Flash-induced Absorption Spectroscopy Studies of ... - Semantic Scholar

2 downloads 0 Views 777KB Size Report
Dec 30, 2017 - Droppa, M., Terry, N., and HorvBth, G. (1984)Proc. Natl. Acad. Sci. U. S. A. 81,. Res. ... Cheniae, G., and Martin,. I. F. (1971) Plant Physiol.
VOl. 269, No. 52, Issue of December 30, PP. 32865-32670,1994 Printed in U.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc

Flash-induced Absorption Spectroscopy Studies of Copper Interaction with PhotosystemI1 in Higher Plants* (Received for publication, August 15, 1994, and in revised form, October 17, 1994)

Wolfgang P. SchroderS, Juan B. Arellano57, Thomas Bittnerll**, Matilde Baron§$*, Hann-Jorg Eckertll, and Gernot Rengerll§§ From the Department of Biochemistry, Arrhenius Laboratories for Natural Sciences, Stockholm University, S-106 91 Stockholm, Sweden, $Plant Biochemistry Department, Estacidn Experimental del Zaidin, CSIC, Profesor Albareda 1, E-18008 Granada, Spain, and ((Max-Volmer-Institut,TU Berlin, Strasse des 17, Juni 135, D-106 23 Berlin, Germany

lutant makes of it great interest to understand and locate to the Measurements of flash-induced absorption changes at process. 325,436, and 830 nm and of oxygen evolution were per- inhibition site in the photosynthetic formed in order to analyze in detail the inhibition of Early studies suggested that copper directly affects the phophotosystem I1 (PS 11) by Cu(I1)in PS I1 membrane frag- tosynthetic light reactions(6) and that light is required for the ments from spinach.(a)The kinetics of P680’ reduction toxic effects of copper ( 7 , 8). The inhibition of photosystem I 100 p~ becomemarkedlyslowerinthepresenceof electron transport has been reported in algae and plants by CuSO,. ( b )The CuS0,-induced kinetics of P680’ reduc- different authors (9, 10) and attributed to interactionwith its tion are dominated by a 140-160-psdecay. ( c ) The extent reducing side (111, probably with ferredoxin (12). The cytoof these 140-160-ps kinetics, normalized to the overall chrome b,/f complex has also been suggested as an inhibitory decay, remains virtually unaffected by addition of the target site (13). exogenous PS I1 donor, =,OH. ( d ) In thoroughly darkThe interaction mechanism of copper with photosystem I1 adapted samples the CuS0,-induced 140-160-ps kinetics and theprecise location of the inhibitory binding site is still a are already observed after the first flash and remain unchanged by a train of excitation flashes. (e) The ex- matter of debate. In earlier reportsmost authors proposed the (7,lO-12,14,15). tent of P680’ and QA- formation under repetitive flash donor side of photosystem I1 as the target site Hsu and Lee (16) and Renganathan and Bose (17) suggested 100 p~ excitation is notdiminished by additionof CuSO,. ( f ) The induction of ps kinetics of P680’ reduc- that the toxic effect of copper creates a lesion close to the tion at the expense of ns kineticsand the inhibition of reaction center. Some recent studies have challenged the idea of a toxic effect located at the donor side and suggesteda copper the saturation rate of oxygen evolution exhibit the same of photosystem 11, at the level ( 8 ) CuSO, also binding siteon the reducing side dependenceonCuSO,concentration. transforms the 10-20-ps reduction of P680+by T y r z in of the Q, site (181, or in the pheophytinQ,-Fe domain (18-20). effects, Le. copper-induced Tris-washedPS I1 membrane fragments into140-160-ps Another report suggests multiple kinetics without any effect on the extent of flash-in- modifications of both the donor (near Tyr,) and the acceptor duced P680+ formation.Theseresultsunambiguously side (near the QB site) (22). show that Cu(I1) does not affect the charge separation This paper presentsan attempt tolocalize the siteof copper (P680’QA-),but instead specifically modifies Tyrz and/or inhibition in PS 11’ more precisely by measuring laser flashits micro environment so that the electron transfer to induced absorption changes a t 830, 436, and 325 nm, which P680’ becomes blocked. provide information on the extentof charge separationbetween Copper is an essential micro nutrient for higher plants and algae (for review see Ref. 1) required for maintaining photosynthetic efficiency. It is directly involved in the electron transport as a cofactor of the blue protein plastocyanin (2). This transition metal indirectly influences the lipid and pigment composition of the thylakoid membrane (3, 4). On the other hand, it is well knownthat copper can also betoxic, and for that reason it is often used as fungicide and algaecide (5). The increasing concentration in the environmentof this potent pol-

P680’ and QA-, the kinetics of the electron transfer from the redox-active Tyrz t o P680’ and from the manganese-containing water oxidase to Tyry. In this paper we show that the charge separation (P680+QA-)is unaffected by copper. The results are interpreted in termsof a main donor-side inhibition by copper, located close to the Tyr, donor in the D l protein.

MATERIALS AND METHODS Spinach was obtained from the local market. Photosystem I1 particles (BBY) were isolated by the method of Berthold et d . (23) with the modification of Arellano et al. (24), in which the Triton-treated thyla4 min at 10,000 x g in ordertoremove koidswerecentrifugedfor the optical * The costs of publication of this article were defrayed in part by the contaminating starch, DNA, andmetalsandtoimprove in mM MES, pH payment of page charges. This article must therefore be hereby marked properties of the sample. Samples were resuspended 20 “advertisement”in accordancewith 18 U.S.C.Section 1734 solely to 6.5, and 10 mM NaC1. Tris treatment wasperformed as inRef. 25. indicate this fact. Photosystem I1 core particles were prepared as in Ref. 26. Copper sul$Supported by the Swedish ForestandAgriculturalResearch fate was used throughout the whole experiment, since no difference Council (SJFR). To whomcorrespondenceshould be addressed. Tel.: between copper chloride and copper sulfate could be detected in the 46-8-16-43-92;Fax: 46-8-15-36-79. inhibition of oxygen evolution. Copper was added in the dark to the 1 Recipient of a predoctoral fellowship fromthe Spanish Science and Education Ministry. Supported by a grant from the Spanish DGICYT (Project PB 88-0092). ’The abbreviations used are: PS 11, photosystem 11; P680,primary ** Supported by a postdoctoral fellowship from Deutsche Forschun- electron donor PSII; DCBQ, 2,6-dichlorobenzoquinone;QA/QB,secondgsgemeinschaft. ary plastoquinon acceptors;Tyr,, tyrosine residue of PSII; Pheo, pheoI$ Supported by a grant from the Spanish DGICYT (Project PB 88- phytin; Chl, chlorophyll; MES, 4-morpholineethanesulfonic acid; BBY, 0092). Triton X-100-based photosystem I1preparation; OEC, oxygen evolving $5 Supported by Deutsche Forschungsgemeinschaft (RE354/11-2). complex.

32865

Copper Inhibition of PS II

32866

sample a few minutes before the measurements.Longer incubations (up to 30 min) were also tested, but significant differences were not observed (not shown). To some samples 2 mM hydroxylamine was added 2 min before performing the measurements. The measurements of the 830-nm absorption changeswith microsecond time resolution were performed at room temperature witha single beam flash photometer (27) with a n optical path length of 1 cm. PS I1 was excited by light pulses from a Q-switched frequency doubled Nd: YAG laser (A = 532 nm, duration3 ns) at a repetition rate of 1Hz. 16or 32 signals were averaged to improve the signal to noise ratio, except in the single flash studies. The flash-induced absorption changesat 325 and 436 nm were measured with an apparatus as described in detailpreviously (28). The dark time between the saturatingactinic laser flashes was700 ms. 90 signals were averaged to improve the signa1:noise ratio. The sample was exchanged for each series of four flashes. The chlorophyll concentration was as indicated in the figure legends. The electron acceptor 2,6-dichloro-benzoquinone (DCBQ) was used at a concentration of 100 p. Oxygen evolution was measured polarographically, using a Clark type electrode a t 10 pg of CWml in a buffer containing 4 mM sucrose, 10 mM NaCl, and 20 mM Mes-NaOH, pH 6.5, using 0.4 mM DCBQ as electron acceptor.

A

B

+ cum4

RESULTS

PS I1 membrane fragments (BBY) with an intactoxygen-evolving complex, the reduction of P680' has been found to occur with multiexponential kinetics within the ns and ps time domain (29-33). About 70% occurs in the nanosecond time range. Measurements of the 830-nm absorptionchangeswith ns time resolution are often complicated because of the scatteringbehavior of the sample. This problem can be circumvented by an indirectmethod that permits determination of the extent of ns kinetics by measurement of absorption changes at a resolution of 1ps. In thiscase, the nanosecond reduction of P68O' by the primarydonor T y r z cannot be detected directly. However, its extent canbe obtained by calculating the difference between the maximal amplitude in the presence of 2 2 mM NH,OH and the amplitude 2 ps after the actinic flash in theabsence of NH,OH (for details see Refs. 27 and 34). Fig. lA depicts the time course of absorption changes at 830 nm induced by repetitive laser flashes in PS I1 membrane fragments (BBY) from spinach in theabsence and in the presence of 2 mM NH,OH. In theabsence of NH,OH (Fig. LA, no addition) the initial amplitude of AA,,, only represents the fraction of P680+that is re-reduced with ps kinetics. After adincreases by a dition of NH,OH, the initial amplitudeof factor of about 2.5, and the decay is dominated by kinetics with ahalf-life of about 10 ps.These kineticsare ascribed to the electron transfer from Tyrz to P680' in PS 11, deprived of their exhiboxygen evolution capacity (35, 36). A small partof its a slower decay (with a calculated t a of 130 ps) caused by a charge recombination between P680' and QA-(37, 38). These decay kinetics of P680' are slow enough to be completely resolved by the time resolution of our equipment. The capability for stable charge separation is not affected by NH,OH (39). Therefore, the initial amplitude AA,, measured in the presence of NH,OH, is a measure of the numberof reaction centers, which are capable of performing a stable charge separation between P680' and QA-. Fig. 1B shows absorption changes in a control sample and PS I1 membrane fragments after incubation with 100 1.1~ CuSO,. A comparison with Fig. LA clearly shows that CuSO, by practically increases theobserved initial amplitudeof the same factor as NH,OH. This finding is an unambiguous indication of a copper-induced shift of the P680' ns reduction t o the ps time domain without any detrimental effect on the capacity of the PS I1 reaction center to perform a stable charge separation between P680' and QA-.A closer inspection of the data reveals that the reduction kinetics of P680+ in coppertreated samples are much slower than after incubation with AAaso Measurements-In

PA,,

AA,

AA,

AA,

AA,

' 1

I

50 ps

time FIG.1. Time course of laser absorption changes at 830 nm induced by repetitive laser flashes in PS I1 membrane fragments. A , with (upper truce) and without (lower truce) the addition of 2 mM NH,OH (2-min incubation). B , with (upper truce) and without (lower truce) 100 PM CuSO,. 16 flashes were averaged at a repetition rate of 1Hz, in the presence of 100 p~ DCBQ, and at a chlorophyll concentration of 50 pg/ml. Time resolution is 1 ps.

2 mM NH,OH. In the presence of copper, P680' becomes predominantly reduced by kinetics with half-lives between 120 and 160 ps. In order to corroborate these findings by independent lines of evidence, flash-induced absorption changes were measured at 436 nm. At this wavelength, P680' has a characteristic negative peak in the absorption differential spectrum (40, 41). The sweep times of these measurements were 4 ms, and, therefore, in Fig. 2 only a small spikeis seen for the control, which reflects the slower components of P68O' reduction in intactPS I1 membrane fragments. The rapidly relaxing negative spike is followed by a positive signal of a rather small amplitude, which remains constant within the4-ms time window. Upon addition of copper (100 p v , lower trace) the extentof the flash-induced negative signal drasticallyincreases, and the relaxation is much slower. As in the830-nm absorption changesin thepresence of copper, AA,,, decays mainly with a half-life of 140 ps. Single flash experiments with dark-adapted samples reveal that the positive signal at 436 nm reflecting the S, 4 S , transition in the water oxidase (42-44) disappears completely (data not shown). The dataof Figs. 1and 2 are allconsistent with the assumption that copper addition a t micromolar concentrations causes a drastic retardationof the reduction kineticsof P680' virtually without any effect on the process of stable charge separation, leading to P680+QA-formation. In addition, the disappearance of the positive absorption change at 436 nm suggests that the electron extraction from the oxygen-evolving complex (OEC) is blocked by Cu". To substantiate this idea, the initial amplitude of AA,,, and its nanosecond decay component (determined according to Ref. 27) were compared with the rate of oxygen evolution, normalized to the respective control values, as a function of copper to chlorophyll ratio. The data obtained are

32867

Copper Inhibition of PS I1

200 FS

H time FIG.4. Single flash-induced absorption change measurements at 830 nm.PSI1 membrane fragments (BBY)are shown in thepresence

4 rns

time FIG.2. Flash-induced absorption changes at 436 nm in PS 11 membrane fragments. Upper trace, control; lower trace, in the presence of 100 p~ CuSO,; single flash experiment, 40 measurements averaged, in the presence of 100 p~ DCBQ, and at a chlorophyll concentration of 10 pg/ml. Time resolution is 50 ps.

of 100 pv CuSO,. Samples (100 pg/ml) were totally dark-incubated with copper for 30 min and then subjected to a train of laser flash (1 Hz).

AA,

vious that in the presence of copper decays mainly with a half-life of about 150 ps already after the first flash and that the reduction of P680' is independent of the number of preflashes. The long time of dark adaptation (30 min) before the first flash makes it very improbable that the incapability of 100 Tyrz to donate electrons to P680' in the presence of Cu2+is simply caused by a prolonged re-reduction time of Tyry, which is known to occur with a time constant of less than 1 s in samples in which the oxygen evolution capacity is selectively inhibited, e.g. by Tris treatment. To illustrate thedifference from samples selectively deprived 40 of their oxygen evolution capacity, Fig. 5A shows for a comparison 830-nm absorption changes induced by repetitive flashes 2Ot in Tris-treatedPS I1 membrane fragments. The repetitive frequency was small enough (1Hz) so that formed via the 0 10-20-ps kinetics of electron abstraction by P680' becomes re0 1 2 3 4 5 reduced between the flashes.Therefore, in the absence of copCu/Chl (mol ratio) per, the relaxation kinetics of P680' are dominated by the 10FIG.3. The extent of nanosecond decay component(AA;;,,, 20-ps decay (35, 45). Addition of copper to Tris-washed PS I1 and the microseconddecaycomponent A), of the 830nm membrane fragments gives rise to a retardation of the reducabsorption changes (from Fig.1) and the rate of oxygen evolution (0) in PS I1 membrane fragments (BBY)as a function of tion kinetics of P680' t o ty, = 160 ps, observed in normal PS I1 copper:chlorophyll ratio. All quantities are normalized to the respec- membrane fragments treated with copper. On the other hand, tive control value. The rate of oxygen evolution of the control sample the initial amplitude of AA,,,,which reflects the number of was 530 pmol of 0, per mg of Chl per h. reaction centers capable of performing a stable charge separation, is not affected by copper. This indicates thatalso in Trisdepicted in Fig. 3, which shows that the initial amplitude in- treated samples copper is able to bind and inhibit theelectron creases and reaches (at a copper:chlorophyll ratio of about 2) donation of T y r z and that copper does not affect the electron the level of hydroxylamine-treated samples (not shown). Fur- transfer from P680 to QA. thermore, within the experimental error range of the data, the AA,,, Measurements-The conclusions described so far can oxygen evolution and the extent of nanosecond decay compo- be independently testedby measurements of flash-induced abnents of A A a a 0 exhibit a n identical decline with respect to the sorption changes in the W region. At 325 nm the difference copper:chlorophyll ratio. This provides a n unambiguous proof extinction coefficient of QA/QA- dominatesthe absorption that copper under the experimental conditions of this study changes. In addition, the transitions between the S states of the exerts only marginal, if any, influence on the capacity of stable water-splitting complex also contributetothese absorption charge separation, whereas oxygen evolution becomes almost changes (44, 46, and references therein). completely suppressed. Fig. 6 shows typical traces of absorption changes at 325 nm Two alternative mechanisms could be responsible for the ( A A 3 2 J , induced by a train of four flashes in PS I1 membrane retardation of P680' reduction and simultaneous inhibition of fragments thatwere dark-adapted in the absence and presence 0, evolution: (i) Cu2+blocks the function of T y r z to act asredox of two different copper concentrations. Within the accuracy of mediator between P680' and the OEC, or (ii) Cu2+specifically the experiment, the initial amplitude