PhotosynthesisResearch 47: 167-173, 1996. t~) 1996KluwerAcademic Publishers. Printedin the Netherlands. Regular paper

Slow exciton trapping in Photosystern II: A possible physiological role R o b e r t C. J e n n i n g s , F l a v i o M. Garlaschi, L a u r a Finzi & G i u s e p p e Z u c c h e l l i Centro CNR sulla Biologia Cellulare e Molecolare delle Piante, Dipartimento di Biologia, Universit~ di Milano, via Celoria 26, 20133 Milano, Italy Received24 July 1995;acceptedin revisedform6 December1995 Key words: antenna-trap equilibration, photoinhibition, photoprotection, Photosystem II, trapping time

Abstract

Photosystem II, which has a primary photochemical charge separation time of about 300 ps, is the slowest trapping of all photosystems. On the basis of an analysis of data from the literature this is shown to be due to a number of partly independent factors: a shallow energy funnel in the antenna, an energetically shallow trap, exciton dynamics which are partly 'trap limited' and a large antenna. It is argued that the first three of these properties of Photosystem II can be understood in terms of protective mechanisms against photoinhibition. These protective mechanisms, based on the generation of non photochemical quenching states mostly in the peripheral antenna, are able to decrease pheophytin reduction under conditions in which the primary quinone, QA, is already reduced, due to the slow trapping properties. The shallow antenna funnel is important in allowing quenching state-protective mechanisms in the peripheral antenna. Abbreviations: chl - chlorophyll; PS I - Photosystem I; PS I I - Photosystem II; QA - the primary quinone acceptor; RC-reaction centre; R T - r o o m temperature Introduction

In the photosystems of photosynthetic organisms light is absorbed by a large array of antenna pigments and transferred rapidly to a special reaction centre pigment (primary donor) where the photochemical reduction of a primary acceptor occurs. The time scale of energy transfer between nearest neighbour antenna pigments is thought to be in the 0.2-5 ps range for higher plant photosystems (for recent reviews see van Grondelle et al. 1994 and Jennings et al. 1996). The time between photon absorption and primary photochemical trapping (Ttr) for bacterial photosystems and plant Photosystem I is in the 50-100 ps range (Trissl 1993). However, in the case of Photosystem II rtr is approximately 300 ps (Roelofs et al. 1988; Leibl et al. 1989), thus making PS II the slowest of all photosystems to date examined. The biological advantage associated with rapid trapping is clearly to be found in the ensuing high quantum efficiency of primary photochemical charge

separation. Thus, the relatively slow trapping of PS II is at the expense of maximising quantum efficiency. It is therefore interesting to analyse this phenomenon in order to understand its physiological importance. In recent years, considerable progress has been made in understanding antenna energy transfer rates and equilibration processes (Kfihlbrandt and Wang 1991; Jennings et al. 1993; D u e t al. 1994) as well as primary RC trapping rates (Roelofs et al. 1993; Wiederrecht et al. 1994) in PS II, thus permitting an analysis of the various factors leading to slow trapping. In the first part of this paper we will present such an analysis in terms of some thermodynamic and kinetic properties of PS II. Successively the physiological implications of these properties will be discussed.

168 Table 1. Absorption (A) and room temperature excited state population (S) in PS II in terms of the chl-protein complexes. The excited state population values have been determined on the basis of a Boltzmann excitation distribution amongst the spectral forms present in the chlprotein complexes (for details see Jennings et al. 1993, 1994) Chl-protein complex

A (% total)

LHC II CP29 CP26 CP24 CP43 CP47 D1/D2/cytb559

55.3 6.9 6.3 3.0 12.3 12.4 3.9

S (% total)

48.8 7.4 5.9 2.6 14.2 15.6 5.0

are present in the core complexes. The free energy differences at RT, calculated from the above S/A values, are (Jennings et al. 1993, 1994): peripheral aO=--0.29kT core AO=--0.07kr Dl/D2/cyfib559 antenna antenna complex

S/A

0.88 1.07 0.94 0.87 1.15 1.26 1.28

Results and discussion

Thus, energy is expected to be substantially delocalised over the entire antenna at RT. It is therefore clear that the spectral forms are not organised to rapidly 'direct' excitation towards the traps of PS II.

ii) Reaction centre trapping - trap~diffusion limited This is a kinetic aspect which has important implications for the trapping rate. It is easily understood in terms of the following, simple three component model (k3 A k l ) P k2) I k-I

Slow trapping in PS H i) Antenna energy properties It has long been known, on the basis of absorption spectroscopy at cryogenic temperatures, that the chlorophylls associated with PS II can be divided into a number of pigment pools with different spectroscopic properties (French et al. 1972; Zucchelli et al. 1994), which are often referred to as spectral forms. This gave rise to the suggestion that these pigments might be arranged as an energy funnel in order to rapidly 'direct' excitation towards the RCS of PS II (Seely 1973b; Shipman and Housman 1979). Model simulations demonstrated that such an organisation could increase the transfer rate to RCs by a factor of five (Seely 1973a, b). This hypothesis has recently been directly investigated by analysing the chl spectral forms in all the chl-protein complexes which make up the peripheral and core antenna, including the D1/D2/cytb559 complex which binds the RC of PS II (Jennings et al. 1993, 1994). On the basis of equilibrium calculations of the excited state population of the spectral forms in each chl-protein complex, it was concluded that the PS II antenna pigment system is organised as a shallow energy funnel. The predicted excited state population at room temperature in each chl-protein complex, together with the calculated absorption population are summarised in Table 1 (for details of these calculations see Jennings et al. 1993, 1994) where it can be seen that excitation is well distributed over the entire antenna at equilibrium though slightly higher 'concentrations' (SIA values)

where A is the antenna system of PS II; P is P680; I is pheophytin. The k values are first order kinetic constants with kl representing the overall transfer rate from A to P and is equivalent to the 'first passage time' of Pearlstein (1982). This assumption of first order kinetics for kl is justified as model calculations for a large (230 chl sites), spectroscopically heterogeneous antenna such as PS II, display essentially single exponential trapping kinetics when pairwise energy transfer rates in the antenna are equal to or faster than the primary reaction centre trapping rate (A. Yu. Borisov, personal communication), k_ 1 is the rate of back transfer from P to A. Conceptually, this can be understood as a single energy transfer step out of P and will thus depend on the specific arrangement of spectral forms around P. In reality k_l may describe more than a single energytransfer step as the exciton moves out of antenna sites in 'direct contact' with P. This inexactness in defining k - l can be tolerated in such a simple, exciton trapping kinetic model for PS II, as it permits a description of energy flowing into and out of P in a straightforward way. k2 is the rate of pheophytin reduction and k3 is the rate of all trivial decay processes in A. The inclusion of other rate processes in this simple model, e.g. reduction of QA or primary charge recombination (Schatz et al. 1988) has been omitted as we are interested only in the primary trapping by pheophytin. When k2> k - i the situation is said to be diffusion limited and P-A equilibration does not occur. Initial attempts to model energy transfer in PS II were based on the assumption of diffusion limited trapping (Butler and Strasser 1977). However, more recently, Holzwarth and coworkers have suggested that PS II is in fact strongly trap limited (Schatz et al. 1988) with excitation flowing in and out of P680 about 20 times prior to trapping. Recently, the suggestion was also made on the basis of kinetic considerations that a mixed diffusion/trap limited situation exists (Jennings et al. 1995). In the following we have investigated this using the above simple three component kinetic model. The values of k2 (330 ns -1) and k3 (0.5 ns -1) were fixed respectively on the basis of widely accepted rates for pheophytin reduction (Schatz et al. 1988, Roelofs et al. 1993; Wiederrecht et al. 1994) and fluorescence lifetimes in PS II with closed reaction centres (Haehnel et al. 1982; Hodges and Moya 1986). k_ l values were varied to simulate various trap and diffusion limited situations, kl values were varied between 10-100 ns -1 which amply encompasses the kinetic range expected (a) on the basis of experimentally determined single site transfer rates (Du et al. 1994) and (b) A-P 'first passage' migration times calculated according to Pearlstein (1982) and Kudzmauskas et al.

(1983). From the Pearlstein formalism: rmig = 0.5 N f(N) rh

(1)

where rmig is the 'first passage' migration time to P680, N is the number of antenna pigments for RC, f(N) is the lattice function which was determined according to Kudzmauskas et al. (1983) and 7"his the pairwise hopping rate. Considering only the chla of PS II antenna N=170, f(N) = 0.91 for a square lattice and rh is taken as 1 ps. This latter value is in accord with recent time resolved measurements of single site transfer rates in LHCII (Du et al. 1994). In this way, rmig comes out as 77 ps. Thus the most reasonable rate values (kl) are between 10-20 ns -1. Calculations of trapping rates were performed using the analytical solution for the system of differential equations = k-i P - (kl -F k3)A dP ~ - = kl a - (k-i + k2)P obtained with the initial conditions A = 1, P = 0 and r l 1 = m + (m 2 - n2) ½ T21 = m - (mz - nz) ½ m = ~ 1 (kl + k-l + k2 + k3)

170 n = (k3 kl + k3 k2 + kl k2) ½ Results of these calculations are given in Figure 1, where it can be seen that, as expected, the overall trapping time is critically dependent on both kl and the k2/k-l ratio. For the experimentally determined trapping time of 300 ps and taking reasonable kl values (kl = 10-20 ns-l), the k2/k-i ratio lies between 0.18 and 0.42. Thus, energy is expected to flow in and out of P680 between 3-7 times prior to trapping and the exciton dynamics are expected to be determined by both the antenna diffusion rate and the RC trapping rate. The importance of this repeated back transfer of energy from P680 to the antenna is readily apparent, as in its absence the primary trapping rate would be determined almost exclusively by the antenna-RC migration rate (kl) and would thus probably be in the 50-100 ps time range.

iii) Antenna-trap energy gap Slow trapping in PS II is also associated with the fact that the primary donor P680, is energetically much closer to the antenna than the traps of other photosystems. This point has been recently discussed by Trissl (1993). For example, the energy gap between P680 and the antenna absorption maximum is about 0.5 kT, while that for PS I is about 2.5 kT. This means that the time integrated probability of P680 being excited (P*) at equilibrium is low with respect to the traps of other photosystems. The importance of this factor in helping to determine the overall trapping rate is readily obvious as the overall rate of charge separation (kes) is given by kcs = P* kpi where kpi is the rate constant for pheophytin reduction. As pointed out by Schatz et al. (1988), in the assumption of trap/antenna equilibration, P* may be calculated from the population weighted Boltzmann expression: P* = N - l eaE/kT, where N is the number of antenna sites per RC and AE is the antenna-trap energy gap. Usually this value is taken as the difference between the absorption maximum of the trap and that of the antenna chls. While this is certainly not the most correct way of determining AE, which should take into account the absorption/fluorescence mirror symmetry point of all chl spectral forms (Jennings et al. 1994), it is an approach which gives a reasonable approximation and therefore, for the sake of simplicity, will be used here. Thus, for PS II and considering only chlorophyll a (N =170, AE = 0.5 kT) we calculate P* ~ 0.01. On the other hand, for PS I (N = 190, AE = 2.5 kT) P* is 0.064. From this albeit oversimplified comparison of PS II and P S I it is evident that an important property

in determining slow PS II trapping is the shallow trap constituted by P680.

iv) Antenna size Another important parameter in determining the trapping rate is the antenna size. As mentioned above (section ii), in purely kinetic terms the antenna-trap migration rate is in part determined by the antenna size (N in Equation (1)). This parameter is also important in the case of the thermally equilibrated antenna-trap photosystem, as discussed above in section iii, where it was argued that the overall trapping rate (kcs) varies inversely with antenna size (N). The large antenna size is important in understanding the differences in trapping rate between purple bacteria, where N is smaller, and PS II (Trissl 1993). Thus, slow trapping in PS II can be attributed to a number of thermodynamic and kinetic parameters. It is therefore reasonable to conclude that the biological selection pressures to achieve this situation were considerable, particularly as slow trapping is at the expense of maximising the quantum efficiency of primary charge separation. In the case of the large antenna size (section iv) it is generally accepted that this is associated with the optical cross section for light absorption. Thus, the influence of this factor on the trapping rate is probably best understood in terms of the 'price to be paid' for ensuring a sufficiently high rate of light absorption. On the other hand, it is not so easy to see the biological advantage of the slow trapping rate associated with (i) the shallow antenna energy gradient (ii) the partially 'trap limited' exciton dynamics (iii) the shallow energy trap. In the following we will argue that this can be understood in terms of the development of excited state quenching mechanisms in PS II which seem to protect RCs against photoinhibition (Figure 2).

Significance of slow reaction centre trapping in PS H for photoinhibition Photoinhibition is usually defined as the process by which PS II photochemistry is inactivated under conditions of 'excess' illumination, i.e. when the rate of photon absorption exceeds the capacity of the system to dissipate excess energy by biologically innocuous mechanisms. Under these conditions RCs are predominantly in the closed (QA reduced) configuration (for recent review see Pr~isil et al. 1992). There is some agreement that the primary site of photoinhibition is

171

•

11

non photochemical "down regulation" of quenching in peripheral excited states by nonantenna photochemical quenching Figure 2. Relations between the various factors giving rise to slow RC trapping and 'down regulation' of excited states by non-photochemical

quenchingin PS II antenna. at the level of QA oxidation by the plastoquinone pool (Gong and Ohad 1995) and that subsequent to this RCs undergo photodamage, probably via triplet generation on P680 following P680/pheophytin charge recombination (MacPherson et al. 1993). P680 triplets are known to interact with oxygen and produce singlet oxygen (Telfer et al. 1994). Thus it is evident that any mechanism which minimises primary charge separation in the presence of reduced QA will protect PS II against photoinhibition. It is well known that powerful protective mechanisms against photoinhibition exist within PS II (Krause and Laasch 1987; Demmig-Adams and Adams 1992). These are based on the development of light induced quenching states which, by lowering the excited state levels, reduce or eliminate photoinhibitory damage of RCs. There has been considerable debate as to whether these quenching states are localised in the antenna complexes or in the RC itself. However, there is now considerable evidence that light induced antenna states are important sites of these processes. Thus, the xanthophyll cycle intermediates, associated with zeaxanthin quenching processes (Demmig-Adams and Adams 1992), are concentrated in the peripheral antenna complexes (Bassi et al. 1993). Recent experiments with leaves demonstrate that important nonphotochemical quenching states exist in both LHC II and the minor peripheral antenna complexes of PS II, while the core antenna complexes appear to be almost inactive (Hartel and Lokstein 1995). It is also known that what appear to be physiologically relevant quenching states can be induced in LHC II in vitro (Jennings et

al. 1991; Ruban et al. 1992) and in vivo (Jennings et al. 1992; Ruban and Horton 1994) by yet unknown molecular processes. Thus, the peripheral antenna complexes seem to be important sites for light-induced nonphotochemical quenching states. As pointed out above, trapping in PS II is several orders of magnitude slower than the individual energy transfer steps. It is therefore expected that excitation will be thermalised within the antenna-RC system before trapping occurs. This assumption is directly supported by the model calculations presented above which suggest that excitation flows in and out of P680 between 3 and 7 times prior to trapping. Thus antenna-P680 equilibration will occur. In such a situation the yield of non-photochemical quenching (CI~p) is approximated by ~h~e =

-1

Ttr

P" kNe + P " kNp

(2)

where PkNp is the overall quenching rate with P being the probability of an exciton residing in a quenching state and kNp the molecular quenching rate. 7-~is the RC trapping time (pheophyin reduction) in the absence of non-photochemical quenching states. Rearrangement of Equation 2 yields P " kNP = Ttr 1" (CI)N1 --

1) -1

(3)

Thus, for a constant yield, a slow trapping time implies a small PkNp and thus increases the sensitivity of PS II to down regulation by this mechanism. In the context of photoinhibition this factor is expected to be important when RCs are closed. Under these conditions of

172 reduced QA, slow trapping will increase the down regulation by non-photochemical quenching before primary charge separation occurs. The suggestion that the rate constant of pheophytin reduction (k2 in the above kinetic model) decreases upon QA reduction (Schatz et al. 1988; Dau and Sauer 1992) is extremely interesting in this context as this would further increase ru, thereby further increasing sensitivity to antenna based non-photochemical quenching prior to primary charge separation. In addition to contributing to slow trapping the shallow antenna funnel is important for regulatory quenching states in the peripheral antenna. As indicated above, to a first approximation the overall rate of a quenching process is given by PkNp (see Equation (3)). In the shallow antenna of PS II we estimate that the probability of an exciton residing in the peripheral antenna is P = 0.65 (Table 1). It is useful to compare this with the situation for a deep antenna funnel. In this case, if all the short wavelength forms (chlb, chla660, chla669) were concentrated in the peripheral antenna and all the long wavelength forms (chla677, chla684) in the core antenna the energy gradient would be about A = 23 kT at RT. This would lead to a markedly different distribution of excited states between peripheral and core antenna, with P = 0.05-0.12. It is therefore clear that the presence of a highly sensitive light-induced quenching mechanism in the peripheral antenna complexes of PS II depends directly on the shallow antenna properties.

References Bassi R, Pineau B, Dainese P and Marquardt J (1993) Carotenoidbinding proteins of Photosystem-lI. Eur J Biochem 212:297-303 Butler WL and Strasser RJ (1977) Tripartite model of the photochemical apparatus of green plant photosynthesis. Proc Natl Acad Sci USA 74:3382-3385 Dau H and Sauer K (1992) Electric field effect on the picosecond fluorescence of Photosystem-II and its relation to the energetics and kinetics of primary charge separation. Biochim Biophys Acta 1102:91-106 Demmig-Adams B and Adams WW (1992) Photoprotection and other responses of plants to high light stress. Ann Rev Plant Physiol Plant Mol Biol 43:599-626 Du M, Xie X, Mets L and Fleming GR (1994) Direct observation of ultrafast energy transfer processes in light harvesting complex II. J Phys Chem 98:4736-4741 French CS, Brown JS and Lawrence MC (1972) Four universal forms of chlorophyll a. Plant Physio149:421-429 Gong HS and Ohad I (1995) Rapid turnover of the RCII-D1 protein in the dark induced by photoinactivation of Photosystem II in Scenedesmus wild type and the PS II-donor defective It mutant cells. Biochim Biophys Acta 1228:181-188

Haehnel W, Nairn JA, Reisburg P and Saner K (1982) Picosecond fluorescence kinetics and energy transfer in chloroplasts and algae. Biochim Biophys Acta 680:161-173 Hartel H and Lokstein H (1995) Relationship between quenching of maximum and dark-level chlorophyll fluorescence in vivo dependence on Photosystem II antenna size. Biochim Biophys Acta 1228:91-94 Hedges M and Moya I (1986) Time resolved chlorophyll fluorescence studies of photosynthetic membranes: Resolution and characterisation of four kinetic components. Biochim Biophys Acta 849:193-202 Jennings RC, Garlaschi FM and Zucchelli G (1991) Light-induced fluorescence quenching in the light-harvesting chlorophyll a/b protein complex. Photosynth Res 27:57-64 Jennings RC, Zucchelli G, Garlaschi FM and Vianelli A (1992) A comparison of the light induced, non-reversible fluorescence quenching in Photosystem II with quenching due to open reaction centres in terms of the chlorophyll emission spectral forms. Biochim Biophys Acta 1101:79-83 Jennings RC, Bassi R. Garlaschi FM, Dalnese P and Zucchelli G (1993) Distribution of the chlorophyll spectral forms in the chlorophyll protein complexes of Photosystem-II antenna. Biochem 32:3203-3210 Jennings RC, Garlaschi FM, Finzi L and Zucchelli G (1994) Spectral heterogeneity and energy transfer in higher plant Photosystem II. Lith J Phys 34:293-300 Jennings RC, Bassi R and Zucchelli G (1996) Antenna structure and energy transfer in higher plants photosystems. In: Mattay J (ed) Topics in Current Chemistry: Electron Transfer pp 147-181. Springer-Verlag Heidelbert, Berlin Krause GH and Laasch H (1987) Photoinhibition of photosynthesis. Studies on mechanism of damage and protection in chloroplasts. In: Biggins J (ed) Progress in Photosynthesis Research, Vol 4, pp 19-26. Martinus NijhoffPublishers, Dordrecht/Boston/Lancaster Kudzmauskas S, Valkunas L and Borisov AY (1983) A theory of excitation transfer in photosynthetic units. J Theor Biol 105: 1323 Kiihlbrandt W and Wang DN (1991) Three-dimensional structure of plant light-harvesting complex determined by electron crystallography. Nature 350:130-134 Leibl W, Breton J, Deprez J and Trissl HW (1989) Photoelectric study on the kinetics of trapping and charge stabilisation in oriented PS II membranes. Photosynth Res 22:257-275 Macpherson AN, Telfer A, Barber J and Truscott TG (1993) Direct detection of singlet oxygen from isolated Photosystem-II reaction centres. Biochim Biophys Acta 1143:301-309 Pearlstein RM (1982) Chlorophyll singlet excitons. In: Govindjee (ed) Cell Biology: A Series of Monographs: Photosynthesis. Energy Conversion by Plants and Bacteria, Vol 1, pp 293-330. Academic Press, New York Pr~il O, Adir N and Ohad I (1992) Dynamics of Photosystem II: Mechanism of photoinhibition and recovery processes. In: Barber J (ed) Topics in Photosynthesis, 11: The Photosystems Structure, Function and Molecular Biology, pp 295-348. Elsevier Science Publishers, Amsterdam/London/New York/Tokyo Roelofs, TA, Lee C-H and Holzwarth AR (1992) Global target analysis of picosecond fluorescence fluorescence kinetics from pea chloroplasts. Biophys J 61, 1147-1163. Roelofs TA, Kwa SLS, van Grondelle R, Dekker JP and Holzwarth AR (1993) Primary processes and Structure of the PhotosystemII reaction center. 2. Low-temperature picosecond fluorescence kinetics of a D1/D2/cytochrome b-559 reaction center complex isolated by short triton exposure. Biochim Biophys Acta 1143: 147-157

173 Ruban AV and Horton P (1994) Spectroscopy of non-photocbemical and photochemical quenching of chlorophyll fluorescence in leaves - evidence for a role of the light harvesting complex of Photosystem II in the regulation of energy dissipation. Photosynth Res 40:181-190 Ruban AV, Ross D, Pascal AA and Horton P (1992) Mechanism of deltapH-dependent dissipation of absorbed excitation energy by photosynthetic membranes. 2. The relationship between LHC II aggregation in vitro and qE in isolated thylakoids. Biochim Biophys Acta 1102:39--44 Schatz GH, Brock H and Holzwarth AR (1988) Kinetic and energetic model for the primary processes in Photosystem 1I. Biophys J 54: 397-405 Seely GR (1973a) Effects of spectral variety and molecular orientation on energy trapping in the photosynthetic unit: A model calculation. J Theor Biol 40:173-187 Seely GR (1973b) Energy transfer in a model photosynthetic unit of green plants. J Theor Biol 40:189-199 Shipman LL and Housman DL (1979) F0rster transfer rates for chlorophyll a. Photochem Photobiol 29:1163-1167

Telfer A, Bishop SM, Phillips D and Barber J (1994) Isolated photosynthetic reaction center of Photosystem II as a sensitizer for the formation of singlet oxygen detection and quantum yield determination using a chemical trapping technique. J Biol Chem 269: 13244-13253 Trissl HW (1993) Long-wavelength absorbing antenna pigments and heterogeneous absorption bands concentrate excitons and increase absorption cross section. Photosynth Res 35:247-263 van Grondelle R, Dekker JP, Gillbro T and Sundstrom V (1994) Energy transfer and trapping in photosynthesis. Biochim Biophys Acta 1187:1-65 Wiederrecht GP, Seibert M, Govindjee and Wasielewski MR (1994) Femtosecond photodichroism studies of isolated Photosystem II reaction centers. Proc Natl Acad Sci USA 91:8999-9003 Zuccbelli G, Dainese P, Jennings RC, Breton J, Garlaschi FM and Bassi R (1994) Gaussian decomposition of absorption and linear dichroism spectra of outer antenna complexes of Photosystem II. Biochem 33:8982-8990