Journal of Photochemistry

and Photobiology,

B:

Biology, 5 (1990) 41 - 41

41

EFFECT OF REPLACING THE PRIMARY QUINONE BY DIFFERENT SPECIES ON THE ULTRAFAST PHOTOSYNTHETIC ELECTRON TRANSFER IN BACTERIAL REACTION CENTRES S. L. LOGUNOV, P. P. KNOX, N. I. ZAKHAROVA, V. Z. PASCHENKO and A. A. KONONENKO Department

of Biology,

M. V. Lomonosov

(Received January 6,1989;

Keywords. Rhodobacter absorption spectroscopy

State

B. N. KORVATOVSKY,

University,

Moscow

119899

(U.S.S.R.)

accepted April 28,1989)

sphaeroides,

quinone,

bacterial reaction centre,

Using picosecond absorption spectroscopy it has been shown that in Rhodobacter sphaeroides reaction centres the substitution of the primary quinone acceptor (QJ, ubiquinone-10, by other quinone species (with redox potentials higher or lower than that of ubiquinone-10) has essentially no modifying effect on the reaction centre protein. The molecular relaxation processes that accompany the localization and stabilization of a photoexcited electron on the intermediate acceptor, bacteriopheophytin (I), are not affected, although the subsequent transfer of the electron from I to QA is slowed down. Consequently, this leads to a lower quantum efficiency of charge separation between the photoactive pigment and QA. Therefore the high rate of direct I- + QA reaction is normally due to the specificity of the primary quinone species and its binding site in the reaction centre protein which provide optimum steric and chemical conditions for an effective interaction between I and QA.

1. Introduction The primary events of bacterial photosynthesis leading to charge separation occur in the reaction centre (RC), a polypeptide complex having a molecular weight of 100 kDa and associated with electron-transfer cofactors. Light activates bacteriochlorophyll dimer (P), causing a fast (r = 3 - 6 ps) transfer of a photoexcited electron from P to bacteriopheophytin (I), and subsequently to the primary quinone acceptor (QA, T = 150 - 200 ps) [l 5 1. The backward electron-transfer reactions are several orders of magnitude slower. Owing to this, the efficiency of charge separation is nearly 100%. Elsevier Sequoia/Printed in The Netherlands

P*I Q,

T

Kl

I 2

KR1

Fig. 1. Electron-transfer transitions in the Rb. sphaeroides reaction centre (data of refs. 6 - 8). Arrows show processes leading to a decrease in the free energy of the system following light excitation. Subscripts i and r denote the initial state of the P+T pair and the state after relaxation respectively (for further details see text).

This pattern of correlation between the direct and backward reactions is ensured by efficient relaxation processes in the RC which lead to the rapid lowering of the free energy in the system after the photoexcited electron has been localized on I and optimum conditions for its transfer to QA have been created (the sequence of events leading to the separation of the positive and negative charges between P and QA is shown in Fig. 1). Disturbances to the relaxation processes make the electron stabilization on I less efficient. Consequently, the subsequent electron-transfer reaction slows and recombination occurs between P+ and I- in the non-relaxed state. Such changes in the stabilization dynamics of the intermediate P+Istate take place after HZ0 substitution by *H,O, exposure to organic solvents or other effects which interfere with the hydrogen bonds and bound water in the macromolecular RC complex [6 - 91. It is not known whether electron localization and stabilization on I are dependent on the type of species that acts as the primary non-porphyrine acceptor. However, it is known that the quantum efficiency of charge separation between P and QA and the backward electron transfer from QA- to P+ depend on the chemical nature of QA [lo, 111. In an attempt to elucidate how different QA species influence the kinetics of direct electron transfer from P to I, and subsequently to QA, we have carried out picosecond measurements of the electrontransfer reactions in RCs containing different quinone species as acceptor. 2. Materials and methods Rhodobacter sphaeroides (wild type) RCs, isolated by the method described in ref. 12, were used in the experiments. Quinone extraction was

43

performed as described in ref. 13. Isolated RCs were incubated in a mixture containing 4% lauryldimethyl-amine N-oxide (LDAO) and lo-* M o-phenanthroline at 18 - 20 “C for 20 h in the dark. The samples were placed in a hydroxylapatite column and were washed with 10 mM sodium phosphate buffer (pH 7.0). They were then eluted with 0.15 M sodium phosphate buffer containing 0.05% LDAO. This procedure leads to a QA (ubiquinone-10, Uq,,) extraction of approximately 85%. To reconstitute the RCs, exogenous quinones were added in tenfold excess and the mixture was sonicated for 2 - 3 min. It was verified that the quinones were incorporated into the RCs by monitoring changes in the kinetics of P+ dark reduction by QA-. In reconstituted RCs of ubiquinone-9 (Uqs) the kinetics were similar to those of the control RCs containing Uq,,. In both types of sample, the half-time of the P’-Q,recombination reaction was 60 - 70 ms. Reconstitution with tetramethylbenzoquinone (duroquinone, Dq) and anthraquinone (Aq) caused changes in the P+ recovery kinetics, namely the appearance of slower components in Dq RCs and faster components in Aq RCs, consistent with the observations reported in refs. 10 and 11. A programmable picosecond absorption spectrometer was used to record fast electron transfers within the PIQA system. Provision was made for pulse quality control by identifying pulses whose duration or energy were beyond pre-selected variances. A pulse width of approximately 30 ps was used. The pulse density, at a pulse repetition frequency of 2 Hz, was 1015 quanta cm-*.

3. Results and discussion Tables 1 and 2 show the kinetic parameters of P+II decay and P+ dark recovery on the subnanosecond and nanosecond time scale in RC preparations with and without quinones, and in RCs reconstituted with quinones of different types. Absorption changes associated with the P+I- decay were measured at 665 nm. The decrease in 665 nm absorbance following light excitation can be described by the expression A&&t)

= Aa exp(-t/ra)

+ A I ew(-t/7d

The kinetics of P+ dark reduction were measured at 870 nm and can be approximated as A&&t)

= A2 exp(-t/r*)

+ As exp(--t/rs)

+ A4

The relative amplitude A4 is a constant absorbance level towards which the photoinduced signal decays during the time of observation. In other words, A4 = 1 means that P+ reduction occurs in a time much greater than the observation period (approximately 50 ns). It can be seen from Tables 1 and 2 that, in initial RCs, the P+I- state decays for r. = 0.2 ns, the time of electron transfer to QA [l - 81. The amplitude of the 870 nm band at the onset of bleaching (reflecting the formation of P’) does not change during

44 TABLE 1 Relative amplitudes A and decay times 7 of the components of P+T decay in the RCs from wild-type Rb. aphoeroides before and after extraction of quinones and after reconstitution with different exogenous quinones (h,, = 665 nm)

Initial RCs (Uqio) Initial RCs (Uqie)” RCs depleted of quinone Quinone-free RCs reconstituted with Uqs Quinone-free RCs + Dq Quinone-free RCs + Aq Vhe ml-‘)

Ao

AI

TOb)

71 (ns)

1.0 f 0.05 0.3 + 0.05 0.15 f 0.05 1.0 ? 0.05

0.7 + 0.05 0.85 f 0.05 -

0.2 0.7 0.2 0.2

10 + 1.0 10 ?r 1.0 -

0.15 + 0.05 0.15 f 0.05

0.85 f. 0.05 0.85 f 0.05

0.2 f 0.03 0.2 + 0.03

f 0.03 f 0.03 ?r 0.03 f 0.03

0.4 f 0.05 0.9 f 0.05

superscript (-) represents quinones reduced by dithionite (approximately and sodium ascorbate ( 10e3 M).

1 mg

TABLE 2 Relative amplitudes A and decay times T of the kinetic components of RC bacteriochlorophyll (P”) dark reduction in RCs from wild-type Rb. sphaeroides before and after extraction of quinones and after reconstitution with different quinones (&,.,, = 870 nm)

A2

Initial RCs (Uq ie) Initial RCs (Uqie)” RCs depleted of quinone Quinone-free RCs reconstituted with Uqg Quinone-free RCs reconstituted with Uq, a Quinone-free RCs + Dq Quinone-free RCs + Dq-a Quinone-free RCs + Aq Quinone-free RCs + Aqea

A3

A4

72 (n3)

73 (ns)

0.3 + 0.05 -

-

1.0 f 0.05 0.15 + 0.05 0.35 + 0.05 1.0 + 0.05

-

-

0.55 + 0.05 0.65 f 0.05 -

0.8 f 0.08 -

11 + 1.0 10 + 1.0 -

0.3 f 0.05

0.55 + 0.05

0.15 f 0.05

0.8 f 0.08

11 f 1.0

0.3 f 0.05 0.1 f 0.05 0.3 f 0.05

-

1.0 f 0.15 0.9 f 0.15

-

-

0.8 f 0.08 0.8 + 0.08 0.8 + 0.08

9.5 f 1.0 10 f 1.0

0.55 * 0.05 0.55 + 0.05

0.05 + 0.05 0.05 + 0.05

aThe superscript (-) represents a chemically reduced state.

the observation period because the P’-Q,decay time is of the order of tens of milliseconds. For Uq, RCs, the electron-transfer parameters are the same as for the initial RCs. This indicates that the extraction procedure does not affect, to any noticeable extent, the electron-transfer processes in the RCs. Reconstitution with Dq or Aq, instead of Uq,, results in a slower P+II decay (0.4 ns and 0.9 ns respectively) (Table 1). In Aq RCs, together with the slower P+I- decay (0.9 ns) (pi, see Table l), a new component with a

45

lifetime of about 0.8 ns (T*) and relative amplitude of about 0.1 appears in the P+ dark recovery kinetics at 870 nm (Table 2). This component is probably associated with the decay of the P+I-Q* state via channel Ka, (see Fig. 1) because of the slower direct electron transfer to QA (rate constant Ks): r2 = l/(K,* + Ks). Since the constant KnZ is known from measurements for RCs depleted of quinone QA (KR, = l/ra, Table 2), then K3 = lo9 s-i and the relative amplitude of the 72 kinetic component is A2 = Kn,I(Ka, + Ks) = 0.1. In RCs containing duroquinone as QA, the slower I- + QA reaction (0.4 ns) does not cause the appearance of the P+ reduction component reflecting the P+-II recombination via channel KR2. Indeed, its contribution is KR2/ (KS + KR,) x 0.04, which is beyond the sensitivity of the measuring equipment. Therefore the substitution of Uq by Aq or Dq does not cause any noticeable change in the relaxation behaviour of the radical pair. However, the substitution of Uq by Aq or Dq significantly slows the subsequent I- --f QA electron transfer. This makes the charge separation process in Aq RCs slightly less efficient (by lo%), consistent with that observed in refs. 10 and 11. Electron localization and stabilization on I are independent of the type of QA species as is evident from the kinetics of the electron-transfer processes in the RCs in which QA has been chemically reduced (Tables 1 and 2). As can be seen, the kinetics are virtually analogous irrespective of the QA species. The kinetic component with an amplitude A2 = 0.3 and a lifetime r2 of 0.8 ns reflects the decay of the non-relaxed state P’I- [6 - 81. In this case, r2 = (K, + KR,), where K2 is the rate constant of the relaxation process; its value becomes smallest presumably when the quinone QA is reduced. The appearance of this component in RCs containing reduced QA may be a reflection of the presence of a carotenoid molecule in the RC structure, as the data of ref. 8 suggest. The change in the rate of I- --f QA transfer in RCs reconstituted with different quinone species may, in principle, be ascribed to the different redox potentials of the quinones. The redox potentials of Aq, Uq and Dq (acting as QA) have been estimated in Rb. sphaeroides RCs as -0.21 V, -0.05 V and 0.02 V respectively [ 141. Hence, in Fig. 1, the P+IAqenergy level will be higher and the P+IDq- level will be lower than that of P+IUq-. However, in both Aq and Dq the I- + QA reaction is slower. This means that in addition to the mutual location of the energy levels of P+I-QA and P+IQ*-, there are other phenomena which influence the rate of I- +=QA electron transfer. These may be a change in the overlapping of the wavefunctions of P+I-Q, and P+IQ*- [15] or different geometries (orientation and distance to I) of quinone binding at the QA locus. As predicted by simple theoretical electron-transfer models [15,16], the changes observed in the rate of I- + QA reaction when naturally occurring quinone is substituted by other species, may result from a change in the distance between the two carriers of only approximately 1 A in Dq RCs and approximately 2 A in Aq RCs. The occurrence of such changes is fairly reasonable because Dq and Aq are very different from Uq sterically and chemically and may have different binding sites in the RC structure.

46

4. Conclusions Substitution of the natural quinone acceptor Uq,, in reaction centres of by other quinone types with higher or lower redox potentials does not modify the RC protein. The molecular relaxation processes that accompany the localization and stabilization of a photoexcited electron on the intermediate acceptor bacteriopheophytin (I) are not affected, although the subsequent transfer of the electron from I to QA becomes slower. Consequently, this leads to a lower quantum efficiency of charge separation between P and QA. It is important to emphasize that the decrease in the quantum efficiency is not a result of changes in the electron stabilization process on I (which takes place in deuterated RCs and in RCs in which the protein has been modified by different exposures). It is a consequence of the competitive P’-I- recombination process in the (P+I-Q,), system which has undergone relaxation and in which the I- + QA reaction is slower. Normally, the high rate of this reaction is due to the specificity of the primary quinone species and its binding site in the RC protein which provide optimum steric and chemical conditions for an effective interaction between I and QA.

Rb.

sphaeroides

References 1 D. Holten, C. Hoganson, M. W. Windsor, C. C. Schenk, W. W. Parson, A. Migus, R. L. Fork and C. V. Shank, Biochim. Biophys. Acta, 592 (1980) 461- 477. 2 C. Kirmaier, D. Holten and W. W. Parson, FEBS Lett., 185 (1985) 76 - 81. 3 V. Z. Paschenko, L. A. Avramov, S. S. Vasilev, V. V. Gorokhov, B. N. Korvatovsky, V. B. Tusov, A. A. Kononenko, N. P. Grishanova, N. I. Zakharova, S. K. Chamorovsky, and A. B. Rubin, Dokl. Akad. Nauk SSSR, 285 (1985) 1004 - 1008. 4 V. Z. Paschenko, B. N. Korvatovsky, A. A. Kononenko, S. K. Chamorovsky and A. B. Rubin, FEBS Lett., 191 (1985) 245 - 248. 5 A. M. Freiberg, V. I. Godik, S. G. Kharchenko, K. E. Timpmann, A. Yu. Borisov and K. K. Rebane, FEBS Lett., 189 (1985) 341- 344. 6 V. Z. Paschenko, B. N. Korvatovsky, S. L. Logunov, S. S. Vasilev, A. A. Kononenko, P. P. Knox, N. I. Zakharova, N. P. Grishanova and A. B. Rubin, Dokl. Akad. Nauk SSSR, 290 (1986) 742 - 747. 7 V. Z. Paschenko, B. N. Korvatovsky, S. L. Logunov, A. A. Kononenko, P. P. Knox, N. I. Zakharova, N. P. Grishanova and A. B. Rubin, FEBS Lett., 214 (1987) 28 - 34. 8 S. L. Logunov, S. S. Vasilev, B. N. Korvatovsky, V. B. Tusov, P. P. Knox, N. I. Zakharova, N. P. Grishanova, V. Z. Paschenko and A. A. Kononenko, Dokl. Akad. Nauk SSSR, 299 (1988) 1004 - 1007. 9 V. I. Godik, P. P. Knox, A, A. Kononenko, A. Yu. Borisov and A. B. Rubin, Mol. Biol. (USSR), 20 (1986) 1305 - 1312. 10 M. R. Gunner, D. E. Robertson and P. L. Dutton, J. Phys. Chem., 90 (1986) 3783 3795. 11 P. Sebban, FEBS Lett., 233 (1988) 331- 334. 12 N. I. Zakharova, M. Fabian, N. Ya. Uspenskaya, A. A. Kononenko and A. B. Rubin, Biokhimiya (U.S.S.R.), 46 (1981) 1703 - 1710.

47 13 C. E. D. Chidsey, C. Kirmaier, D. Holten and S. G. Boxer, Biochim. Biophys. Acta, 766 (1984) 424 - 437. 14 N. W. Woodbury, W. W. Parson, M. R. Gunner, P. C. Prince and P. L. Dutton, Biochim. Biophys. Acta, 851 (1986) 6 - 22. 15 J. Jortner, Biochim. Biophys. Acta, 594 (1980) 193 - 230. 16 B. J. Hales, Biophys. J., 16 (1976) 471 - 480.