Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase

FEBS Letters xxx (2015) xxx–xxx

journal homepage: www.FEBSLetters.org

Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase Artem V. Dyuba, Tatiana Vygodina, Natalia Azarkina, Alexander A. Konstantinov ⇑ A.N. Belozersky Institute of Physico-Chemical Biology, Moscow State University, Moscow 119 992, Russia

a r t i c l e

i n f o

Article history: Received 8 August 2015 Revised 15 November 2015 Accepted 16 November 2015 Available online xxxx Edited by Peter Brzezinski Keywords: Cytochrome oxidase Calcium Electron transfer Respiratory chain H proton channel

a b s t r a c t The effect of Ca2+ on the rate of heme a reduction by dithionite and hexaammineruthenium (RuAm) was studied in the cyanide-complexed bovine cytochrome oxidase (CcO). The rate of heme a reduction is proportional to RuAm concentration below 300 lM with kv of 0.53  106 M 1s 1. Ca2+ inhibits the rate of heme a reduction by dithionite by 25%. As the reaction speeds up with increased concentrations of RuAm, the inhibition by Ca2+ disappears. The inhibition of heme a reduction may contribute to recently described partial inhibition of CcO by Ca2+ in the enzymatic assays. The inhibitory effect of Ca2+ on heme a reduction indicates that ET through heme a may be coupled to proton movement in the exit part of the proton channel H. Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Cytochrome c oxidase (CcO) is a terminal component of the respiratory chain of mitochondria and many bacteria. The enzyme gates access of oxidative phosphorylation to oxygen as the final electron acceptor. CcO catalyzes electron transfer (ET) from ferrous cytochrome c to oxygen, and free energy of this thermodynamically favourable process is conserved in a form of transmembrane difference of proton electrochemical potential, DlH+ (reviewed, [1–3]). ET in CcO occurs through a series of redox carriers, two copper centers (CuA and CuB) and two hemes (a and a3): External electron donor (cyt c or an artificial donor) ? CuA ? heme a ? {heme a3/CuB} ? O2 The exergonic ET to oxygen is coupled to electrogenic translocation of protons across the membrane [4] via a system of proton channels as proposed originally in [5] and studied since then in great detail (reviewed, [1–3,6,7]). Operation of the proton pump

Author contributions: Artem V. Dyuba, planned and made experiments, processed and analyzed the data, discussed the paper. Tatiana Vygodina, planned and made experiments, discussed the paper; Natalia Azarkina, planned and made experiments, processed and analyzed the data, discussed the paper. Alexander A. Konstantinov, designed and planned the experiments, processed and analyzed the data, wrote the paper. ⇑ Corresponding author at: A.N. Belozersky Institute, Moscow State University, Leninskie Gory 1, Bld. 40, 4 Khokhlova Str., Room 428, 119 992 Moscow, Russia. E-mail addresses: [email protected], [email protected] (A.A. Konstantinov).

may be linked primarily to the steps of heme a reduction and oxidation [3,5,7–9]. Besides the redox metals, CcO contains bound non-redox metal ions, and, in particular, there is a special cation binding site (CBS) in subunit I of the enzyme [10–12] located near heme a and CuA. In CcO from vertebrate mitochondria, the site binds reversibly Ca2+ and Na+ [13–15] as well as some other metal cations, albeit with lower affinity [16]. Ca2+ interaction with the mitochondrial CcO was first described long ago [17] and was studied since then in considerable detail with both the mitochondrial CcO [13–15,18] and bacterial mutant oxidase [12,18–20]. However, functional significance of the cation binding remained unknown for a long time, until it was found that Ca2+ binding at the CBS may result in 2–3-fold inhibition of CcO activity, the inhibition being prevented by excess Na+ [16,21,22]. The mechanism of the Ca2+-induced inhibition of CcO is now to be disclosed. Interestingly, the cation binds at the exit part of the transmembrane ‘‘proton channel H”, associated with heme a [3,22]. Therefore, effect of Ca2+ on proton movement within the H-channel coupled to redox reactions of heme a is a tempting hypothesis [3,16,22]. In order to establish the inhibition mechanism, it is necessary first of all to reveal the steps of intramolecular ET affected by Ca2+ binding. It was reported earlier that Ca2+ slows down electron transfer from heme a to heme a3 in bovine heart CcO [23]. The inhibition of ET between the hemes a and a3 has been recently confirmed and characterized in more detail by this group using dithionite, ascorbate and different concentrations of ruthenium

http://dx.doi.org/10.1016/j.febslet.2015.11.023 0014-5793/Ó 2015 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

Please cite this article in press as: Dyuba, A.V., et al. Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase. FEBS Lett. (2015), http://dx. doi.org/10.1016/j.febslet.2015.11.023

2

A.V. Dyuba et al. / FEBS Letters xxx (2015) xxx–xxx

hexaammine (RuAm) complex, [RuII/III(NH3)6]2+/3+ [24] as the reductants (paper in preparation). In the course of these experiments, we noticed that under some conditions Ca2+ appeared to slow down not only the oxidation, but also the reduction of heme a. However, the kinetics of absorption changes in ligand-free CcO is complicated by contribution from the two hemes, a and a3, which renders interpretation of the data less certain. In order to better resolve potential effect of Ca2+ on heme a reduction, we choose to simplify the experimental model and to explore the reaction with the cyanide complex of CcO, a3+a3+ 3 -CN. In this complex heme a3 is trapped by cyanide in a ferric state so that ET from heme a to heme a3 is blocked and contribution of heme a3 spectral changes is eliminated. In this work the kinetics of heme a reduction has been studied in the cyanide-complexed CcO, a3+a3+ 3 -CN, with the aid of a stopped-flow diode array spectrophotometer. Dithionite was used as the electron donor in the absence or in the presence of different concentrations of RuAm. Ca ions are found to bring about modest but distinct partial inhibition of heme a reduction by dithionite or RuAm. The inhibition is lifted as the electron input rate increases in the presence of high concentrations of RuAm. 2. Materials and methods 2.1. Chemicals Sodium dithionite, calcium chloride and bovine liver catalase were from Sigma–Aldrich. Hexaammineruthenium (III) chloride was purchased from Alfa-Division. Dodecyl maltoside (SOLGRADE) was a product of Anatrace. pH buffers, EGTA and magnesium sulfate were from Amresco. 2.2. Preparation Cytochrome c oxidase was prepared from bovine heart mitochondria by a method of Fowler et al. [25] modified by Dr. A. Musatov (Institute of Experimental Physics Slovak Acad. Sci, Kosice) as described in [16]. The basic medium contained 100 mM HEPES/Tris buffer, pH 8, and 0.1% dodecyl maltoside with other additions as indicated. Care was taken to avoid Na+ in the solutions (except for using sodium dithionite, Na2S2O4, the only dithionite salt available) because Na+ competes with Ca2+ for the CBS in CcO [14,15]. Cyanide complex of CcO was obtained by incubation of the enzyme (24 lM) in the basic buffer with 2 mM KCN in a closed tube for 20 h at 4 °C. The reaction with cyanide was essentially complete (De434–412 ca. 50 mM 1cm 1), but a minor shoulder at 444 nm was present indicating partial reduction of heme a. Complete oxidation of heme a was then achieved by addition of 40–60 lM ferricyanide. 2.3. Measurements Static spectra were recorded in a CaryBio 300 spectrophotometer (Varian). The kinetics of CcO reduction was studied in an Applied Photophysics SX-20 stopped-flow spectrophotometer operated in a diode array mode. The instrument allows to record the spectra (190–720 nm range) in a 20 lL cell with 1 cm optical pathway with a minimal interval between the spectra of 1 ms. Unless indicated otherwise (cf. Fig. 3), one syringe contained 12 lM aerobic oxidized cyanide complex of CcO (a3+a3+ 3 -CN) in the basic buffer with 100 lM EGTA and 1 mM KCN and supplemented where indicated with 0.5 mM CaCl2 or 0.5 mM MgSO4. The second syringe contained 10 mM sodium dithionite in the same 100 mM HEPES/Tris basic buffer, pH 8, with 100 lM EGTA, 8 nM catalase (to remove H2O2 formed upon reaction of dithionite

with oxygen) and appropriate concentrations of RuAm. Hereafter, final concentrations after mixing are referred throughout the text. These were for the main reactants: CcO, 6 lM; sodium dithionite, 5 mM; catalase, 4 nM; EGTA, 100 lM and, where indicated, Ca2+ or Mg2+, 250 lM (i.e., 150 lM excess over EGTA). RuAm concentration was varied between 0 and 3000 lM. In Figs. 1 and 2 spectral baseline was measured by mixing blank solution in syringe 1 (the basic buffer with 100 lM EGTA), with blank solution in syringe 2 (10 mM sodium dithionite, 100 lM EGTA, and 8 nM catalase in the basic buffer without dodecyl maltoside). To prevent dithionite consumption by oxygen, syringe 2 was bubbled with argon prior to addition of pre-weighed dithionite powder and kept plugged. In the experiments shown in Fig. 3, the baseline was measured without dithionite in syringe 2. This allowed us to monitor dithionite radical concentration directly in the cell after mixing (the peak at 315 nm) and check its stability.

2.4. Data processing The spectra were collected in a 190–720 nm nominal range, 1000 scans for the observation period with minimal interval between the points of 1 ms and logarithmic spread of the time points where the observation period exceeded 1 s. To determine the rate constants, global analysis of the spectra/time surface files was made with ProKineticist software provided with the APL SX-20 instrument. The Soret and a-band parts of the files were excised with MATLAB (MathWorks, Inc.) and subjected to global analysis in ProKineticist program separately in order to eliminate influence of the extensive Soret band changes on the visible part data fitting. Alternatively, the kinetic curves at selected wavelengths were extracted from the spectra/time surface files and DA444–480 or DA605–630 curves were fitted to 1 or 2 exponentials with MATLAB or Origin 7E (Microcal) software. In general, the results obtained by different procedures of analysis were in good agreement. Single exponential approximation was sufficient to fit the kinetics in the a-band, at least for the purposes of this work (e.g., see Supplementary Data Fig. S1A–C). The changes in the Soret revealed sometimes minor contribution of the slower changes.

3. Results Fig. 1A and B shows typical sets of spectra evolving after mixing the a3+a3+ 3 -CN complex with 5 mM dithionite and 10 lM RuAm (final concentrations). The reduction is complete in less than 2 s. The spectra sets reveal clear isobestic points in the c- and a-absorption bands of the hemes. In particular, the isobestic point at 437.5 nm persists in the Soret during the entire observation period confirming that heme a reduction is the only obvious process observed (Fig. 1A). This is at variance with the data obtained in the absence of cyanide, in which the initial isobestic point at 432 moves to 426 nm at the later stages of the reduction [23]. No evidence for significant reduction of the cyanide complex of heme a3 was obtained during the observation period of up to 20 s. Interestingly, the spectral changes in the b-region of the hemes are not homogenous and no unique isosbestic points are retained below the one at 585 nm (Fig. 1B). Development of the narrow heme peaks at 520 and 555 nm is overlaid by bleaching of absorption between the isobestic points at 585 and 463.5 nm with a minimum in the difference spectrum around 480 nm (cf. [26,27]). Identity of this band in the absolute spectra of the oxidized ligandfree and cyanide-complexed CcO [26,28] is not quite certain. According to the data in [29,30], it could belong to the oxidized CuA (cf. a difference spectrum of CuA obtained for R54M mutant

Please cite this article in press as: Dyuba, A.V., et al. Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase. FEBS Lett. (2015), http://dx. doi.org/10.1016/j.febslet.2015.11.023

3

A.V. Dyuba et al. / FEBS Letters xxx (2015) xxx–xxx

430

A

603.5

B

437.5 443

time

time

0,10

0,4

absorption

absorption

0,5

0,3 time

0,2

463.5

555

585

521 time 565 596 575

0,05

621.5 642

0,1 0,0

400

425

450

475

500

0,00

560

wavelength, nm

630

700

wavelength, nm

Fig. 1. Evolution of spectral changes during the reduction of the oxidized cyanide complex of CcO. 12 lM oxidized CcO-CN complex was mixed with dithionite and RuAm in a buffer with 100 lM EGTA. For detailed description, see Section 2. The first spectrum, 1 ms after mixing; the last spectrum shown, 2000 ms. 1000 spectra collected. One of each 50 spectra is shown in the figure. (A) Soret band; (B) visible part. Dithionite, 5 mM; RuAm, 10 lM.

600

c f

RuAm, 10 μΜ

rate constant, s

e

0,05

d a

no RuAm -1

b

1.4 s

0.89 s

-1

0,1

0,2

0,3

kv (Ca) / kv (control)

400 300

kv = 0.53 x 106 M-1s-1

200

0 0

0,4

500

1000

1500

2000

2500

3000

[RuAm], μ M

time, s

1,00

500

100 Ca EGTA

0,00 0,0

B

700

RuAm, 30μΜ

-1

Δ A at 605-630 nm

A

C

0,75

0,50

0,25

0,00 0

250

500

750

1000

[RuAm], μM 2+

Fig. 2. Effect of RuAm and Ca on the rate of heme a reduction in the cyanide complex of CcO. (A) Representative kinetic traces of heme a reduction by 5 mM dithionite and the indicated concentrations of RuAm. The kinetic curves at 605 minus 630 nm were constructed by extracting the traces at 605 and 630 nm from the overall files. Traces (a, c, e), reaction in the presence of 100 lM EGTA. Traces (b, d, f), reaction in the presence of 100 lM EGTA and 250 lM Ca2+. (B) Dependence of the rate of heme a reduction on concentration of RuAm. Experiments like that shown in Fig. 1 were performed at RuAm concentrations increasing from 0 to 3000 lM. Pseudo-first order rate constants for heme a reduction were determined by global analysis of the spectra/time surface files in the visible range (500–720 nm) assuming a single exponential transition (model a > b in the ProKineticist program). The data from several series of measurements performed in the presence of 100 lM EGTA (filled circles), 100 lM EGTA + 250 lM Ca2+ (open circles) and 100 lM EGTA + 250 lM Mg2+ (squares) are included in the figure. The kv values obtained at 3 mM RuAm are very approximate (as only the last 5% of the reaction is resolved, see the text and Supplementary Data, Fig. S1D) and therefore are given in brackets, just as a guide to the workers who may wish to further explore the reaction. (C) Dependence of the inhibition of heme a reduction by Ca2+ on RuAm concentration. The data in Fig. 2B replotted to show the effect of Ca2+. The rate constants in the presence of Ca2+ were measured with 250 lM Ca2+ on a background of 100 lM EGTA. The control values correspond to experiments in the presence of 100 lM EGTA (squares) or 100 lM EGTA + 250 lM Mg2+ (circles). The line has been drawn through the points to guide the eye.

of CcO from Paracoccus denitrificans [31] and a spectrum of a model CuA protein [32]). Reduction of heme a by dithionite goes much faster in the presence of RuAm (Fig. 2A and B). The apparent pseudo-first order rate constant was around 1 s 1 at zero concentration of RuAm and increased to  600 s 1 at 3 mM of the compound. Above 10 lM of RuAm, direct reduction of CcO by 5 mM dithionite contributes but little to the observed reaction rate. The plot of the effective rate constant vs concentration of RuAm is linear up to 300 lM in

agreement with [24] but tends to saturate or, rather, change slope above 500 lM of RuAm (Fig. 2B). At 1 mM RuAm, the recordings catch the last 20% of the reduction process, sufficient to determine accurately the effective rate constant. However, at 3 mM of RuAm, only the last 5% are recorded and there are just 7–10 points before the final level is reached (e.g., Supplementary Data, Fig. S1D). Therefore, the rate constant values determined by global analysis at 3 mM RuAm may be considered only as provisional as indicated by brackets in the figure panel. The points are included

Please cite this article in press as: Dyuba, A.V., et al. Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase. FEBS Lett. (2015), http://dx. doi.org/10.1016/j.febslet.2015.11.023

A.V. Dyuba et al. / FEBS Letters xxx (2015) xxx–xxx

Δ A at 605-630 nm

4

0,03

A + RuAm 2

0,02

3 2+

+ Ca after RuAm 0,01

dith 1 0,00 0,0

0,5

1,0

1,5

2,0

2,5

3,0

Δ A at 605 - 630 nm

time, s

B 0,02

dith 1 2

+ Ca2+

0,01

4. Discussion

0,00 0

10

20

30

40

100 150

time, s 0,03

ΔA at 605-630 nm

the plot is 0.53  106 M 1 s 1, in very good agreement with the value of 0.5  106 M 1 s 1 reported for the rapid phase of free CcO reduction by Scott and Gray [24]. Detailed account on the kinetics of heme a reduction in the free and cyanide-inhibited CcO by dithionite and RuAm will be reported elsewhere (paper in preparation). Effect of calcium ions on the rate of heme a reduction by dithionite or dithionite + RuAm was studied. In the absence of RuAm, Ca2+ was found to bring about modest but distinct deceleration of heme a reduction by dithionite (Fig. 2A, traces a, b). The inhibition was fully prevented by 50 mM Na+ that competes with calcium for the specific binding site [15,18] (e.g., Fig. 3C and see also Supplementary Data, Fig. S3). This important control supports specificity of the inhibitory effect of Ca2+. No inhibition but rather a slight stimulation was observed with Mg2+ added instead of Ca2+ (data not included). The inhibition by Ca2+ reproduced in several independent measurements in Fig. 2 corresponds on the average to 25 ± 5% (maximal scatter, 19–37%) which is similar to the inhibition of the a-to-a3 electron transfer rate by Ca2+ (20–30%) reported in [23]. Interestingly, we found that as the reduction of heme a becomes faster in the presence of increasing concentrations of RuAm, the inhibition by calcium disappears (Fig. 2A and C). Possible reasons for this effect are discussed below.

C

dith (black) 1

0,02

2

2+

+ Ca (red)

0,01

in the presence of 50 mM NaCl 0,00 0

10

20

30

40

50

time, s Fig. 3. Ca2+ inhibits heme a reduction by RuAm at low rate of electron input. Experiments were made essentially as those in Fig. 2 with the following changes in the conditions. Dithionite concentration was lowered to 0.4 mM; the buffer in both syringes, the syringe compartment of the instrument and all the reactants to be added were flushed thoroughly with argon to prevent oxidation of added dithionite. CcO concentration was 3 lM. All the concentrations are final after mixing. Stability of dithionite concentration was checked in the same spectra sets by a peak at 315 nm. Panels A, B and C correspond to 3 separate experiments. Logarithmic spread in time of the spectra acquisition points has been used. Since Ca2+ brings about red shift of heme a spectrum, it can affect slightly the magnitude of the response, depending on a particular wavelength. Therefore, the traces at 605– 630 nm in the presence of EGTA and Ca2+ have been normalized to each other (correction by 5%). (A). Trace 1, kinetics of heme a reduction by 0.4 mM dithionite in the presence of 100 lM EGTA. Trace 2, as 1 but after addition of 1.5 lM RuAm. Trace 3, as 2, but after addition of 250 lM Ca2+ (150 lM excess over EGTA). The small lag phase in the curves with RuAm is presumably due to the rapid initial reaction of RuAm with the small amount of ferricyanide used to prevent autoreduction of the aa3-CN complex (see the Methods). (B). Trace 1, reduction of heme a by 0.4 mM dithionite in the presence of 100 lM EGTA. Trace 2, as 1 but after addition of 250 lM Ca2+ (150 lM excess over EGTA). (C) As (B), but in the presence of 50 mM NaCl. The traces with EGTA (black) and Ca2+ (red) coinside.

because, though provisional, they can serve a useful guide for the workers who may wish to make experiments on this issue. The apparent second order rate constant for the initial linear part of

In order to reveal possible effect of Ca2+ on the kinetics of heme a reduction, we have used cyanide-bound ferric CcO as a relatively simple experimental model with dithionite and ruthenium hexaammine complex, [Ru(NH3)6]2+/3+, as the electron donors. Dithionite is a traditional reductant widely used in the CcO kinetics studies [27,33–36]. RuAm was introduced as an efficient electron donor to CcO by Scott and Gray [24] and since then proved to be a useful redox mediator in several CcO studies [23,37–42]. The compound is a very efficient electron donor to CcO competing with cytochrome c for the binding site at CcO [37] and its oxidation by CcO is coupled to generation of membrane potential [39], so oxidation of RuAm most probably goes the physiological route. The kinetics of heme absorption changes during reduction of free ferric cytochrome oxidase is not elementary. Reduction of purified CcO by either dithionite or RuAm is at least biphasic, the rapid phase associated with heme a reduction and the slow phase with ET from heme a to heme a3 (e.g., [24,34,38]). Moreover, the slow phase itself can be heterogenous [41,42] due to heterogenous state of the binuclear site in the oxidized CcO. Finally, reduction of heme a3 shifts redox midpoint potential of heme a to the negative and, accordingly, may affect the kinetics of heme a reduction. Accordingly, rigorous analysis of the kinetics pattern of absorption changes in the Soret upon mixing of the free CcO with dithionite and RuAm is rather complicated. In order to better evaluate a rather small effect of Ca ions on the kinetics of heme a reduction, we choose to simplify experimental model by eliminating spectral contribution of heme a3. To this end, heme a reduction was measured in the aa3-CN complex of the enzyme, and the kinetics was analyzed primarily in the a-band where contribution of heme a3 is small. Under these conditions, the kinetics of heme a reduction by dithionite or dithionite + RuAm is essentially monophasic (e.g., Supplementary Data Fig. S1A–C). This is at variance with the original paper [24] in which a large second slow phase of absorption changes was observed in the cyanide-complexed CcO with the effective rate constant independent of RuAm concentration. We did not observe such a phase in our experiments. Presumably, cyanide binding by the ferric CcO in [24] could be incomplete. At the same time, the second order rate constant of CcO reduction by

Please cite this article in press as: Dyuba, A.V., et al. Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase. FEBS Lett. (2015), http://dx. doi.org/10.1016/j.febslet.2015.11.023

A.V. Dyuba et al. / FEBS Letters xxx (2015) xxx–xxx

RuAm obtained in our work (0.53  106 M 1 s 1) is in very good agreement with that reported for the rapid phase of CcO reduction in [24]. 4.1. Inhibitory effect of calcium As found in this work, Ca2+ inhibits heme a reduction by dithionite by 25%. At the same time, it was reported earlier [23] that electron transfer from heme a to heme a3 in the unliganded bovine CcO was decelerated by Ca2+ by about the same value. Thus, calcium ions can inhibit the steps of electron transfer both to and from heme a. The inhibition of heme a reduction by Ca2+ proves to be maximal at low rates of electron input (ca. 1 s 1 or less), as observed for the reduction of CcO by dithionite with no RuAm. At similar low rates of electron input, the inhibition could also be observed in pilot experiments with ascorbate, a sluggish electron donor (Supplementary Data, Fig. S2). At the same time, the inhibition decreases in the presence of 10 lM RuAm and vanishes as the rate of the reduction increases to 100 s 1 at high concentrations of RuAm (Fig. 2A and C). This observation may be related to the earlier finding that the inhibition of CcO turnover by Ca2+ is maximal at low rates (few per second) of electron input [16,22]. However, a simple alternative explanation is possible, namely, that reduction of heme a by the anionic dithionite (or ascorbate) and by the cationic RuAm involves non-identical pathways, reduction by RuAm per se being not inhibited by Ca2+. Taking into account crystal structure of CcO, it is difficult to visualize how the enzyme can accept electrons via a redox center other than CuA, regardless of the charge of an electron donor. Nevertheless, it was important, as recommended by one of the reviewers, to test experimentally whether reduction of heme a by RuAm may be resistant to inhibition by Ca2+. To this end, concentrations of RuAm and dithionite were varied widely to select conditions at which the reduction of heme a by RuAm occurs slowly, but still very much faster than reduction by dithionite itself. Results of an experiment under such optimized conditions are presented in Fig. 3. As shown in Fig. 3A, reduction of heme a by 400 lM dithionite is very slow (Trace 1). Upon addition of 1.5 lM RuAm, the reaction rate increases about 30-fold (from 0.050 s 1 to 1.8 s 1, Trace 2), so that virtually all the reduction of heme a proceeds now via RuAm. Under these conditions, Ca2+ brings about distinct inhibition of heme a reduction by RuAm (Fig. 3A, Trace 3). About the same inhibition has been observed in this sample for the slow reduction of heme a by dithionite itself (Fig. 3B). In these experiments, made with a new batch of CcO, the calcium inhibition of heme a reduction by dithionite varied in the range 18–20% as observed in several probes with either 0.4 mM or 5 mM of dithionite, that is somewhat less than the average inhibition in Fig. 2 (25%). However, this modest inhibition was reproducible and it was fully prevented by 50 mM Na+ with either 0.4 mM dithionite (Fig. 3C) or 5 mM dithionite as the reductant (Supplementary Data, Fig. S3) supporting specificity of the inhibitory effect. These experiments show that the above described disappearance of the calcium inhibition upon increased concentrations of RuAm (Fig. 2) can hardly be explained by different reaction pathways of heme a reduction by dithionite and RuAm, the latter being insensitive to Ca2+. Rather, the extent of inhibition may depend indeed on the rate of electron input in accordance with the results of the turnover experiments [16,22]. Discussion of possible reasons of such a dependence would be premature at this stage of the research. The molecular mechanism of the inhibition exerted by Ca2+ on the redox reactions of heme a remains to be established. The inhibition of CcO turnover by Ca2+ is rather modest (ca. 50% in bovine heart CcO). Such an effect may be quite significant for physiological tuning of CcO turnover in mitochondria, but it makes it more

5

difficult to explore the mechanism of the effect. That Ca2+ is expected to inhibit proton transfer in the exit part of the ‘‘H-proton channel” was pointed out [22]. However, the specific role of the H-channel in CcO has not been established yet. This putative proton-conducting pathway may perform transmembrane proton transfer from inside to the outside (proton pumping) creating DlH+ [3,43], or conduct protons in the opposite, outsideto-inside, direction dissipating the DlH+ (a mild-uncoupling module built in CcO [18,20]) or, finally, it may not be a transmembrane proton channel at all, but just a pair of ‘‘dielectric wells” opening to the opposing sides of the membrane and facilitating ET through heme a [44]. Whatever the specific function, the inhibitory effect of Ca2+ on ET to heme a (this work) and from heme a to heme a3 [23] points out functional significance of the H-channel in bovine CcO. The inhibition of redox reactions of heme a by Ca2+ suggests that proton movement in the exit part of the H-channel may be important for facilitating electron transfer through heme a. As a minimal hypothesis, this fact would be in general agreement with the proposal of Rich and Marechal [44] that charge displacement within the H-channel can assist ET through heme a. A specific hypothetical mechanism can also be considered. There are indications that reduction of heme a and/or CuA is coupled to proton binding not only from the N-side [5] but also from the P-side of the membrane [45,46]. The redox-linked H+-binding from the P-phase is much faster than ET to heme a [46] and so is likely to be initiated by the reduction of CuA. As argued in [8], there is a cluster of amino acid residues in vicinity of the CuA/heme a domain coupled simultaneously to CuA and heme a (and cf. Refs. [47,48] for discussion of potential redox-linked proton movements coupled to CuA). The cluster accepts proton from the P-phase when CuA is reduced. Subsequent electron transfer from CuA to heme a may be accompanied by parallel proton displacement within the cluster closer to heme a, assisting reduction of the heme. It is possible that such proton displacement occurs within the exit part of the H-channel, so that its inhibition by Ca2+ results in deceleration of heme a reduction in bovine CcO. Inward displacement of H+ within the exit of the H-channel could help to explain diminished magnitude of the microsecond electrogenic phase coupled to vectorial CuA ? heme a ET in bovine CcO as compared to the bacterial CcO which lacks the H-channel exit [7]. Acknowledgements This work was supported in part by a grant No. 14-04-00425 from Russian Fund for Basic Research. The stopped-flow spectrophotometer was bought with the funds from HHMI International Research Scholar Award 55005615 to AAK. We would like to thank the reviewers of the paper for a number of useful suggestions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2015.11.023. References [1] Belevich, I. and Verkhovsky, M.I. (2008) Molecular mechanism of proton translocation by cytochrome c oxidase. Antioxid. Redox Signal. 10, 1–29. [2] Kaila, V.R.I., Verkhovsky, M.I. and Wikstrom, M. (2010) Proton-coupled electron transfer in cytochrome oxidase. Chem. Rev. 110, 7062–7081. [3] Yoshikawa, S. and Shimada, A. (2015) Reaction mechanism of cytochrome c oxidase. Chem. Rev. 115, 1936–1989. [4] Wikstrom, M. (1977) Proton pump coupled to cytochrome c oxidase in mitochondria. Nature 266, 271–273. [5] Artzatbanov, V.Y., Konstantinov, A.A. and Skulachev, V.P. (1978) Involvement of intramitochondrial protons in redox reactions of cytochrome a. FEBS Lett. 87, 180–185.

Please cite this article in press as: Dyuba, A.V., et al. Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase. FEBS Lett. (2015), http://dx. doi.org/10.1016/j.febslet.2015.11.023

6

A.V. Dyuba et al. / FEBS Letters xxx (2015) xxx–xxx

[6] Brzezinski, P. and Gennis, R.B. (2008) Cytochrome c oxidase: exciting progress and remaining mysteries. J. Bioenergy Biomembr. 40, 521–531. [7] Siletsky, S. and Konstantinov, A.A. (2012) Cytochrome c oxidase: charge translocation coupled to single-electron partial steps of the catalytic cycle. Biochim. Biophys. Acta 1817, 476–488. [8] Capitanio, N., Palese, L.L., Capitanio, G., Martino, P.L., Richter, O.M., Ludwig, B. and Papa, S. (2012) Allosteric interactions and proton conducting pathways in proton pumping aa3 oxidases: heme a as a key coupling element. Biochim. Biophys. Acta 1817, 558–566. [9] Egawa, T., Yeh, S.R. and Rousseau, D.L. (2013) Redox-controlled proton gating in bovine cytochrome c oxidase. PLoS ONE 8, e63669. [10] Ostermeier, C., Harrenga, A., Ermler, U. and 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. U.S.A. 94, 10547–10553. [11] Yoshikawa, S., Shinzawa-Itoh, K., Nakashima, R., Yaono, R., Inoue, N., Yao, M., Fei, M.J., Libeu, C.P., Mizushima, T., Yamaguchi, H., Tomizaki, T. and Tsukihara, T. (1998) Redox-coupled crystal structural changes in bovine heart cytochrome c oxidase. Science 280, 1723–1729. [12] Pfitzner, U., Kirichenko, A., Konstantinov, A.A., Mertens, M., Wittershagen, A., Kolbesen, B.O., Steffens, G.C.M., Harrenga, A., Michel, H. and Ludwig, B. (1999) Mutations in the Ca2+-binding site of the Paracoccus denitrificans cytochrome c oxidase. FEBS Lett. 456, 365–369. [13] Saari, H., Pentilla, T. and Wikstrom, M. (1980) Interaction of Ca2+ and H+ with heme A in cytochrome oxidase. J. Bioenergy Biomembr. 12, 325–338. [14] Mkrtchyan, H., Vygodina, T. and Konstantinov, A.A. (1990) Na+-induced reversal of the Ca2+-dependent red shift of cytochrome a. Is there a hydronium output well in cytochrome c oxidase? Biochem. Int. 20, 183–190. [15] Kirichenko, A., Vygodina, T.V., Mkrtchyan, H.M. and Konstantinov, A.A. (1998) Specific cation-binding site in mammalian cytochrome c oxidase. FEBS Lett. 423, 329–333. [16] Vygodina, T.V., Kirichenko, A. and Konstantinov, A.A. (2014) Cation binding site of cytochrome c oxidase: progress report. Biochim. Biophys. Acta 1837, 1188–1195. [17] Wikstrom, M. and Saari, H. (1975) A spectral shift of heme a induced by calcium ions. Biochim. Biophys. Acta 408, 170–179. [18] Kirichenko, A.V., Pfitzner, U., Ludwig, B., Soares, C.M., Vygodina, T.V. and Konstantinov, A.A. (2005) Cytochrome c oxidase as a calcium binding protein. Studies on the role of a conserved aspartate in helices XI–XII cytoplasmic loop in cation binding. Biochemistry 44, 12391–12401. [19] Riistama, S., Laakkonen, L., Wikstrom, M., Verkhovsky, M.I. and Puustinen, A. (1999) The calcium binding site in cytochrome aa3 from Paracoccus denitrificans. Biochemistry 38, 10670–10677. [20] Lee, A., Kirichenko, A., Vygodina, T., Siletsky, S.A., Das, T.K., Rousseau, D.L., Gennis, R.A. and Konstantinov, A.A. (2002) Ca2+-binding site in Rhodobacter sphaeroides cytochrome c oxidase. Biochemistry 41, 8886–8898. [21] Vygodina, T.V. and Konstantinov, A.A. (2010) Ca2+-induced inhibition of the mammalian cytochrome c oxidase. Biochim. Biophys. Acta 1797, 102. [22] Vygodina, T., Kirichenko, A. and Konstantinov, A.A. (2013) Direct regulation of cytochrome c oxidase by calcium ions. PLoS ONE 8, e74436. [23] Vygodina, T.V., Dyuba, A.V. and Konstantinov, A.A. (2012) Effect of calcium ions on electron transfer between hemes a and a(3) in cytochrome c oxidase. Biochemistry (Moscow) 77, 901–909. [24] Scott, R.A. and Gray, H.B. (1980) Cytochrome aa3 electron-transfer reactions. Kinetics of hexaammineruthenium(II) reduction of the beef heart enzyme. J. Am. Chem. Soc. 102, 3219–3224. [25] Fowler, L.R., Richardson, S.H. and Hatefi, Y. (1962) A rapid method for the preparation of highly purified cytochrome oxidase. Biochim. Biophys. Acta 64, 170–173. [26] Liao, G.-L. and Palmer, G. (1996) The reduced minus oxidized spectra of cytochromes a and a3. Biochim. Biophys. Acta 1274, 109–111. [27] Halaka, F.G., Babcock, G.T. and Dye, J.L. (1981) Kinetic distinction between cytochromes a and a3 in cytochrome c oxidase: rapid scanning stopped flow study of anaerobic reduction by a neutral and a negatively charged donor. J. Biol. Chem. 256, 1084–1087.

[28] Yonetani, T. (1960) Studies on cytochrome oxidase. I. Absolute and difference absorption spectra. J. Biol. Chem. 235, 845–852. [29] Greenwood, C., Hill, B.C., Barber, D., Eglinton, D.G. and Thomson, A.J. (1983) The optical properties of CuA in bovine cytochrome c oxidase determined by low-temperature magnetic-circular-dichroism spectroscopy. Biochem. J. 215, 303–316. [30] Thomson, A.J., Greenwood, C., Peterson, J. and Barrett, C.P. (1986) Determination of the optical properties of CuA(II) in bovine cytochrome c oxidase using magnetic circular dichroism as an optical detector of paramagnetic resonance. J. Inorg. Biochem. 28, 195–205. [31] Jasaitis, A., Backgren, C., Morgan, J.E., Puustinen, A., Verkhovsky, M.I. and Wikstrom, M. (2001) Electron and proton transfer in the arginine-54methionine mutant of cytochrome c oxidase from Paracoccus denitrificans. Biochemistry 40, 5269–5274. [32] Hwang, Y.I. and Lu, Y. (2004) PH-dependent transition between delocalized and trapped valence states of CuA center and its possible role in protoncoupled electron transfer. Proc. Natl. Acad. Sci. U.S.A. 101, 12842–12847. [33] Jones, G.D., Jones, M.G., Wilson, M.T., Brunori, M., Colosimo, A. and Sarti, P. (1983) Reactions of cytochrome c oxidase with sodium dithionite. Biochem. J. 209, 175–182. [34] Nicholls, P. and Wrigglesworth, J.M. (1988) Routes of cytochrome a3 reduction. The neoclassical model revisited. Ann. N.Y. Acad. Sci. 550, 59–67. [35] Malatesta, F., Sarti, P., Antonini, G. and Vallone, B. (1990) Electron transfer to the binuclear center in cytochrome oxidase: catalytic significance and evidence for an additional intermediate. Proc. Natl. Acad. Sci. U.S.A. 87, 7410–7413. [36] Orii, Y. (1998) Determination and novel features of the absolute absorption spectra of the heme a moieties in cytochrome c oxidase. J. Bioenergy Biomembr. 30, 47–53. [37] Hochman, J.H., Partridge, B. and Ferguson-Miller, S. (1981) An effective electron donor to cytochrome oxidase. Purification, identification, and kinetic characterization of a contaminant of ruthenium red, hexaammineruthenium II/III. J. Biol. Chem. 256, 8693–8698. [38] Reichardt, J.K. and Gibson, Q.H. (1982) Spectra of intermediates in oxidation and reduction of cytochrome c oxidase. J. Biol. Chem. 257, 9268–9270. [39] Tsofina, L.M., Liberman, E.A., Vygodina, T.V. and Konstantinov, A.A. (1986) Hexaammineruthenium as an electron donor to mitochondrial cytochrome oxidase: membrane potential generation in the absence of cytochrome c. Biochem. Int. 12, 103–110. [40] Forte, E., Barone, M.C., Brunori, M., Sarti, P. and Giuffre, A. (2002) Redox-linked protonation of cytochrome c oxidase: the effect of chloride bound to CuB. Biochemistry 41, 13046–13052. [41] Antalik, M., Jancura, D., Palmer, G. and Fabian, M. (2005) A role for the protein in internal electron transfer to the catalytic center of cytochrome c oxidase. Biochemistry 44, 14881–14889. [42] Jancura, D., Antalik, M., Berka, V., Palmer, G. and Fabian, M. (2006) Filling the catalytic site of cytochrome c oxidase with electrons. J. Biol. Chem. 281, 20003–20010. [43] Yoshikawa, S. (2003) A cytochrome c oxidase proton pumping mechanism that excludes the O2 reduction site. FEBS Lett. 555, 8–12. [44] Rich, P.R. and Marechal, A. (2013) Functions of the hydrophilic channels in protonmotive cytochrome c oxidase. J. R. Soc. Interface 10, 20130183. [45] Mitchell, R. and Mitchell, P. (1988) The redox midpoint potential of haem a in cytochrome c oxidase is sensitive to extramitochondrial pH. Biochem. Soc. Trans. 17, 892–894. [46] Kirchberg, K., Michel, H. and Alexiev, U. (2012) Net proton uptake is preceded by multiple proton transfer steps upon electron injection into cytochrome c oxidase. J. Biol. Chem. 287, 8187–8193. [47] Sharpe, M.A. and Ferguson-Miller, S. (2008) A chemically explicit model for the mechanism of proton pumping in heme-copper oxidases. J. Bioenergy Biomembr. 40, 541–549. [48] Sugitani, R. and Stuchebrukhov, A.A. (2009) Molecular dynamics simulation of water in cytochrome c oxidase reveals two water exit pathways and the mechanism of transport. Biochim. Biophys. Acta 1787, 1140–1150.

Please cite this article in press as: Dyuba, A.V., et al. Calcium ions inhibit reduction of heme a in bovine cytochrome c oxidase. FEBS Lett. (2015), http://dx. doi.org/10.1016/j.febslet.2015.11.023