that the photoinactivation of Photosystem II in vivo is a probability and light-dosage event which depends on the photons absorbed and not the irradiance per se.
Photosynthesis Research 56: 1–13, 1998. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.
1
Regular / hypothesis paper
Unifying model for the photoinactivation of Photosystem II in vivo under steady-state photosynthesis Jan M. Anderson1,2, Youn-Il Park2 & Wah Soon Chow1 1 Photobioenergetics
Group, Research School of Biological Sciences, Institute of Advanced Studies, Australian National University, GPO Box 475, Canberra ACT 2601, Australia; 2 Department of Plant Physiology, Umeå University, S901 87 Umeå, Sweden Received 26 August 1997; accepted in revised form 10 November 1997
Key words: photoinhibition, Photosystem II, primary radical pair, singlet oxygen, triplet P680
Abstract We present a unifying mechanism for photoinhibition based on current obsevations from in vivo studies rather than from in vitro studies with isolated thylakoids or PS II membranes. In vitro studies have limited relevance for in vivo photoinhibition because very high light is used with photon exposures rarely encountered in nature, and most of the multiple, interacting, protective strategies of PS II regulation in living cells are not functional. It is now established that the photoinactivation of Photosystem II in vivo is a probability and light-dosage event which depends on the photons absorbed and not the irradiance per se. As the reciprocity law is obeyed and target theory analysis strongly suggests that only one photon is required, we propose that a single dominant molecular mechanism occurs in vivo with one photon inactivating PS II under limiting, saturating or sustained high light. Two mechanisms have been proposed for photoinhibition under high light, acceptor-side and donor-side photoinhibition [see Aro et al. (1994) Biochim Biophys Acta 1143: 113–134], and another mechanism for very low light, the low-light syndrome [Keren et al. (1995) J Biol Chem 270: 806–814]. Based on the exciton-radical pair equilibrium model of exciton dynamics, we propose a unifying mechanism for the photoinactivation of PS II in vivo under steady-state photosynthesis that depends on the generation and maintenance of increased concentrations of the primary radical pair, P680+ Pheo− , and the different ways charge recombination is regulated under varying environmental conditions [Anderson et al. (1997) Physiol Plant 100: 214–223]. We suggest that the primary cause of damage to D1 protein is P680+ , rather than singlet O2, formed from triplet P680, or other reactive oxygen species. Abbreviations: Chl – chlorophyll; D1 – protein, psbA gene product; NPQ – non-photochemical quenching; P680 – primary electron donor in PS II; Pheo – pheophytin; PS II (PS I) – Photosystem II (Photosystem I); QA (QB ) – primary (secondary) quinone electron acceptor of PS II; qP – photochemical quenching coefficient; Tyrz – redoxactive tyrosine in PS II reaction centre Introduction Studies in photoinhibition, the loss of PS II photochemical efficiency under excess light, have mainly been concerned with the loss of function in isolated thylakoid membranes and PS II membrane fragments under extremely high light, with photon exposures rarely experienced by PS II in leaves and algae. Also, in the field, the inevitability of photoinactivation of
*157970*
PS II is matched by multiple, interacting photoprotective strategies that help protect PS II from excess photons. Hence, to understand how PS II functions in vivo the opposing effects of PS II photoinactivation and its complex photoprotection must be considered (see Chow 1994). Photoprotective strategies include physiological responses which decrease incident light, such as leaf and chloroplast movement, waxy cuticles and so on. Also, when plants are ex-
PIPS NO.:157978 (M) (preskap:bio2fam) v.1.15 pres775b.tex; 30/06/1998; 12:21; p.1
2 posed to light intensities in excess of those that can be used in photosynthetic electron transport, nonphotochemical dissipation of excess excitation energy from the light-harvesting antenna of PS II, partly achieved by the xanthophyll cycle (Demmig-Adams and Adams 1992), is a major photochemical photoprotective mechanism (Horton et al. 1996). In higher plants, under excess light, non-functional, photoinhibited PS II centres accumulate in stacked membrane regions where D1 protein degradation is prevented: when normal irradiance is received again, these PS II centres are speedily repaired (Anderson and Aro 1994). In the short term, other molecular mechanisms such as spillover, state transitions and cyclic electron flow around PS II help to fine-tune the regulation of PS II function. In the longer term, light and metabolic acclimation drive dynamic adjustments of the photosynthetic apparatus in response to varying environmental stimili (Anderson et al. 1995, 1997). All of these photoprotective strategies acting together with varying extents in different plants under different environmental conditions help plants to balance energy supply with energy consumption under all light levels (Anderson et al. 1995; Huner et al. 1996; Anderson et al. 1997). However, the extremely high oxidising potential of P680+ required for the oxidation of water (Klimov et al. 1969), ensures that despite the catena of photoprotective strategies which regulate and optimise PS II function, there is an inevitable probability that PS II is photoinactivated in vivo. Arising mainly from in vitro photoinhibition studies, two mechanisms have been proposed for in vivo photoinhibition in high light, known as acceptor-side and donor-side photoinhibition (Aro et al. 1993; Ohad et al. 1994; Barber 1995). Another mechanism that does not require prior impairment of electron transport through PS II has been proposed to be operational at very low light only, the low-light syndrome (Ohad et al. 1994; Keren et al. 1995, 1997). However, the photoinactivation of PS II is a probability and light-dosage event (Park et al. 1995, 1996a; Baroli and Melis 1996; Tyystjärvi and Aro 1996) which obeys the reciprocity law; it depends only on absorbed photons, and not the irradiance rate per se (Park et al. 1996a). Being a light-dosage effect, PS II inactivation occurs in limiting, saturating and super-saturating light, strongly suggesting that only one mechanism predominates in vivo (Park et al. 1995, 1996a; Baroli and Melis 1996; Park et al. 1996a; Tyystjärvi and Aro 1996). Further, target theory analysis demonstrates that only one photon is required for PS II photoinactivation (Sinclair et
al. 1996; Park et al. 1997). Hence, mechanisms requiring two photons for the double reduction of QA and subsequent QA H2 formation are unlikely for the in vivo situation. We propose a unifying mechanism for the photoinactivation of PS II in vivo during steady-state photosynthesis which requires only one photon. Our model is based on a simplified reversible excitonradical pair equilibrium model (Schatz et al. 1988; Dau 1994; Schreiber and Krieger 1996). It depends on the probability of charge recombination of the primary radical pair, P680+ Pheo− and the various regulatory ways this occurs under varying environmental conditions (Anderson et al. 1997). The generation and/or maintenance of the primary radical pair is equilibrated with antenna chlorophyll and its higher concentration causes a higher rate of PS II inactivation leading to D1 protein turnover. We suggest that the photoinactivation of PS II in vivo occurs by one predominant molecular mechanism involving one partner of the primary radical pair, P680+ , as the active species which causes oxidative damage to the D1 protein. Any triplet P680 generated via charge recombination of the primary radical pair is likely to be actively quenched; the more QA is reduced, the more the triplet state will be effectively dissipated (van Mieghem et al. 1995). We suggest that 3 P680 interacting with oxygen to produce singlet oxygen as proposed in the mechanism for acceptor side-photoinhibition is unlikely to play a predominant role in the photoinactivation of PS II in vivo.
Materials and methods Pea (Pisum sativum L. cv. Greenfeast) was grown in growth cabinets (12 h light/ 22 ◦ C; 12 h dark/18 ◦ C) illuminated by fluorescent lights at 50 (low-light), 250 (medium -light) and 700 (high-light) µmol photons m−2 s−1 (Park et al. 1995). For inhibitor treatments, leaf petioles were cut under water and transferred to small tubes containg water (control), 1 mM dithiothreitol or 1 µM nigericin and allowed to imbibe water or an inhibitor for 2 h in a gentle air stream at low light (20 µmol photons m−2 s−1 ) as described previously (Park et al. 1996a). The number of functional PS II reaction centres (mmol PS II mol−1 Chl) was determined as a function of photon exposure according to Park et al. (1995) using a leaf-disc oxygen electrode system (Hansatech). Light treatments were performed in two ways to give
pres775b.tex; 30/06/1998; 12:21; p.2
3 idential photon exposures of 0, 0.36, 1.53, 3.06, 4.68, 6.48 and 10.8 mol photons m−2 , either by variable irradiances of 1, 100, 425, 850, 1800 and 3000 µmol photons m−2 s−1 given for 1 h, or by a fixed irradiance of 1800 µmol photons m−2 s−1 applied for 0, 3.3, 14.2, 28.3, 43.3, 60 and 100 min. During illumination, a constant flow of a humidifed gas mixture containing 1.1% CO2 in air was passed through the chamber at 25 ◦ C. The light was provided by a slide projector, and various irradiances were achieved with appropriate neutral-density filters.
Results and discussion The photoinactivation of PS II in vivo depends on light dosage The classical paper of Jones and Kok (1966) on the kinetics and action spectra of photoinhibition of PS II is still quoted for its detailed action spectra. Oddly, their demonstration that the photoinactivation of PS II in vitro with isolated spinach thylakoids and in vivo with the cyanobacterium, Anacystis nidulans, is a lightdosage effect that obeyed the reciprocity law has been buried in the literature. Jones and Kok (1966) showed that the photoinactivation of PS II electron transport in isolated spinach thylakoids followed the general exponential attenuation law, Rt = Ro e−ct , where Ro is the initial rate of PS II electron transport, Rt is the rate of electron transport following photoinhibitory exposure during time t, and c is a constant related to irradiance. The product ct is related to photon exposure measured as mol photons m−2 . R is also observed with respect to the photoinactivation of PS II between the irradiance level and the time of illumination in a cyanobacterium, Synechocystis sp. PCC 6803 (Nagy et al. 1995) and for photosynthesis of Rumex patientia L. leaves exposed to UV-B irradiation during growth (Sisson and Caldwell 1977). As PS II inactivation is a light-dosage effect, Park et al. (1995, 1996a) reasoned that leaves might also show reciprocity, especially light-acclimated pea leaves which despite having different light-harvesting PS II antennae size, had the same amount of chlorophyll on a leaf area basis and so absorbed the same fraction of incident light (almost 90%). High- and low-light pea leaf discs were illuminated with identical photon exposures obtained with either a constant illumination time of one h at varying irradiances of 0 to 3000 µmol m−2 s−1 or a constant irradiance of
Figure 1. Photon dose-dependence of photoinactivation of functional PS IIs in leaves of peas grown in low (dotted lines) and high (solid lines) light. Leaf discs pretreated with water (Cont; , ) or 0.6 mM lincomycin (Linc; B , E ) were illuminated in a Hansatech leaf-disc electrode system supplied with humidified air containing 1.1% CO2 . Photon exposures of 0, 0.36, 1.53, 3.06, 4.68, 6.48 and 10.8 mol photons m−2 were given at varying irradiances of 0, 100, 425, 800, 1300, 1800 or 3000 µmol m−2 s−1 for 1 h ( , B ) or at a fixed irradiance of 1800 µmol m−2 s−1 ( , E ) for 0, 3.3, 14.2, 28.3, 49.3, 60 and 100 min, respectively.
#
#
1800 µmol m−2 s−1 for varying times from 0 to 2 h. As shown in Figure 1, within the range of photon exposures tested up to 13 mols photons m−2 , exactly the same amounts of functional PS II reaction centres were inhibited at each photon exposure generated in the two ways. With high-light peas, no photoinactivation of PS II was observed until the photon exposure exceeded 3 mol photons m−2 , indicating that at low photon exposures the rate of D1 protein synthesis exactly matched the rate of D1 protein degradation and no net PS II inactivation occurred. Reciprocity was observed in all acclimated pea leaves, but the photon dosage required for PS II photoinactivation was much lower in low-light peas (Figure1) which have lower capacity for both maximal photosynthesis and photoprotective strategies (Park et al. 1996a, b). In order to prevent de novo D1 protein synthesis and observe PS II inactivation at low photon exposure it was necessary to block D1 protein synthesis, one of the main protective strategies elicited even under low light (Aro et al. 1993; Anderson and Aro 1994). With high- and low-light peas pretreated with lincomycin (a chloroplast-encoded protein synthesis inhibitor, which at low concentrations specifically prevents D1 protein synthesis), the functional PS IIs were decreased
pres775b.tex; 30/06/1998; 12:21; p.3
4 more or less biphasically (Figure 1). Again, there is a reciprocity between the irradiance level and time of irradiance. Thus, for both control leaves and those inhibited in D1 protein synthesis, PS II inactivation depends on the number of photons received, and not the rate of delivery of photons. Despite being counterintuitive, reciprocity holds not only for limiting light through to saturating light for photosynthetic capacity, but also under sustained high light under both aerobic (Park et al. 1995a, 1996a) and anaerobic (Park et al. 1997) conditions. Photoinactivation of PS II is a light-dosage effect; when a certain number of photons have been absorbed by the PS IIs in a leaf there is a finite probability of an individual PS II being inactivated. Note that this does not mean that the probability of PS II photoinactivation will be the same with different leaves which have varying extents of multiple photoprotective strategies. There is a functional heterogeneity of PS IIs in vivo with respect to chlorophyll fluorescence induction (Melis 1991). There is also a functional heterogeneity in vivo with respect to the susceptibility of PS II to photoinactivation, with the quantum yields of PS II inactivation in acclimated peas being higher at low photon exposure than at high photon exposure (Park et al. 1996a). In high-light peas, the apparent quantum yield of D1 protein degradation was also higher at low than at high photon exposure (Park et al. 1996a), and the same was true in Chlamydomonas cells at low light (Keren et al. 1995). Given that between 105 to 107 photons, when absorbed by leaves, isolated thylakoids and Chlamydomonas cells, will lead to the loss of function in one PS II (see refs. in Anderson et al. 1997), it is important to re-state that the photoinactivation of PS II is a rare but finite probability event arising from the production of toxic species or unstable states. This probability is not unique, but varies with the type of PS II being inactivated, the state of acclimation of PS II and other thylakoid components, as well as the varying extents of multiple photoprotective strategies operational in particular leaves. One photon is required for the photoinactivation of PS II in vivo Target theory was adopted to describe the damage incurred by living cells when exposed to ultraviolet or ionizing radiation (Hutchinson and Pollard 1961). Target theory uses the Poisson distribution (a probability phenomenon) to describe the frequency distribution of the number of photons absorbed per living cell or
target in an irradiated sample. In target theory four assumptions are made with respect to the living organism: (i) targets are present; (ii) they are bombarded with random radiation; (iii) a hit or interaction with the target kills, inactivates or induces a radiobiological effect; and (iv) the probability of producing a hit is proportional to the target size (Altmann et al. 1970). The first three criteria needed to apply target theory to PS II photoinactivation are satisfied as the target of photoinactivation is the PS II reaction centre, photons absorbed by thylakoid membranes interact randomly with PS IIs in the thylakoid membranes, and some fraction of these photons will inactivate PS II. With respect to criterion (iv), by analogy it is reasonable to equate target size with the size of the light-harvesting antenna of PS II. However, the influence of the lightharvesting antenna size of PS II on photoinhibition is controversial. Some investigators have suggested that photoinhibition is proportional to antenna size (Cleland and Melis 1987; Park et al. 1997) while others claim it is not (Tyystjärvi et al. 1991, 1994, 1996; Sinclair et al. 1996). Recently, we demonstrated in acclimated peas that low-light peas with larger antenna size showed greater PS II photoinactivation than highlight pea under conditions when other photoprotective effects were eliminated or minimized (anaerobic conditions and inhibition of D1 protein repair and protonmediated events); this is the strongest evidence yet that PS II photoinactivation in vivo indeed depends on antenna size (Park et al. 1997). Being a probability event, the application of target theory to the photoinactivation of PS II is valid only if irradiated samples contain many sensitive targets compared to the irradiated area. This holds as there are about 0.7 to 1.0 µmol PS II per m2 of leaf area (Park et al. 1996b), and the approx. area of a PS II is 200 (nm)2 (Boekema et al. 1995). Of course, target theory is valid only if reciprocity holds with respect to the biological effect between the irradiance level and time of irradiation. As detailed above, where tested, reciprocity has been demonstrated for photoinhibition by visible light both in vitro and in vivo, as well as for UV-B irradiation. The number of photons required to photoinactivate PS II can be determined unequivocally by target theory. When target theory was applied to the loss of PS II function in acclimated pea leaves, as measured by O2 evolution, under both normal aerobic or anaerobic conditions, only one photon was required for the loss of PS II function, i.e. it is a one-hit process in all cases (Sinclair et al. 1996; Park et al. 1997). This has impor-
pres775b.tex; 30/06/1998; 12:21; p.4
5 tant implications for the mechanism involved in PS II inactivation. Three mechanisms proposed for the photoinhibition of PS II in vivo Arising mainly from extensive in vitro studies where PS II thylakoids or isolated PS II membrane fragments were bombarded with very high irradiance for a long time, often with extremely high photon exposure of 100 mol photons m−2 or more, a consensus for two molecular mechanisms for photoinhibition in vitro soon emerged, termed acceptor-side and donor-side photoinhibition (see Aro et al. 1993; Ohad et al. 1994; Barber 1995). However, with isolated thylakoids or PS II complexes there is no possibility for D1 protein synthesis and hence for D1 protein turnover, a major strategy for the maintenance of functional PS II (Aro et al. 1993; Anderson and Aro 1994; Ohad et al. 1994). Moreover, the relevance of such high photon dosage in the in vitro studies to the natural environment is questionable: in nature most PS IIs in leaves would rarely be exposed to such high photon exposures as used routinely for in vitro studies. Nevertheless these two mechanisms provided impetus to investigate their relevance to in vivo photoinhibition. In acceptor-side photoinhibition, high light leads to an alteration of the acceptor side of the PS II reaction center which blocks electron flow from QA − to QB . In turn, this would increase the possibility of charge recombination of the primary radical pair P680+ Pheo− which may facilitate the formation of triplet chlorophyll, 3 P680, which then reacts with O2 to generate singlet oxygen (1 O2 ) (Vass et al. 1992). The oxidative damage due to 1 O2 with an extremely short life-time is presumed to be confined to PS II, thereby targeting D1 protein for degradation (Ohad et al. 1994). Alternatively, the donor-side photoinhibition mechanism would operate under high light when the supply of electrons from Tyrz − to P680+ does not match the rate of electron removal from Pheo− . For example, a high transthylakoid pH gradient causes overacidification of the thylakoid lumen, Ca2+ is partly released from the oxygen evolving complex, the redox potential of QA is lowered and PS II is down-regulated (Krieger and Weis 1993; Johnson et al. 1995). With inactivation of the oxygen evolving complex, the cation radicals Tyrz + and P680+ that persist for longer periods may cause oxidative damage in the presence or
Figure 2. A model for the photoinactivation of PS II in vivo and energy conversion in PS II based on the exciton-radical pair equilibrium model, featuring rapid excitation equilibration between PS II antenna and the reaction centre chlorophyll P680. Light absorbed by PS II antenna including P680 will be dissipated by three reactions: (i) excitation trapping (k1 ); (ii) fluorescence (kF ); and (iii) non-radiative heat dissipation mainly by PS II light-harvesting antenna (kDA ), but also by triplet P680 quenching (k2DT ) and possibly reaction centre quenching (k2D ). The primary radical pair, P680+ Pheo− , will decay by four reactions: (i) k−1 , charge recombination; (ii) k2Q , charge stabilization by electron transfer to QA ; (iii) k2D , nonradiative decay to ground state; and (iv) k2T , spin dephasing to triplet state of radical pair followed by formation of triplet P680. The dotted lines depict possible pathways of PS II photoinactivation. Photoinactivation of PS II in vivo leading to D1 protein damage may result either from the oxidative pathway from P680+ or from triplet P680 giving rise to singlet oxygen. We strongly favour the oxidative pathway with P680+ , rather than singlet O2 or reactive oxygen radicals, being the damaging species responsible for D1 protein degradation.
absence of oxygen that triggers D1 protein degradation. It is important to note that the donor-side and acceptor-side mechanisms for PS II photoinactivation are distinctly different. Both mechanisms form highly reactive but different oxidants, P680+ or 1 O2 , formed from 3 P680, that are assumed to cause damage to PS II reaction centre leading to D1 protein degradation. It is not clear which of these mechanisms is dominating in vivo, or if other mechanisms are involved. Identification of the proteolytic fragments of D1 protein arising from its presumed cleavage on the lumenal or stromal side of thylakoid membranes, corresponding to those expected from donor- or acceptor-side photoinhibition (Aro et al. 1993; Barber 1995), has not given clear results. Kettunen et al. (1996) conclude that an in vivo photoinhibition mechanism cannot be deduced on the basis of fragmentation patterns from D1 protein degradation. Another variant of the acceptor side mechanism has been introduced by Ohad et al. (1994) to account for the degradation of D1 protein induced by very low light. Based on studies with Chlamydomonas rein-
pres775b.tex; 30/06/1998; 12:21; p.5
6 hardtii and spinach thylakoids, Keren et al. (1995, 1997) proposed a third specific mechanism for photoinhibition in very low light (the ‘low light syndrome’) in which D1 protein degradation is triggered by back electron flow from Q·− B to the S2,3 states of the water oxidation complex in the dark interval between two consecutive flashes or under very low, continuous light. Such charge re-transfer from Q·− B to the S2,3 states may generate 3 P680 and in turn 1 O2, thereby causing damage to D1 protein (Keren et al. 1995, 1997). In another very low light study, consecutive flashing light with long dark intervals which gave very low cumulative photon exposure of pea leaves, photosynthesis did not reach steady-state (Shen et al. 1996). Shen et al. (1966) showed that during photosynthetic induction, there was no reciprocity of irradiance and time of illumination. During photosynthetic induction only limited photoprotective strategies are developed: while the trans-thylakoid pH gradient conferred some photoprotection, the xanthophyll cycle was inoperative. Thus, PS II is more easily photoinactivated during photosynthetic induction, a phenomenon that may have relevance for understory leaves experiencing infrequent, brief sunflecks (Shen et al. 1996; Keren et al. 1997). The strongest evidence for our model (Figure 2) is based on photosynthesis where reciprocity between irradiance and the time of illumination is observed, that is, with steady-state photosynthesis; however, it is not relevant for photosynthetic induction, nor for comment on the low-light syndrome mechanism (Ohad et al. 1994; Keren et al. 1995, 1997; Shen et al. 1996). One mechanism only for in vivo photoinactivation of PS II during steady-state photosynthesis Several recent findings strongly indicate that there is only one mechanism for in vivo photoinhibition during steady-state photosynthesis. These include the following: 1. PS II photoinactivation shows a reciprocity between irradiance level and time of illumination in acclimated pea leaves, demonstrating that the photoinactivation of PS II is a light dosage event depending on the number of photons absorbed and not the rate of photon absorption (Park et al. 1995, 1996a, b). Thus, it is extremely unlikely that one mechanism is operational at limiting to saturating light and donor-side or acceptor-side would be operational at high sustained light. Being a light dosage effect, PS II inactivation in vivo occurs in limiting, saturating and super-saturing light (Park
et al. 1995, 1996a, b; Anderson et al. 1997). Currently, this observed reciprocity is the best evidence to strongly suggest that only one mechanism predominates in photoinhibion in vivo during steady-state photosynthesis. 2. The rate constant of photoinhibition, measured from fluorescence changes of lincomycin-treated pumpkin leaves, was directly proportional to irradiance from very low to high light (6.5 to 2000 µmol m−2 ) (Tyystjärvi and Aro 1996). This finding was interpreted as a constant quantum yield for photoinhibition, independent of light intensity; if this interpretation is valid, only a single mechanism for photoinhibition in vivo would be needed (Tyystjärvi and Aro 1996). 3. The rate constant for D1 protein photodamage was also a linear function of the growth irradiance of Dunaliella salina (Baroli and Melis 1996), again suggesting a single in vivo mechanism for photoinhibition. 4. Importantly, the acceptor-side mechanism requires two consecutive photochemical events since electron flow from QA − is gated at the secondary electron acceptor, QB , and its double reduction requires the supply of two electrons from the donor side of PS II. However, target theory has established unequivocally that only one photon is required for the photoinhibition of light-acclimated pea leaves (Sinclair et al. 1996; Park et al. 1977), strongly suggesting that the double reduction of QA − does not occur during in vivo photoinhibition. 5. At the lowest intensity at which we have tested the validity of the reciprocity law (100 µmol m−2 s−1 ) (Park et al. 1996a), each PS II on average absorbed about 50 photons per second. Thus, the average interval between the trapping of photons at each PS II is about 20 ms. Since the lifetime of QA − is about 0.3 ms, a multiple-hit mechanism which requires two photons for the double reduction of QA − as proposed in the acceptor-side mechanism is extremely unlikely to be operational in vivo. These points taken together strongly suggest there may be no valid reason to invoke different mechanisms for in vivo photoinhibition during steady-state photosynthesis. Under infrequent flashing light or very low continuous illumination when the light is insufficient for steady-state photosynthesis, i.e. during photosynthetic induction, a different mechanism may be operational as proposed for the low-light syndrome (Keren et al. 1995, 1997).
pres775b.tex; 30/06/1998; 12:21; p.6
7 Unifying model for the photoinactivation of PS II in vivo We consider that the photoinactivation of PS II in vivo during steady-state photosynthesis is an inevitable consequence arising from the presence of the primary radical pair, P680+ Pheo− . The radical pair formed by charge separation can be dissipated by charge stabilization with no loss of PS II function, charge recombination, or reaction of P680+ with a neighbouring amino acid will occur with a certain probability, resulting in photodamge to PS II. But the probabilty must be very low as the quantum requirement for PS II photoinactivation ranges from 106 to 107 photons per PS II inactivated, although these values can be altered by varying the physical and functional antenna sizes of PS II, the extents of multiple photoprotective strategies as well as environmental conditions (Anderson et al. 1997). We propose that PS II photoinactivation is a function of primary radical pair concentration: the more radical pairs generated or maintained, the greater the extent of PS II photoinactivation. Figure 2 shows a unifying model for the photoinactivation of PS II in vivo based on the reversible exciton-radical pair equilibrium model of Schatz et al. (1988), as further elaborated by Dau (1994), and Schreiber and Krieger (1996). Energy transfer and trapping are in an equilibrium that involves both the exciton and the primary radical pair, P680+ Pheo− (Schatz et al. 1988). Since the primary charge separation is reversible and P680 is a shallow trap, an exciton may visit a reaction center several times before being finally trapped (Schatz et al. 1988), an important point often overlooked in photoinhibition studies. Our model (Figure 2) incorporates the competing processes of utilization of photons and the wastage of photons as heat dissipation in PS II against the background of the probability of rare but finite inevitable photoinactivation occurring with one photon as discussed. Light absorbed by PS II antenna including P680 will be dissipated by three reactions: (i) excitation trapping (k1 ); (ii) fluorescence (kF ); and (iii) nonphotochemical dissipation of excess photons as heat in PS II. This occurs mainly by PS II antenna dissipation (kDA ) (Horton et al. 1996), but 3 P680 quenching (k2TD ) and possibly limited reaction center quenching (k2D ) are also feasible (Weis and Berry 1987) (Figure 2). There are two situations to consider. When electron transport is fully functional and photons are being
fully utilised, the primary radical pair, P680+ Pheo− , will decay by four reactions: (i) k−1 , charge recombination; (ii) k2Q , charge stabilization by electron transfer to QA ; (iii) k2D , nonradiative decay to ground state; and (iv) k2T , spin dephasing to the triplet state of the radical pair followed by formation of triplet P680. On the other hand, when photon utilization is inhibited and QA is prereduced or removed, electron transfer is blocked at the level of P680+ Pheo− . Under these conditions, P680+ Pheo− decays by three ways (Figure 2): (i) to the ground state directly (k2D); (ii) to the excited singlet state (k−1 ); and (iii) to the triplet state, 3 P680 (k2T ). The triplet state is formed after singlet-triplet mixing in the radical pair (van Mieghem et al. 1995). Despite these energy dissipative pathways for excess photons, D1 protein damage is presumed to arise by two oxidative pathways shown by dotted lines in Figure 2, either from P680+ , the strongest oxidant in photosynthesis, and the cation radical Tyrz +, or from triplet P680, which in the presence of O2, forms highly toxic singlet oxygen, 1 O2 . We strongly favour the oxidative pathway of P680+ being the main, perhaps only, pathway in vivo for Photosystem II from point (1) detailed below. We advance below arguments that the critical factor for the photoinactivation of PS II depends on the generation and maintenence of the primary radical pair [P680+ Pheo− ] and the various ways charge recombination is regulated under varying environmental conditions. Experimental evidence relevant to the model for PS II photoinactivation in vivo 1. Enhanced radical pair concentration may lead to damage to D1 protein, via (i) the oxidative pathway of P680+ or (ii) due to the triplet state of P680 generating singlet oxygen (Figure 2). We suggest P680+ as the more likely starting route to D1 protein degradation for six reasons. i) Electron transfer from the lowest singlet state of P680 to Pheo takes about 10 ps, then Pheo− transfers its electron to QA within 400 ps, while the reduction of P680+ occurs within 20–200 ns by Tyrz (van Mieghem et al. 1995). The slower electron transfer from the secondary donor to P680+ strongly favours the likelihood of the damaging species for D1 protein being P680+ rather than triplet 3 P680. ii) Van Mieghem et al. (1995) using PS II membrane fragments showed that the yield of the primary
pres775b.tex; 30/06/1998; 12:21; p.7
8 radical pair is high, irrespective of the redox state of QA , and its decay is faster with QA − than with QA H2 . They also demonstrated that the yield of the triplet state 3 P680 was high with both QA − and QA H2 , but significantly the rate of triplet P680 decay was two orders of magnitude greater in the presence of reduced QA . Thus, the rate of triplet state P680 quenching is strongly stimulated as more QA is reduced (van Mieghem et al. 1995). The mechanism of triplet P680 quenching in the presence of QA − is not understood but it is unlikely to involve carotenoid quenching. The active quenching of triplet P680 may represent yet another protective mechanism in PS II. Thus, short-lived triplet P680 may be more likely to be quenched also in vivo and less likely to react with oxygen; and may not necessarily increase the danger of irreversible damage to PS II via singlet oxygen. iii) There is no evidence to demonstrate that indeed triplet state P680 is formed in leaves or algae, although there are as yet no ways to measure its concentration in vivo. Similarly, there is scant evidence for the formation of singlet oxygen in vivo under realistic light treatments comparable to those obtained in nature. The many in vitro studies relating to triplet P680 formation used isolated PS II membrane fragments under conditions of very high photon exposure, much greater than encountered by most PS IIs in leaves, and these isolated complexes also have only limited photoprotective strategies. iv) It seems to us very unlikely that oxygen is evolved in close vinicity of P680, otherwise the production of singlet oxygen via triplet P680 would be very difficult, perhaps impossible, to control. In this regard it is important to consider the beautiful structural evidence that cytochrome c oxidase has an oxygen channel allowing oxygen to reach its site of reduction (Tsukihara et al. 1996; Riistama et al. 1996). Within PS II complex, the unique oxidation of water must also occur at a very specific site (Wydrzynski et al. 1996). By analogy to the structure of cytochrome c oxidase, we suggest that the oxygen evolved from water will be released within PS II complex via a specific oxygen channel thereby preventing its reaction with transient triplet P680 in vivo under normal conditions, and even during photoinhibition unless very major secondary effects occur. On the other hand, with isolated PS II reaction centres are bombarded
with enormous photon exposure the situtation is very different. Large conformational changes of PS II reaction centers could prevent efficient 3 P680 quenching, while free acess to oxygen could permit the formation of singlet oxygen. v) Given the diminished likelihood of photoinactivation of PS II by singlet oxygen as argued above in points (i) to (iv), one is led to consider whether other reactive oxygen species may cause PS II photoinactivation. Reactive oxygen species may be generated through the Asada or Mehler ascorbate peroxide pathway (Asada 1994), potentially leading to photoinactivation of PS II. As H2 O2 is generated via electron donation to molecular oxygen in PS I, one may expect that highly oxidising radicals such as • OH arising from a Fenton reaction would damage PS I locally ahead of PS II, especially as most PS I complexes are not in close contact with PS II complexes in higher plants. Under chilling conditions when the scavenging enzyme systems are presumably sluggish, this is indeed the case in chilling-sensitive plants (see Sonoike 1996). However, at more favorable temperatures, it is expected that scavenging enzyme systems remove reactive oxygen species sufficiently rapidly, so that even PS I is adequately protected. Thus, at 23 ◦ C, PS II in tobacco leaf discs can be extensively photoinactivated (in the presence of lincomycin to inhibit repair) without any effect on the function of PS I complex or the cytochrome b/f complex (Chow and Hope submitted). Therefore, we suggest that the photoinactivation of PS II may not be primarily caused by oxygen-containing free radicals, nor, as argued above, by singlet oxygen. vi) On the other hand, P680+ , which has a very high oxidising potenial of 1.175 V (Klimov et al. 1979) required to extract electrons from water, is formed every time a primary radical pair is generated. If the lifetime of P680 is lengthened as may occur when multiple charge recombinations occur when forward electron transport is blocked (as detailed in points 2 and 3 below), there is an inherent but rare probability it will react with neighbouring amino acids. Taken together these points provide compelling evidence which strongly favours P680+ as the damaging species for D1 protein, during steadystate photosynthesis, rather than singlet oxygen (formed from triplet P680) or other oxygencontaining free radicals.
pres775b.tex; 30/06/1998; 12:21; p.8
9 2. Any model for PS II photoinactivation in vivo has to accommodate the remarkable yet counterintuitive reciprocity between irradiance and time of illumination for PS II inactivation in vivo. Given the clear demonstration that the probability of PS II inactivation in vivo varies according to the state of acclimation of PS II to growth irradiance, to the capacity of varying interacting photoprotective strategies, and to other factors influencing stress tolerance, we considered whether the critical factor might be the concentration of the primary radical pair (P680+ Pheo− ) and the various ways it can recombine harmlessly or lead to potential damage (Anderson et al. 1997) (Figure 2). When charge stabilization is prevented by blocking of electron transport, there needs to be a central component capable of being regulated by multiple diverse pathways; the obvious candidate is the primary radical pair. For antenna quenching there are two pathways for exciton dissipation: either (i) heat is directly dissipated via excitons that have not visited a reaction centre or alternatively, (ii) dissipation takes place after formation of a primary radical pair, followed by delivery of an exciton back to the antenna bed by charge recombination in an exciton-radical pair equilibrium (Horton et al. 1996). While the former point (direct dissipation of an exciton prior to visiting a reaction centre) is often assumed, the role of charge recombination has been largely ignored when dealing with variable fluorescence as pointed out by Schreiber and Krieger (1996); this is true also for PS II photoinactivation, except for the low-light syndrome (Keren et al. 1995, 1997). Possibly, the equilibrium between excitons and the primary radical pair (and associated charge recombination) is responsible for the reciprocity between irradiance and duration of illumination. That is, under a given set of photoprotective conditions, each photon, regardless of the rate of photon arrival, has the same chance of causing a charge separation, rather than some photons being dissipated from the antenna without having visited a reaction centre. Further, for reciprocity to hold, it is also imperative that plants are able to balance any irradiance-dependent damage with protection, over a wide irradiance range (Anderson et al. 1995). 3. For reciprocity to hold, we postulate that almost every absorbed photon needs to be involved in forming a primary radical pair, one partner (P680+ ) of the pair being capable of photoinactivation of PS II. Even excitons which are dissipated as heat in antenna pigment molecules may have made a number of visits to a re-
action centre, each time forming a radical pair, prior to final dissipation in the antenna. It has been generally assumed that photoinhibition in vivo occurs as a result of excessive excitation pressure on PS II with QA being mainly reduced due to sustained high light (Aro et al. 1993; Ohad et al. 1994; Huner et al. 1996). However, overreduction of QA is not a prequisite for in vivo photoinhibition since plants begin to lose PS II function when only 40% of their PS II centres are closed (Öquist et al. 1992). In this framework, the enhanced photoinactivation of PS II as a linear function of QA reduction at a fixed irradiance (Öquist et al. 1992) can now be understood. We suggest that blockage of forward electron flow from QA − to QB does not inhibit stable charge separation, but rather helps maintain exciton-primary radical pair equilibrium for a longer time. This increase of the life-time of P680+ Pheo− , in the presence of singly reduced QA when electron flow beyond QA is inhibited, could promote damage due to greater P680+ . In other words, when electron flow beyond QA is inhibited, charge separation is followed by charge recombination, giving rise to excitation of the antenna pigments which are de-excited via charge separation and so on. In this way the repeated generation of P680+ in the presence of singly reduced QA , as opposed to fewer charge separations per photon when QA is oxidized, increases the probability that an inadvertant oxidation causes photodamage. Baroli and Melis (1996) also suggested there is an inherent probability of PS II photodamage every time there is a charge separation between P680 and pheophytin, though they did not specify a mechanism of damage. Quenching of chlorohyll fluoresecence during steady-state photosynthesis may be caused by utilisation of photons by photosynthetic electron transport (photochemical quenching coefficient qP) or by events concerned with photon wastage by non-photochemical quenching (NPQ). The parameter, 1–qP, is an indicator of the amount of reduced QA (Huner et al. 1996). At any fixed irradiance, the higher the value of (1 – qP), the greater is the extent of photoinactivation of PS II (Ögren 1991; Öquist et al. 1992). It is remarkable that most leaves are able to balance energy consumed by PS II with energy dissipated by non-radiative dissipation, monitored by the fluorescence parameters, photochemical quenching or its complement (1 – qP) and non-photochemical quenching (NPQ), respectively. The value of (1 − qP), however, increases with increase in irradiance. Why then, is there reciprocity between irradiance and duration
pres775b.tex; 30/06/1998; 12:21; p.9
10 of illumination – that is an equivalence between high and low irradiances that give the same photon exposure? To answer this question, note that as irradiance is increased, not only is (1 – qP) increased, but the transthylakoid 1pH is also increased, as assayed by an increase in non-photochemical quenching (NPQ) of chlorophyll fluorescence or its equivalent qE (Horton et al. 1996). Hence, an increase in irradiance also leads to enhanced dissipation of excitons in the antenna, thereby minimizing the total time over which P680+ exists through exciton-radical pair equilibrium. That is, there are compensating effects when irradiance is increased: viz. greater PS II photoinactivation associated with greater (1 – qP), together with enhanced photoprotection through increased NPQ. Significantly, for most leaves the quotient (1 – qP)/NPQ is approximately constant over a wide range of irradiance up to full-sun levels, the absolute value of (1 – qP)/NPQ being dependent on the irradiance under which plants had been grown (Park et al. 1996b; Shen et al. 1996). This means that during steady-state photosynthesis, PS II has the ability to balance light supply with light consumption over the entire wide irradiance range tested. During photosynthetic induction, however, this is not the case and the quotient (1 – qP)/NPQ varied by two orders of magnitude and reciprocity was not observed (Shen et al. 1996). 4. Target theory analysis demonstrates that only one photon is required for PS II photoinactivation in vivo (Sinclair et al. 1996; Park et al. 1997). Thus, the proposed acceptor-side mechanism which requires the double reduction of QA isno longer relevant (Vass et al. 1992; Ohad et al. 1994). The likelihood of double reduction of QA in vivo is also not favored on the time scale: even at 2000 µmol photons m−2 s−1 , an individual PS II has a chance to receive only one photon about every ms which is far slower than reoxidation of QA − ; charge recombination is favored. Moreover, there is as yet no evidence for the double reduction of QA in vivo in steady-state photosynthesis. As seen in Figure 2, the arrival of one electron at QA already has a profound effect on PS II as 3 P680 is very effectively quenched (van Mieghem et al. 1995). Even with O2 -evolving PS II membrane fragments, there was no evidence to suggest that QA H2 was formed during photoinhibition, and the authors inferred that a mechanism which comprises double reduction and protonation of QA is ‘only of marginal-if any-relevance for PS II photoinhibition under aerobic and anaerobic conditions’ (Napiwotzki et al. 1997).
Figure 3. Varying extents of PS II photoinactivation in pea leaves at a given exposure of 4.7 mol photons m−2 . The fraction of functional PS IIs were determined as repetitive flash-induced O2 evolution and expressed as mmol PS II (mol Chl)−1 : 100% in low-, medium- and high-light pea correspond to 2.30, 2.66, 2.98, respectively (Park et al. 1996a). Note that low-light grown plants are the most susceptible to light stress. Further, the probability of PS II photoinactivation is variable depending which photoprotective strategies which have been inhibited by pretreatment of pea leaves prior to the photoinactivation treatment: D1 protein synthesis (inhibited by lincomycin); transthylakoid pH gradient (partly inhibited by nigericin) and the protective xanthophyll cycle (inhibited by dithiothreitol) (Park et al. 1996a). The photochemical utilization of absorbed photons was also varied by the availability of electron acceptors during illumination in various gas mixtures (Park et al. 1996c). Again, the probabilty of PS II photoinactivation is not constant but dependent also on the extent of photochemical utilization of absorbed photons.
5. The use of photons in photochemistry can be altered by variations in the availability of final electron acceptors in vivo, thereby altering the extent of PS II photoinactivation in vivo (Park et al. 1996c). Accompanying carbon assimilation, photosynthetic electron transport to oxygen also occurs, via the oxygenase activity of Rubisco (photorespiration) as well as the Mehler reaction. The probability of PS II inactivation in medium-light peas with the same photon exposure of 4.7 mol photons m−2 is differently affected with illumination of leaves in various gases, as shown in Figure 3. As expected, PS II is most inactivated (63%) in N2 , since electron acceptors are practically absent. In low (2%) or high (60%) O2 without any CO2 , less PS II is inactivated compared to N2 , demonstrating that electron transport to O2 in vivo is important in utilising photons and mitigating against PS II photoinactivation. While 47% of PS II were photoinactivated in 2% oxygen, only 38% were inactivated in 60% oxygen, demonstrating that O2 is involved as an electron acceptor, rather than as a substrate for 1 O2 production. We postulate that the greater photoinactivation of PS
pres775b.tex; 30/06/1998; 12:21; p.10
11 II under low O2 or CO2 (Figure 3) is attributable to limited charge stabilization (photon utilization) due to a limitations in availability of final electron acceptors. 6. The state of PS II acclimation to light is also very important for the susceptibility of plants to photoinhibition. Relative to PS I, shade and low-light plants have fewer, but larger PS IIs than sun and high-light plants (Anderson et al. 1995). Shade and low-light plants are more photoinhibited at the same photon exposure than sun and high-light plants (e.g. Figure 3), because they have less capacity for both exciton dissipation at the PS II antenna, or exciton utilization by photosynthesis, resulting in more reduced QA − (Park et al. 1996b). More reduced QA , will allow more charge recombinations to occur and lengthen the exciton-radical pair equilibrium (as detailed in point 3 above) in shade and low-light plants leading to an increased probability of formation of P680+ and damage to D1 protein via the oxidative pathway. 7. The use of leaves which had been pretreated with inhibitors also demonstrated that the photoinactivation of PS II is a probability event dependent on the sum of multiple, interacting photoprotective strategies. Nigericin, an uncoupler of photosynthesis, at partly uncoupling concentrations, caused an increase in the number of non-functional PS IIs at all photon exposures compared to control acclimated pea leaves (Park et al. 1996a): after a photon exposure of 4.7 mol m−2 , 28% of PS IIs are inactivated compared to 10% in control leaves (Figure 3). With partial inhibition of proton-mediated dissipative processes by nigericin, the enhanced photoinactivation is attributable to decreased rate constants of nonradiative energy loss processes (kDA and k2D ) as well as k2T , and the resulting increase in [P680+Pheo− ] will stimulate the oxidative process initiated by P680+ . In contrast, when the protective xanthophyll cycle (violaxanthin conversion to antheraxanthin and zeaxanthin) (Demmig-Adams and Adams, 1992) is inhibited, only the rate constant for antenna quenching (kDA ) is affected (Figure 2). The effect of low concentrations of dithiothreitol, which specifically inhibited the protective xanthophyll cycle and no zeaxanthin was formed (Park et al. 1996b), led to an 18% increase in non-functional PS IIs compared to control plants (10%). This result which clearly demonstrates that the xanthophyll cycle is only part of protection afforded by the transthylakoid pH gradient, can be readily explained by our model (Figure 2).
In conclusion We present a unifying model for Photosystem II photoinactivation in vivo under steady-state photosynthesis which depends on the generation and/or maintenance of increased concentrations of the primary radical pair and the different ways charge recombination is regulated under varying environmental conditions. We propose that one dominant molecular mechanism in vivo takes place during steady-state photosynthesis with one photon inactivating PS II over the entire irradiance range: limiting, saturating, and even sustained high light. Although charge recombination of the primary radical pair has the potential to form damaging triplet P680, we suggest that rather than react with oxygen to form singlet O2 , 3 P680 is effectively quenched in the presence of singly reduced QA (QA − ). This undefined quenching mechanism may be yet another photoprotective strategy of PS II. We propose that one partner of the primary radical pair, P680+ , is the more likely active species to cause oxidative damage to D1 protein. If P680+ is the predominant agent responsible for the photoinactivation of PS II, a ‘necessary evil’ required for the photooxidation of otherwise stable water molecules, then the existence of an exciton-primary radical pair equilibrium can be easily rationalized. This exciton-radical pair equilibrium ensures that, all else being equal, P680+ persists for shorter periods of time than in the absence of such an equilibrium: however, if electron donation from the redox-active tyrosine (Tyrz ) in PS II reaction center to P680+ is delayed, charge recombination will convert the oxidising potential of P680+ back to excitonic energy of pigment molecules.
Acknowledgements We ( Y.-I. P., J.M.A.) are very grateful to Prof. Gunnar Öquist for support, to Profs. Peter Horton and Gunnar Öquist and Dr Tom Wydrzynski for discussion, and as always to our other research collaborators.
References Altmann KI, Gerber GB and Okada S (1970) Radiation Biology, Vol I, Cells, pp 329-336. Academic Press, New York Anderson JM and Aro E-M (1994) Grana stacking and protection of Photosystem II in thylakoid membranes of higher plant leaves under sustained high irradiance: An hypothesis. Photosynth Res 41: 315–326
pres775b.tex; 30/06/1998; 12:21; p.11
12 Anderson JM, Chow WS and Park Y-I (1995) The grand design of photosynthesis: Acclimation of the photosynthetic apparatus to environmental cues. Photosynth Res 46: 129–139 Anderson JM, Park Y-I and Chow WS (1997) Photoinactivation and photoprotection of Photosystem II in nature. Physiol Plant 100: 214–223 Aro E-M, Virgin I and Andersson B (1993) Photoinhibition of Photosystem II. Inactivation, protein damage and turnover. Biochim Biophys Acta 1143: 113–134 Asada K (1994) Production and action of active oxygen species in photosynthetic tissues. In: Foyer CH (ed) Causes of Photooxidative Stress and Amelioration of Defence System in Plants, pp 77–104. CRC Press, Baton Rouge Barber J (1995) Molecular basis of the vulnerability of Photosystem II to damage by light. Aust J Plant Physiol 22: 201–208 Baroli I and Melis A (1996) Photoinhibition and repair in Dunaliella salina acclimated to different growth irradiances. Planta 198: 640–646 Boekema EJ, Hanaker B, Bald D, Kruip J, Nield J, Boonstra AF, Barber J and Rögner M (1995) Supramolecular structure of the Photosystem II complex from green plants and cyanobacteria. Proc Natl Acad Sci USA 92: 175–179 Cleland RE, Melis A and Neale PJ (1986) Mechanism of photoinhibition: photochemical reaction center inactivation in system II of chloroplasts. Photosynth Res 9: 79–88 Chow WS (1994) Photoprotection and photoinhibitory damage. Adv Mol Cell Biol 10: 151-196 Dau H (1994) Molecular mechanisms and quantitative models of variable Photosystem II fluorescence. Photochem Photobiol 60: 1–23 Demmig-Adams B and Adams WW (1992) Photoprotection and other other responses of plants to high light stress. Annu Rev Plant Phys Plant Mol Biol 43: 599-626 Horton P, Ruban AV and Walters RG (1996) Regulation of light harvesting in green plants. Annu Rev Plant Physiol Plant Mol Biol 47: 655-684 Huner NPA, Maxwell DP, Gray GR, Savitch LV, Kroll M, Ivanov and Falk S (1996) Sensing environmental temperature change through inbalance between energy supply and energy consumption: Redox state of Photosystem II. Physiol Plant 98: 358–364 Hutchinson F and Pollard E (1961) Target theory and radiation effects on biological molecules. In: Errera M and Foressberg A (eds) Mechanisms in Radiobiology, Vol I, pp 71–92. Academic Press, New York Johnson GN, Rutherford AW and Krieger A (1995) A change in the midpotential of the quinone QA in Photosystem II associated with photoinactivation of oxygen evolution. Biochim Biophys Acta 1229: 202–207 Jones LW and Kok B (1966) Photoinhibition of chloroplasts reactions. Kinetics and action spectra. Plant Physiol 41: 1037–1043 Keren N, Gong H and Ohad I (1995) Oscillations of reaction centerD1 protein degradation in vivo induced by repetitive flashes. Correlation between the level of RCII-QB − and protein degradation in low light. J Biol Chem 270: 806–814 Keren N, Berg A, van Kan PJM, Levanon H and Ohad I (1997) Mechanism of Photosystem II photoinactivation and D1 protein degradation at low light: The role of back electron flow. Proc Natl Acad Sci USA 94: 1579–1584 Kettunen R, Tyystjärvi E and Aro E-M (1996) Degradation pattern of Photosystem II reaction center protein D1 in intact leaves: The major photoinhibition-induced cleavage site on A1 polypeptide is located amino terminally of the DE loop. Plant Physiol 111: 1183–1190
Klimov VV, Allakhverdiev SI, Demeter S and Krasnovskii AA (1979) Pheophytin reduction in Photosystem II of chloroplasts in relation to oxidation-reduction potential of the medium. Dokl Akad Nauk SSR 249: 227–230 Krieger A and Weis E (1993) The role of calcium in the pHdependent control of Photosystem II. Photosynth Res 37: 117– 130 Melis A (1991) Dynamics of photosynthetic membrane composition and function. Biochim Biophys Acta 1058: 87–101 Nagy L, Balint E, Barber J, Ringler A, Cook KM and Maroti P (1995) Photoinhibition and law of reciprocity in photosynthetic reactions of Synechocystis sp. PCC 6803. J Plant Physiol 145: 410–415 Napiwotzki A, Bergmann A, Decker K, Legall H, Eckert H-J, Eichler H-J and Renger G (1997) Acceptor side photoinhibition in Photosystem II: On the possible effects of the functional integrity of the PS II donor side of photoinhibition of stable charge separation. Photosynth Res 52: 199–213 Ohad I, Keren N, Zer H, Gong H, Mor TS, Gal A, Tal S and Domovich Y (1994) In: Baker NR and Bowyer JR (eds) Photoinhibition of Photosynthesis: From Molecular Mechanisms to the Field, pp 161–177. Bios Scientific Publications, Oxford Ögren E (1991) Prediction of photoinhibition of photosynthesis from measurements of fluorescence quenching components. Planta 184: 538–544 Öquist G, Chow WS and Anderson JM (1992) Photoinhibition of photosynthesis represents a mechanism for the long-term regulation of photosynthesis. Planta 186: 450–460 Park Y-I, Chow WS and Anderson JM (1995) Light inactivation of functional Photosystem II in leaves of peas grown in moderate light depends on photon exposure. Planta 196: 401–411 Park Y-I, Anderson JM and Chow WS (1996a) Photoinactivation of functional Photosystem II and D1-protein synthesis in vivo are independent of the modulation of the photosynthetic apparatus by growth irradiance. Planta 198: 300–309 Park Y-I, Chow WS, Anderson JM and Hurry VM (1996b) Differential susceptibility of Photosystem II to light stress in light-acclimated peas leaves depends on the capacity for photochemical and non-radiative dissipation of light. Plant Sci 115: 137–149 Park Y-I, Chow WS, Osmond CB and Anderson JM (1996c) Electron transport to oxygen mitigates against the photoinactivation of Photosystem II in vivo. Photosynth Res 50: 23–32 Park Y-I, Chow WS and Anderson JM (1997) Antenna size dependency of photoinactivation of Photosystem II in light-acclimated pea leaves. Plant Physiol 115: 151–157 Riistama S, Puustinen A, Garcia-Horsmann A, Itwata S, Michel H and Wikström (1996) Chanelling of dioxygen into the respiratory enzyme. Biochim Biophys Acta 1275: 1–4 Schatz GH, Brock H and Holzwarth AR (1988). Kinetic and energy model for the primary processes of Photosystem II. Biophys J 54: 397–405 Schreiber U and Krieger A (1996) Two fundamentally different types of variable fluorescence in vivo. FEBS Lett 397: 131–135 Shen Y-K, Chow WS, Park Y-I and Anderson JM (1996) Photoinactivation of Photosystem II by cumulative exposure to short light pulses during the induction period of photosynthesis. Photosynth Res 47: 51–59 Sinclair J, Park Y-I, Chow WS and Anderson JM (1996) Target theory and the photoinactivation of PS II. Photosynth Res 50: 33–40 Sisson WB and Caldwell MM (1977) Atmospheric ozone depletion: Reduction of photosynthesis and growth of a sensitive
pres775b.tex; 30/06/1998; 12:21; p.12
13 higher plant exposed to enhanced UV-B radiation. J Expt Bot 28: 691–705 Sonoike K (1996) Photoinhibition of Photosystem I: Its physiological significance in the chilling sensitivity of plants. Plant Cell Physiol 37: 239–247 Tsukihara T, Aoyama H, Yamashita E, Tomizaki T, Yamaguchi H, Shinzawa-Itoh K, Nakashima R, Yaono R and Yoshikawa S (1996) The whole structure of the 13-subunit oxidized cytochrome c oxidase at 2.8 A. Science 272: 1136–1144 Tyystjärvi E and Aro E-M (1996) The rate constant of photoinhibition, measured in lincomycin-treated leaves, is directly proportional to light intensity. Proc Natl Acad Sci USA 93: 2213–2218 Tyystjärvi E, Kettunen R and Aro E-M (1991) Small lightharvesting antenna does not protect from photoinhibition. Plant Physiol 97: 477–483 Tyystjärvi E, Koivuniemi A, Kettunen R and Aro E-M (1994) The rate constant of photoinhibition in vitro is independent of the
antenna size of Photosystem II but depends on temperature. Biochim Biophs Acta 1186: 177–185 van Mieghem F, Brettel K, Hillmann B, Kamlowski A, Rutherford WA and Schlodder E (1995) Charge recombination reactions in Photosystem II. I. Yields, recombination pathways, and kinetics of the primary pair. Biochem 34: 4798–4813 Vass I, Styring S, Hundahl T, Koivuniemi A, Aro E-M and Andersson B (1992) Reversible and irreversible intermediates during photoinhibition of Photosystem II: Stable reduced QA species promote chlorophyll triplet formation. Proc Natl Acad Sci USA 89: 1408–1412 Weis A and Berry J (1987) Quantum efficiency of Photosystem II in relation to energy dependent quenching of chlorophyll fluorescence. Biochim Biophys Acta 894: 198–208 Wydrzynski T, Hillier W and Messinger J (1996) On the functional significance of substrate accessibility in the photosynthetic water oxidation mechanism. Physiol Plant 96: 342–350
pres775b.tex; 30/06/1998; 12:21; p.13
