Morgan et al. [14]. CCO acts as a proton pump [1]. It actively transfers about one proton across the protein for each electron that is transferred to oxygen. Under most conditions, there is no Correspondence to J. A. Kornblatt, Enzyme Research Group, Concordia University, 1455 de Maisonneuve, Montreal, Quebec, CANADA H3G 1M8. Fax: + 33 4 67 52 36 81 (until May 2003), + 1 514 848 2881 (after May 2003), Tel.: + 1 514 848 3404, E-mail: krnbltt@vax2.concordia.ca Abbreviations:CCO,cytochromec oxidase; TMPD, tetramethyl phenylene diamine; ET, electron transfer. (Received 2 October 2002, accepted 20 November 2002) Eur. J. Biochem. 270, 253–260 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03381.x slippage. ET and proton transfer are tightly coupled [16]. This gives rise to the phenomenon of respiratory control. In terms of the electron transfer reactions, proton transfer occurs during the reduction of the binuclear center and during the reduction of oxygen from the peroxy to the oxyferryl intermediates [17]. A water cycle during turnover has been established [18,19]. The application of hydrostatic-pressure influences CCO only during turnover [20]; the current view is that hydrostatic pressure exerts its effects on proteins by influencing hydration. Reducing the activity of water also influences CCO but only during turnover [18]. A series of small hydrophilic molecules is capable of inhibiting CCO activity and the inhibition scales with the size of the small molecule effector [21]. The most potent inhibitor is that whose size is closest to that of water. Based on a stochastic model, the inhibitor studies indicate that around one water molecule enters and leaves the oxidase for every proton that is transported. Other techniques indicate that between one and two water molecules enter and leave the CCO for every proton transported [22]. The water cycle is coupled to ET andprotontransfer. The work presented here was based on the view that pure ET reactions should be relatively insensitive to temperature changes whereas combined ET, proton and water transfer should be more sensitive. As temperatures are reduced, the motions of waters internal to protein structures, are reduced. To study enzyme catalysis at low temperature, it is necessary to include ÔantifreezeÕ or osmolyte in the enzyme mixture. This lowers water activity and imposes a new rate limitation on catalytic turnover. We use the term Ôrate limitationÕ to indicate that it is the net effect of an unknown number of ÔslowÕ steps and that it is not the effect of temperature on a single rate constant. Northrop [23], Brown and Cooper [24], and Ray [25] have shown that enzymes cannot usually be analyzed in terms of a single, slowest, rate determining step. All enzymes, though, are rate limited. Temperature effects on the rate limitation have been exploited over the years through the use of the Arrhenius equation. Temperature sensitivity of the reaction rate when that rate is not limited by substrate availability yields an Arrhenius activation energy which is one among many characteristics of the enzyme. For CCO which has four well defined metal centers that act as electron acceptors and donors, ET rates into and out of these centers are influenced by temperature and can be analyzed using the Arrhenius relation [26]. In this work we show, as have many other studies, that steady state ET from cytochrome c to oxygen decreases as temperature decreases. Furthermore, steady state sensiti- vity at low temperature with osmolyte present is greater than that exhibited for the internal ET reactions during flash induced oxidations. This implies that the rate limitation in the steady state, low temperature ET reaction is developed only during the reductive phase of the catalytic cycle. This idea is in keeping with the results of other studies [12,27–32]. In terms of the comparisons made in this work, it is only with the start of a second cycle ) ET from cytochrome c to the oxidized oxid- ase ) that low water activity causes this rate limitation to shift from one set of steps to another. Materials and methods The purification of CCO has been described previously [33]. The protein is prepared using cholate, and then suspended in 1% (v/v) Tween-80. Before use, the protein is mixed with an equivalent weight of Tween-80, dialyzed to equilibrium vs. 40 m M phosphate, pH 6.9 and then frozen. The preparation is stable for a period of months. Cytochrome c (prepared without trichloroacetic acid) was purchased from Sigma. Tween-80, the detergent used throughout the study, was from Fluka and was their highest grade. Ethylene glycol, enzyme grade, was from Fisher. All other chemicals were from Fluka and were the highest grade available. Steady state assays of CCO were carried out in 40 m M phosphate pH 6.9. The oxidation of cytochrome c was monitored at 550 nm. The complete assay system contained approximately 40 l M reduced cytochrome c and a variable amount of CCO (with its equivalent weight of Tween-80) depending on the temperature. Below 273 K, the assay contained 44% (w/v) 40 m M phosphate and 56% (w/v) ethylene glycol. The paH of this solution is 7.75 at 273 K and 8.1 at 243 K; the term paH indicates that the activity of hydrogen ions in the mixed solvent solution is known; when no osmolyte is present, paH and pH are the same [34]. The concentration of reduced cytochrome c was the same as in the high temperature samples. Temperature inside the assay cuvette was monitored continuously with a T-type thermo- couple (Barrant Co., Barrington, IL, USA). Flash photolysis at 532 nm was carried out with a Quantel laser with a 10-ns pulse. The cuvette holder was cooled with a double Peltier junction, the lower of which was cooled with a refrigerated bath. It was relatively easy to get as low as 240 K. The buffer system contained 44% (w/v) 40 m M phosphate pH 6.9 and 56% (w/v) ethylene glycol, 10 m M ascorbate and 50 l M tetramethyl phenylene diamine (TMPD). The oxidase, final concentration  5 l M ,was added to the buffer system kept at about 270 K. The total volume was 1 mL. The solution was gassed with CO and allowed to sit on ice until such time as it was completely reduced. The cuvette was placed in the refrigerated holder. When the cuvette attained thermal equilibrium, 87 lLof oxygenated solution of 44% (w/v) 40 m M phosphate/56% (w/v) ethylene glycol was added with a chilled syringe; the cuvette contents were mixed with the same syringe and the CO flashed off. The temperature of the cuvette wall was monitored before and after flashing. The samples were monitored at a wavelength of 442 nm and the data stocked in a LeCroy 9400 digital oscilloscope; they were treated as described below. Because these are single runoff experiments, it was not possible to average multiple flash-induced events. A total of 32 000 data points were collected for each flash and these were converted to fewer than 200, with the points at greater times being the average of neighboring points centered at the times indicated. For example, if the data are collected at 1 ls per point, the total scan covers 32 ms. The data point at 10 ms is the average of 64 points from 9.969 ms to 10.032 ms. Shorter times use fewer points to avoid a large spread compared to the time after the flash [35]. An equation containing the sum of three exponential terms was fit to the data using SIGMA PLOT . 254 J. A. Kornblatt et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Results Figure 1 shows the response of CCO to temperature changes. The assay used is spectrophotometric and meas- ures the disappearance of reduced cytochrome c absorbance at 550 nm. The assay consists of cytochrome c, CCO with its equivalent weight of Tween-80, phosphate buffer and oxygen. The data were collected from 280 to 306 K. Above this value the slope of the Arrhenius plot started to change. As we were not interested in the elevated temperatures, we did not take data far above the straight portion of the Arrhenius plot. We note that many other workers have found a break in the curve at about the same temperature [36,37]. In standard assays at 273 K, the oxidase turnover number is close to 10 s )1 which is similar to that found in the older work. Kinetic and thermodynamic parameters are summarized in Table 1. The Arrhenius activation energy calculated from the data of Fig. 1 is 12 kcalÆmol )1 ,also about the same as found in earlier work. Under conditions similar to those of Fig. 1, there is little indication that the rate limitation is a slow ET step between cytochrome a and the binuclear center. In steady state spectra in the presence of cytochrome c and TMPD/ ascorbate, the predominant form of the oxidase appears to be a mixture of the pulsed form of the oxidized oxidase and cytochrome a 2 a 3 3 . Figure 2 shows the spectral approach to the anaerobic state as electrons are transferred from TMPD/ascorbate to oxidase to oxygen. Qualitatively, the significant aspect of the spectra is that there is only a minor peak that grows in at 443 nm between 2 min and 6 min. If cytochrome a were completely reduced, the 443 nm peak would be consider- ably higher as would the 605 nm peak (vida infra). The addition of ethylene glycol to the buffers at 56% (w/v) allows one to work at temperatures as low as 230 K. Provided the phosphate concentration is kept < 0.1 M , there is no precipitation of phosphate even at the lowest temperatures. At temperatures < 290 K the response of the oxidase to changing temperature appears to follow a linear Arrhenius relation (Fig. 3). Above 290 K there is a slow (minute time scale) precipitation of the CCO which complicates the kinetic analysis and prevents us from being able to make a direct comparison between CCO at high temperature (> 290 K) with and without ethylene glycol. The bending of the curve in Fig. 3 at the higher temperatures Fig. 1. CCO was assayed spectrophotometrically in 40 m M phosphate pH 6.9. The initial reduced cytochrome c concentration was 40 l M ;its oxidation was followed at 550 nm. The temperature in the cuvette was monitored during the assay with a T-type thermocouple; the tem- perature was constant within ± 0.1 K over the course of the meas- urement. The rate constant on the ordinate is a turnover number, expressed on a per second time scale. Table 1. Kinetic and thermodynamic constants for the steady state activity and flash induced oxidation activity of cytochrome c oxidase. No ethylene glycol 56% ethylene glycol Steady state turnover number at 280 K 12 s )1 1s )1 Steady state E à (T > 273 K) 12 kcalÆmol )1 (T < 273 K) 30 kcalÆmol )1 Single kinetic constants for the three identified kinetic processes T ¼ 300 K a T ¼ 273 K 25 000 s )1 1500 s )1 10 000 s )1 150 s )1 800 s )1 30 s )1 Flash induced oxidation E à for the three distinguished processes T > 273 K a T < 273 K 3 kcalÆmol )1 5 kcalÆmol )1 7 kcalÆmol )1 7 kcalÆmol )1 13 kcalÆmol )1 8 kcalÆmol )1 a Data taken from Oliveberg et al. 1989 [26]. Fig. 2. The CCO aerobic steady state at 276 K. The oxidase concen- tration was  5 l M , the cytochrome c was 3 l M . The buffer was 40 m M phosphate pH 6.9. Spectrum 1, before the addition of TMPD and ascorbate; spectra 2–6, after the addition of 3 m M ascorbate and 300 l M TMPD; spectra 7–10, progression to the totally reduced oxidase. Ó FEBS 2003 Cytochrome c oxidase activity at low temperature (Eur. J. Biochem. 270) 255 reflects the fact that multiple processes are occurring. Below 290 K there is only one discernable process and it has an activation energy of 30 kcalÆmol )1 . Qualitatively, the bottleneck that slows the catalytic activity is between cytochrome a and cytochrome a 3 (Fig. 4) as was shown in earlier work [18]. The data were collected at 250 K in the presence of low concentrations of cytochrome c plus TMPD and ascorbate. As the concen- tration of cytochrome c is increased, the fraction of cytochrome a that is reduced increases. The fraction of cytochrome a that is reduced is, in part, represented by the peak that grows in at 443 nm and that which disappears at 416 nm. The spectra of Figs 2 and 4 are compared in Fig. 5 where the emphasis is on the changes occurring at 443 nm and 605 nm. The progression to and out of the steady state is shown in Fig. 5. Both monitoring wavelengths show that the steady state is reached within 5 min of TMPD/ascorbate addition to either high or low temperature samples. The steady state is then maintained over the course of 6 min (276 K) or 600 min (250 K) until the preparations become anaerobic. The time course of the absorbance changes at 276 K (d) and 250 K (s) are shown in (A) and (B) based on the data of the absorption spectra of Figs 2 and 4. In both panels, the data have been normalized to a concentration of 1 l M oxidase and have been corrected at 605 nm for the absorbance contributed by oxidized cytochrome c (< 5%). Figure 5A shows the time course of the absorb- ance change at 443 nm. The difference between the steady state values of the 276 K sample and the totally reduced sample at 276 K are clearly much larger than the compar- able difference seen for the 250 K sample. This difference reflects the extent to which cytochrome a is reduced; it is more reduced in the 250 K sample than in the 276 K sample. An approximation [38] of the extent to which cytochrome a is reduced can be obtained from the data of (B) which shows the time course of changes at 605 nm. This approximation is based on the fact that the 605 nm band is almost exclusively the result of cytochrome a absorption. Under the conditions used here the extinction coefficient for the oxidized oxidase is 26 m M )1 Æcm )1 ; this is the pulsed form. The extinction coefficient for the totally reduced oxidase at605nm is 40m M )1 Æcm )1 . A 605 for the 276 K sample starts at  0.034 in the steady state and yields a value of 0.042 for the totally reduced oxidase. This corresponds to  50% reduction of cytochrome a in the steady state at 276 K. At 250 K, in the presence of ethylene glycol, A 605 is  0.038 in the steady state and is 0.042 when the sample goes totally reduced. This corresponds to 75% of the cytochrome a reduced in the steady state. The presence of ethylene glycol inhibits the oxidase by reducing electron flow between cytochrome a and cytochrome a 3 . In order to study the nature of the block, we carried out flash induced oxidation experiments in the presence of 56% (w/v) ethylene glycol at temperatures below 273 K. Reduced CO oxidase was mixed with oxygen and the CO flashed off with a 10-ns pulse (532 nm). Fig. 6 shows the evolution of absorption changes at 442 nm as a function of time. Two typical data sets are shown. One was collected at 268 K (s) and the second at 238 K. An equation containing the sum of three exponential rates was fit to the data thereby yielding three rate constants at each temperature. The fitted lines are included in the figure. The typical R 2 was 0.998. The three sets of rate constants from temperatures between 278 and 240 K were plotted as shown in Fig. 7. The Arrhenius energies (Table 1) are 5 kcalÆmol )1 (fastest Fig. 4. The CCO aerobic steady state at 250 K. The oxidase concen- tration was  6.3 l M ,thecytochrome c was 3 l M . The buffer was 44% (w/v) 40 m M phosphate pH 6.9, 56% (w/v) ethylene glycol. Spectrum 1, before the addition of TMPD and ascorbate; spectra 2–19, after the addition of 3 m M ascorbate and 300 l M TMPD; spectra 20–30, pro- gression to the totally reduced oxidase. Fig. 3. CCO was assayed spectrophotometrically in a mixed solvent system consisting of 44% (w/v) 40 m M phosphate, pH 6.9 and 56% (w/v) ethylene glycol. The initial reduced cytochrome c concentration was 40 l M ; its oxidation was followed at 550 nm. The temperature in the cuvette was monitored during the assay with a T-type thermo- couple and was constant within ± 0.1 K over the course of the measurement. The rate constant on the ordinate is a turnover number, based on a per minute time scale. 256 J. A. Kornblatt et al. (Eur. J. Biochem. 270) Ó FEBS 2003 process), 7 kcalÆmol )1 and 8 kcalÆmol )1 (slowest process). These numbers are much smaller than those obtained in the steady state assay; the activation energies obtained in the flash induced oxidation experiments cannot account for the rate limitation in the steady state. Discussion The catalytic activity of CCO varies from preparation to preparation and from laboratory to laboratory. It is a function of ionic strength, pH, detergent and detergent concentration, temperature, cosolvents, hydrostatic pres- sure, osmotic pressure and probably other factors. None- theless, catalytic activity is still one of the few characteristics that reflect the actual role of the oxidase in the tissues. What is that role? The oxidase catalyzes the transfer of electrons from cytochrome c to oxygen. It couples the energy liberated in the process to the generation of an electrochemical gradient of protons. The two appear to be tightly coupled under most circumstances. Even under conditions where the oxidase is not contained within a membrane, protons are still pumped from one side of the protein to the other. There is also a water cycle that is coupled to electron and proton transfer. Water must enter and leave the protein for the oxidase to turnover [18,21]. The function of this last cycle is unknown but it may be part of the proton transfer machinery. The overall rate constant for any process is determined by all the rate constants that contribute to the reaction [23–25]. InthecaseofETfromcytochrome c to oxygen catalyzed by CCO, rate constants as a function of temperature have been Fig. 6. Flash photolysis of Cu 1 A a 2 Cu 1 B a 2 3 CO + O 2 . Five l M reduced CO CCO was in 44% (w/v) 40 m M phosphate pH 6.9, 56% (w/v) ethylene glycol. The temperature was 268 K (s)or238 K(d). Oxygen was added with a prechilled syringe and the CO was flashed off with a 10-ns pulse at 532 nm; 32 000 data points were collected and treated as detailed in Materials and methods. An equation containing the sum of three exponentials was fit to the data. The R 2 forthefitwasbetterthan 0.998; the fit is shown as the solid line through each set of data points. Fig. 5. The time course of the absorbance changes at (A) 276 K (d)and (B) 250 K (s). These data are taken from the absorption spectra of Figs 2 and 4. In both panels, the data have been normalized to a concentration of 1 l M oxidase and have been corrected for the absorbance contributed by cytochrome c (< 5%). (A) Time course of the absorbance change at 443 nm. (B) Time course of absorbance changesat605nm.Thedatain(B)canbeusedtoapproximatethe extent of cytochrome a reduction during the steady state. It is 50% reduced at 276 K and 75% reduced at 250 K. Fig. 7. Evaluation of Arrhenius activation energies of the three processes discernible in the low temperature flash photolysis experiments. The activation energies are: k1, 5 kcalÆmol )1 (d); k2, 7 kcalÆmol )1 (s); k3: 8.0 kcalÆmol )1 (.) based on the assumption that the individual data sets could be fit to straight lines. Ó FEBS 2003 Cytochrome c oxidase activity at low temperature (Eur. J. Biochem. 270) 257 measured several times. An early study by Smith and Newton [37] showed that simple Arrhenius behavior was not followed over an extended temperature range. Two processes were evident with a break at about 30°C. At temperatures above the break, the activation energy was  8kcalÆmol )1 ; at temperatures below the break the acti- vation energy was  12.5 kcalÆmol )1 . The Smith/Newton data was not substantially different from that collected earlier by Minnaert (quoted in [36]). Other workers have subsequently reported similar activation energies in the same temperature range [39,40]. Under similar conditions, we find  12 kcalÆmol )1 at temperatures < 30 °C. The turnover numbers determined in the above mentioned papers and ours are about the same. Interestingly, the acti- vation energies for the bimolecular rate constants between cytochrome c and CCO are also about 16 kcalÆmol )1 for the temperature range < 20 °C[41]. The inclusion of high concentrations of the cryoprotec- tant, ethylene glycol, inhibits the catalytic activity of CCO [18]. In order to work at temperatures as low as 235 K, it was necessary to include 56% (w/v) ethylene glycol in the solution; this results in  90% inhibition of the overall activity of the cytochrome c oxidase, an energetic difference of  1.2 kcalÆmol )1 . Part of the inhibition probably stems from a slight weakening of the interaction between cytochrome c and the oxidase [42]; however, halving the cytochrome c concentration in the steady state assays had no effect on the measured activity. Ethylene glycol and low temperature also increase the viscosity of the medium; if there are large conformational changes that occur in the oxidase during the catalytic cycle, these would be influenced by the increased viscosity. Certainly, there are conforma- tional changes that occur here. Lowering temperature increases both the dielectric coefficient of the medium and the dielectric coefficient ÔinsideÕ the protein: both can be expected to influence catalytic activity. The majority of the inhibition arises from blocking an internal ET step located between cytochrome a and cytochrome a 3 [18]; this is clearly shown in Figs 4 and 5. The influence of temperature on the ethylene glycol- inhibited protein was studied in both the steady state condition and the flash induced oxidation condition. Steady state condition. During steady state turnover, the activation energy for ET from cytochrome c to oxygen is 30 kcalÆmol )1 (Fig. 3), 2.5 times the value found in the temperature range between 273 and 300 K in the absence of ethylene glycol. The block is not at the delivery of electrons into Cu A as the steady state spectrum of the cytochrome a shows it to be about 75% reduced (Figs 4 and 5). Low temperature that results in freezing is capable of inducing the same inhibition. Nicholls and Kimelberg [43] showed that a solution of oxidase and TMPD/ ascorbate would yield a mixed valence oxidase (cyto- chrome a reduced, cytochrome a 3 oxidized) at 77 K. It took between 40 and 240 s for the samples to freeze; during the freezing process, the sample undergoes osmotic stress. The vapor pressure of the ice surrounding the oxidase is lower than the vapor pressure of water in the protein. The Nicholls/Kimelberg experiment is therefore similar to the ones carried out here using ethlyene glycol as the osmotic stress agent [44]. Flash induced oxidation condition. During the oxidative phase of CCO, three processes can be easily seen after flash induction of oxidation. The three steps have energy barriers which are substantial (5–8 kcalÆmol )1 ). These energy barriers are relatively independent of the presence or absence of ethylene glycol and the temperature at which the measurements are made (Table 1). The reac- tions occurring during the oxidative phase cannot account for the increased energy barrier occurring during steady state turnover. The influence of temperature and ethylene glycol on the three rate constants was also studied. The rate constants decrease by > 90% when ethylene glycol is added [8,28] an energetic difference of about 1.5 kcalÆmol )1 These decrease by another 80% as the temperature is lowered from 273 to 248 K [45] or 240 K (this work). Our rate constants for the oxidation phase are still faster than the overall rate constant during the steady state. As the latter is the complex function of all the internal and intermolecular rate constants, the difference between the two sets of numbers is expected. In summary, at room temperature when no osmolyte is present, the activation energies of the individual steps in ET are comparable to the activation energy for the overall reaction. The limitation on the steady state, catalytic rate, is a reasonably fast reaction associated with ET. The inclusion of 56% (w/v) ethylene glycol changes that. The individual ET steps are slowed by the imposition of low temperature. The activation energies are about the same as they are in the absence of ethylene glycol but the activation energy for steady state turnover is far higher than it is for any of the individual oxidative steps. This can only mean that a new rate limitation has been introduced into the catalytic cycle and that this new rate limitation occurs only after the completion of the flash induced oxidation of the reduced oxidase. The situation is somewhat, but not quite, analogous to that seen by Karpefors et al. [46] when they found that the onset of a large deuterium isotope effect was not seen immediately after mixing with D 2 O but rather occurred only after a lag time. It is more in keeping with the idea that something rate limiting occurs at the onset of a second catalytic cycle of reduction of CCO by cytochrome c. Mitchell and Rich [47] proposed that two protons were taken up on reduction of CCO and that these were taken up concomitantly with reduction of the binuclear cytochrome a 3 -Cu B site. Our results could also be explained by proton uptake during reduction of the cytochrome a-Cu A site as proposed by Capitano et al.[48]. Ethylene glycol is a cryoprotectant. It acts by influen- cing the colligative properties of water. The inclusion of ethylene glycol in our solutions lowers not only the freezing point of our solutions, but also the activity of water. If water entry and exit are necessary for catalytic ET to occur, then we have shown here that this cycle is initiated only during the reductive phase of the catalytic cycle. Reduced oxidase would therefore start as a hydrated molecule. There would be no impediment to ET within the oxidase nor from the oxidase to oxygen. The rate limitation in the overall steady state process would be the entry of water accompanying reduction and ET within the protein. 258 J. A. Kornblatt et al. (Eur. J. Biochem. 270) Ó FEBS 2003 Acknowledgements We wish to thank M. J. Kornblatt for helpful discussions. This work was generously supported by Natural Sciences and Engineering Research Council (Canada) and the Institut National de la Sante ´ et la Recherche Medicale (France). References 1. Wikstrom, M.K. (1977) Proton pump coupled to cytochrome c oxidase in mitochondria. Nature 266, 271–273. 2. Mitchell, P. (1979) Keilin’s respiratory chain concept and its chemiosmotic consequences. Science 206, 1148–1159. 3. Tsukihara, T., Aoyama, H., Yamashita, E., Tomizaki, T., Yam- aguchi, H., Shinzawa-Itoh, K., Nakashima, R., Yaono, R. & Yoshikawa, S. (1996) The whole structure of the 13-subunit oxidized cytochrome c oxidase at 2.8A ˚ . Science 272, 1136–1144. 4. Yoshikawa, S., Shinzawa-Itoh, K., Nakashima, R., Yaono, R., Yamashita,E.,Inoue,N.,Yao,M.,Fei,M.J.,Libeu,C.P., Mizushima,T.,Yamaguchi,H.,Tomizaki,T.&Tsukihara,T. (1998) Redox-coupled crystal structural changes in bovine heart cytochrome c oxidase. Science 280, 1723–1729. 5. Ostermeier, C., Harrenga, A., Ermler, U. & Michel, H. (1997) Structure at 2.7 A ˚ resolution of the Paracoccus denitrificans two- subunit cytochrome c oxidase complexed with an antibody FV fragment. Proc. Natl Acad. Sci. USA 94, 10547–10553. 6. Hill, B.C. (1988) Electron transfer from cytochrome c to O2. Ann. NY Acad. Sci. 550, 98–104. 7. Hill, B.C. (1991) The reaction of the electrostatic cytochrome c-cytochrome oxidase complex with oxygen. J. Biol. Chem. 266, 2219–2226. 8. Hill, B.C. (1994) Modeling the sequence of electron transfer reactions in the single turnover of reduced, mammalian cytochrome c oxidase with oxygen. J. Biol. Chem. 269, 2419– 2425. 9. Chance, B., Saronio, C. & Leigh, J.S. Jr (1975) Functional inter- mediates in reaction of cytochrome oxidase with oxygen. Proc. NatlAcad.Sci.USA72, 1635–1640. 10. Chance, B., Saronio, C., Leigh, J.S. Jr, Ingledew, W.J. & King, T.E. (1978) Low-temperature kinetics of the reaction of oxygen and solubilized cytochrome oxidase. Biochem. J. 171, 787–798. 11. Gibson, Q.H. & Greenwood, C. (1963) Reactions of cytochrome c oxidase with oxygen and carbon monoxide. Biochem. J. 86,541– 555. 12. Branden, M., Tomson, F., Gennis, R.B. & Brzezinski, P. (2002) The entry point of the k-proton-transfer pathway in cytochrome c oxidase. Biochemistry 41, 10794–10798. 13. Einarsdottir, O. (1995) Fast reactions of cytochrome oxidase. Biochim. Biophys. Acta 1229, 129–147. 14. Morgan, J.E., Verkhovsky, M.I., Palmer, G. & Wikstrom, M. (2001) Role of the PR intermediate in the reaction of cytochrome c oxidase with O2. Biochemistry 40, 6882–6892. 15. Antalis, T.M. & Palmer, G. (1982) Kinetic characterization of the interaction between cytochrome oxidase and cytochrome c. J. Biol. Chem. 257, 6194–6206. 16. Chance, B., & Williams, G.R. (1956) The respiratory chain and oxidative phosphorylation. Adv. Enzymol. 17, 65–134. 17. Fabian, M. & Palmer, G. (2001) Proton involvement in the tran- sition from the ÔperoxyÕ to the ferryl intermediate of cytochrome c oxidase. Biochemistry 40, 1867–1874. 18. Kornblatt, J.A. & Hoa, G.H. (1990) A nontraditional role for water in the cytochrome c oxidase reaction. Biochemistry 29, 9370–9376. 19. Kornblatt, J.A., Kornblatt, M.J., Rajotte, I., Hoa, G.H. & Kahn, P.C. (1998) Thermodynamic Volume cycles for electron transfer in the cytochrome c oxidase and for the binding of cytochrome c to cytochrome c oxidase. Biophys. J. 75, 435–444. 20. Kornblatt, J.A., Hui Bon, H.G. & Heremans, K. (1988) Pressure- induced effects on cytochrome oxidase: the aerobic steady state. Biochemistry 27, 5122–5128. 21. Kornblatt, J.A. (1998) The water channel of cytochrome c oxidase: inferences from inhibitor studies. Biophys. J. 75, 3127–3134. 22. Kornblatt, J.A. & Kornblatt, M.J. (1992) Cytochrome c oxidase: the presumptive channel holds at least four waters. Biochim. Bio- phys. Acta 1099, 182–184. 23. Northrop, D.B. (1981) Minimal kinetic mechanism and general equation for deuterium isotope effects on enzymic reactions: uncertainty in detecting a rate-limiting step. Biochemistry 20, 4056–4061. 24. Brown, G.C. & Cooper, C.E. (1993) Control analysis applied to single enzymes: can an isolated enzyme have a unique rate-limiting step? Biochem. J. 294, 87–94. 25. Ray, W.J. Jr (1983) Rate-limiting step: a quantitative definition. Application to steady-state enzymic reactions. Biochemistry 22, 4625–4637. 26. Oliveberg, M., Brzezinski, P. & Malmstrom, B.G. (1989) The ef- fect of pH and temperature on the reaction of fully reduced and mixed-valence cytochrome c oxidase with dioxygen. Biochim. Biophys. Acta 977, 322–328. 27. Adelroth, P., Gennis, R.B. & Brzezinski, P. (1998) Role of the pathway through K (I-362) in proton transfer in cytochrome c oxidase from R. sphaeroides. Biochemistry 37, 2470–2476. 28. Adelroth, P., Karpefors, M., Gilderson, G., Tomson, F.L., Gennis, R.B. & Brzezinski, P. (2000) Proton transfer from glu- tamate 286 determines the transition rates between oxygen inter- mediates in cytochrome c oxidase. Biochim. Biophys. Acta 1459, 533–539. 29. Ruitenberg, M., Kannt, A., Bamberg, E., Ludwig, B., Michel, H. & Fendler, K. (2000) Single-electron reduction of the oxidized state is coupled to proton uptake via the K pathway in Paracoccus denitrificans cytochrome c oxidase. Proc. Natl Acad. Sci. USA 97, 4632–4636. 30. Junemann, S., Meunier, B., Gennis, R.B. & Rich, P.R. (1997) Effects of mutation of the conserved lysine-362 in cytochrome c oxidase from Rhodobacter sphaeroides. Biochemistry 36, 14456– 14464. 31. Konstantinov, A.A., Siletsky, S., Mitchell, D., Kaulen, A. & Gennis, R.B. (1997) The roles of the two proton input channels in cytochrome c oxidase from Rhodobacter sphaeroides probed by the effects of site-directed mutations on time-resolved electrogenic intraprotein proton transfer. Proc. Natl Acad. Sci. USA 94, 9085– 9090. 32. Namslauer, A., Branden, M. & Brzezinski, P. (2002) The rate of internal heme-heme electron transfer in cytochrome c oxidase. Biochemistry 41, 10369–10374. 33. Yonetani, T. Cytochrome oxidase from beef heart. Biochemical Preparations 11, 14–20 (1966). 34. Douzou, P. (1977) Cryobiochemistry. Academic Press, New York, USA. 35. Marden,M.C.,Kister,J.,Bohn,B.&Poyart,C.(1988)T-state hemoglobin with four ligands bound. Biochemistry 27, 1659–1664. 36. Minneart, K. (1961) The kinetics of cytochrome c oxidase. Bio- chim. Biophys. Acta 50, 23–34. 37. Smith, L. & Newton, N. (1968) Structure and Function of Cyto- chromes. University of Tokyo Press, Tokyo, Japan. 38. Kornblatt, J.A. & Luu, H.A. (1986) The interactions of cyto- chrome c and porphyrin cytochrome c with cytochrome c oxidase. The resting, reduced and pulsed enzymes. Eur. J. Biochem. 159, 407–413. 39. Malatesta,F.,Sarti,P.,Antonini,G.,Vallone,B.&Brunori,M. (1990) Electron transfer to the binuclear center in cytochrome Ó FEBS 2003 Cytochrome c oxidase activity at low temperature (Eur. J. Biochem. 270) 259 oxidase: catalytic significance and evidence for an additional intermediate. Proc. Natl Acad. Sci. USA 87, 7410–7413. 40. Wilms, J., Dekker, H.L., Boelens, R. & van Gelder, B.F. (1981) The effect of pH and ionic strength on the pre-steady-state reac- tion of cytochrome c and cytochrome aa3. Biochim. Biophys. Acta 637, 168–176. 41. Veerman, E.C., Wilms, J., Dekker, H.L., Muijsers, A.O., van Buuren, K.J., van Gelder, B.F., Osheroff, N., Speck, S.H. & Margoliash, E. (1983) The presteady state reaction of chemically modified cytochromes c with cytochrome oxidase. J. Biol. Chem. 258, 5739–5745. 42.Kornblatt,J.A.,Kornblatt,M.J.,Hoa,G.H.&Mauk,A.G. (1993) Responses of two protein-protein complexes to solvent stress: does water play a role at the interface? Biophys. J. 65, 1059– 1065. 43. Nicholls, P. & Kimelberg, H.K. (1968) Cytochromes a and a3. Catalytic activity and spectral shifts in situ and in solution. Biochim. Biophys. Acta 162, 11–21. 44. Parsegian, V.A., Rand, R.P., Fuller, N.L. & Rau, D.C. (1986) Osmotic stress for the direct measurement of intermolecular for- ces. Methods Enzymol. 127, 400–416. 45. Morgan, J.E., Verkhovsky, M.I. & Wikstrom, M. (1996) Observation and assignment of peroxy and ferryl intermediates in the reduction of dioxygen to water by cytochrome c oxidase. Biochemistry 35, 12235–12240. 46. Karpefors, M., Adelroth, P. & Brzezinski, P. (2000) The onset of the deuterium isotope effect in cytochrome c oxidase. Biochemistry 39, 5045–5050. 47. Mitchell, R. & Rich, P.R. (1994) Proton uptake by cytochrome c oxidaseonreductionandonligandbinding.Biochim. Biophys. Acta 1186, 19–26. 48. Capitanio, N., Capitanio, G., Minuto, M., De Nitto, E., Palese, L.L., Nicholls, P. & Papa, S. (2000) Coupling of electron transfer with proton transfer at heme a and Cu(A) (redox Bohr effects) in cytochrome c oxidase. Studies with the carbon monoxide inhibited enzyme. Biochemistry 39, 6373–6379. 260 J. A. Kornblatt et al. (Eur. J. Biochem. 270) Ó FEBS 2003 . concentrations of cytochrome c plus TMPD and ascorbate. As the concen- tration of cytochrome c is increased, the fraction of cytochrome a that is reduced increases. The fraction of cytochrome a. second catalytic cycle of reduction of CCO by cytochrome c. Mitchell and Rich [47] proposed that two protons were taken up on reduction of CCO and that these were taken up concomitantly with reduction. The influence of temperature and osmolyte on the catalytic cycle of cytochrome c oxidase Jack A. Kornblatt 1 , Bruce C. Hill 2 and Michael C. Marden 3 1 Enzyme Research Group, Concordia