photosynthesis and photoprotection

Journal of Experimental Botany, Vol. 56, No. 411, Light Stress in Plants: Mechanisms and Interactions Special Issue, pp. 365–373, January 2005

doi:10.1093/jxb/eri023

Advance Access publication 22 November, 2004

Molecular design of the photosystem II light-harvesting antenna: photosynthesis and photoprotection Peter Horton* and Alexander Ruban Department of Molecular Biology and Biotechnology, University of Sheffield, Western Bank, Sheffield S10 2TN, UK Received 21 April 2004; Accepted 2 September 2004

Abstract The photosystem II (PSII) light-harvesting system carries out two essential functions, the efficient collection of light energy for photosynthesis, and the regulated dissipation of excitation energy in excess of that which can be used. This dual function requires structural and functional flexibility, in which light-harvesting proteins respond to an external signal, the thylakoid DpH, to induce feedback control. This process, referred to as non-photochemical quenching (NPQ) depends upon the xanthophyll cycle and the PsbS protein. In nature, NPQ is heterogeneous in terms of kinetics and capacity, and this adapts photosynthetic systems to the specific dynamic features of the light environment. The molecular features of the thylakoid membrane which may enable this flexibility and plasticity are discussed.

Key words: Light-harvesting complex, non-photochemical quenching, photoprotection, thylakoid membrane, xanthophyll cycle.

Introduction During the evolution of oxygenic photosynthesis, there have been two conflicting demands placed upon the molecular machinery of photosystem II (PSII). On the one hand efficient collection of sunlight is needed to deliver excitation energy to the photosynthetic reaction centre at a rate sufficient to drive electron transport at the highest sustainable rate. On the other hand, excited states of chlorophyll and the presence of molecular oxygen provide a potentially lethal cocktail that can irreversibly damage the proteins, lipids, and pigments of the photosynthetic membrane. To combat this, constitutive features of the photosystem, such as the presence of carotenoids to scavenge

singlet oxygen and trap triplet states of chlorophyll, are combined with other photoprotective mechanisms which are induced under high levels of illumination to bring about the dissipation of excess chlorophyll excited singlet states.

In this paper those features of molecular design that promote efficient light harvesting will be reviewed, and then it will be shown how the dynamic nature of this system allows for an inducible enhancement of the rate of dissipation of excited states. It will be shown how these features of design differ in different classes of organism, and these differences will be rationalized in terms of evolution within particular ecological niches.

The molecular principles of light harvesting The chlorophyll a and b molecules, bound to the pigment– protein complexes of the light-harvesting system, are arranged in three-dimensional space to serve as an antenna; gathering light over a large absorption cross-section, and ‘funnelling’ the absorbed energy to the reaction centres.

Energy transfer between pigment molecules, both between and within individual complexes is highly efficient. Equilibration of energy within the antenna is rapid, and tends towards the red-shifted lower energy excited states of the reaction centre core complex (van Amerongen and van Grondelle, 2001). This highly efficient antenna depends entirely on the specific structures of the apo-proteins of the light-harvesting complexes: they must bind sufficient pigments to maximize absorption; they must provide for the optimal orientation and configuration of pigments for energy transfer; they must tune the excited state energy levels of the pigments to promote energy transfer; and they must permit protein–protein contacts to allow appropriate routes of energy transfer between complexes.

There are important constraints placed upon this design.

Most notable is that the chlorophyll molecules must be

* To whom correspondence should be addressed. Fax: +44 (0)114 222 2787. E-mail: p.horton@sheffield.ac.uk Journal of Experimental Botany, Vol. 56, No. 411, ª Society for Experimental Biology 2004; all rights reserved

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

366 Horton and Ruban

arranged so as to prevent the tendency for excited states to dissipate by non-radiative decay. In the main lightharvesting complex, LHCII, the chlorophyll concentration is approximately 0.6 M: if chlorophyll was dissolved in an organic solvent at this concentration all absorbed energy would be dissipated in this way, a process described as concentration quenching (Beddard and Porter, 1976). This occurs via close chlorophyll–chlorophyll interactions, dimers or excimers. Clearly the apoprotein provides a specific ‘solvent’, an environment that allows a high chlorophyll concentration without the probability of quenching by such associations. This environment arises from the formation of hydrophobic binding pockets, liganding of the chlorophyll Mg atoms by amino acid residues of the alpha helices and hydrogen bonding to the porphyrin ring (Liu et al., 2004). It is useful to regard chlorophylls as the ‘cofactors’ of the lightharvesting ‘enzymes’, and many parallels exist between these and other proteins with essential cofactors, such as haem proteins and flavoproteins.

PSII macro-organization The basic structural unit of PSII is the so-called LHCII–PSII supercomplex (Hankamer et al., 1997). This is a macromolecular dimer, consisting of the D1/D2/CP43/CP47 PSII core complex and several light-harvesting proteins, one copy of the minor monomeric complexes CP26 and CP29, and one of the major trimeric complex, LHCII. In the granal membrane of the chloroplast, the supercomplex is associated with further LHCII trimers and the monomeric CP24, to form ‘megacomplexes’, which frequently display a semicrystalline order (Boekema et al., 2000).

The subunits of the PSII antenna are therefore associated together with increasing levels of complexity. For example, LHCII monomers form trimers, which are themselves associated, on the one hand, with the monomeric complexes and core complexes and on the other, with more LHCII trimers, to form a 2-D organization in the thylakoid membrane. Finally, the stromal-facing surfaces of membranes are appressed together to form the grana, with contacts between LHCII and between LHCII and PSII cores of the adjacent membranes. This three-dimensional macrostructure has been referred to as the macrodomain (Garab and Mustardy, 1999). The tendency for oligomerization is displayed by LHCII, which can in vitro self-assemble into 2-D and 3-D structures that have a strong resemblance to thylakoid membranes (Burke et al., 1987). Equally, oligomers of LHCII can be prepared following mild detergent treatment of thylakoid membranes (Ruban et al., 1999). The trimeric state of LHCII appears to be essential for this organization of PSII. When the main proteins that comprise LHCII are removed by expression of antisense genes, it was found that the amount of the normally monomeric CP26 dramatically increased in level and it became trimeric (Ruban et al., 2003). This change in expression of photosynthetic genes

must be a response to the LHCII deficiency, as the plant attempts to compensate for the loss of an essential protein.

Amazingly, this response allows the wild-type oligomeric PSII structure and the granal membrane to be maintained, emphasizing the importance and robustness of the thylakoid macrostructure.

Molecular accessories for regulating light harvesting In addition to the main protein complexes, both the PSII core and the light-harvesting system contain a number of other proteins, pigments, and lipids (Hankamer et al., 1997).

Whereas some of these may have essential roles in the structure and function of the light-harvesting systems (e.g. a specific phospholipid is required for LHCII trimerization; Remy et al., 1982) and DGDG may enable association between trimers (Liu et al., 2004), others appear to have another role, that of tuning the properties of light-harvesting proteins for regulation. Specific lipids appear to be necessary for certain dynamic events within the 3-D organization of LHCII (Simidjiev et al., 2000). Peripherally associated with LHCII are the carotenoids of the xanthophyll cycle (Ruban et al., 1999), which plays a vital role in photoprotective energy dissipation (Demmig-Adams, 1990), observed as the non-photochemical quenching of chlorophyll fluorescence (NPQ). This cycle is the reversible de-epoxidation of violaxanthin (found under light-limiting conditions) into zeaxanthin (found under light-saturating conditions) that is correlated with the induction of NPQ. Violaxanthin deepoxidase catalyses this reaction, and this enzyme may have a specific docking site on LHCII (Liu et al., 2004). A group of LHC-related proteins encoded by the members of the lhc gene super family (Jansson, 1999), including the ELIPS, the Lil proteins, and PsbS are also involved in stress responses (Heddad and Adamska, 2002), with PsbS having a key role in NPQ. Protein kinases also interact with LHCII (Bennett, 1983) and CP29 (Croce et al., 1996), phosphorylating them.

In the case of LHCII it is well-known that the redox regulation of kinase activity provides the basis for regulating the distribution of absorbed energy between PSII and PSI (Horton and Black, 1980), phosphorylation modulating the structure of the complex, and its interaction with other photosystem proteins (Allen and Forsberg, 2001). Phosphorylation of CP29 appears to be activated under cold stress, causing a structural change in the protein, which may have a photoprotective role (Croce et al., 1996).

Changes in structure as the basis for regulation of light harvesting Whereas the higher order structure of PSII was first thought to be important only in increasing the efficiency of light harvesting, more recently it has been suggested that it provides the essential dynamic properties involved in

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

Molecular design of the PSII light-harvesting antenna

regulation (Horton, 1999). When isolated light-harvesting complexes self-assemble into oligomers in vitro, there is a decrease in fluorescence yield (Mullet and Arntzen, 1980; Ruban et al., 1992), resulting from a decrease in excitation lifetime by non-radiative decay. Since, in vivo, the complexes also exist in an oligomeric state (Dekker et al., 1999; Ruban et al., 1999), it could be proposed that increased energy dissipation would result. Indeed, it has been recognized that the maximum fluorescence lifetime of chlorophyll in vivo is always less than free chlorophyll or chlorophyll in unaggregated LHCII (Horton and Ruban, 1994). Perhaps the advantages of oligomeric organization outweigh the disadvantage of some waste of absorbed excitation energy.

More interesting, however, is the notion that structural changes within the oligomeric antenna could provide a physiological mechanism for regulating the partitioning of energy between utilization in photosynthesis and dissipation by NPQ (Horton et al., 1991, 1996). Indeed, there are many remarkable similarities between NPQ and in vitro fluorescence quenching in LHCII and other antenna complexes (Horton et al., 1999). Whilst the exact mechanism of energy dissipation in these complexes remains to be determined, it almost certainly is equivalent to the process of ‘concentration’ quenching discussed above. Hence, given that the precise 3-D structure of the protein prevents such quenching, it is easy to imagine how (relatively subtle) changes in the structure could induce quenching. Examination of the structural model of LHCII reveals the presence of several candidate quenching pairs of chlorophyll molecules (Kuhlbrandt et al., 1994; Liu et al., 2004). Although the pair of Chl a 613/614 has been put forward as a potential quenching site (Liu et al., 2004), it has previously been suggested that the Chl a 610/611/612/lut1 domain was the site of quenching (Wentworth et al., 2003). It was shown how the differing tendencies of trimeric and monomeric LHCII to quench could be explained if the conformation of quenching locus was controlled by the Lut 2 domain (Wentworth et al., 2003). This idea was consistent with earlier observations of altered fluorescence yield of complexes when the Lut 2 site binds different xanthophylls (Moya et al., 2001; Morosinotto et al., 2002; Formaggio et al., 2001; Caffarri et al., 2004).

As a guiding principle, it was suggested that NPQ was a response of the structure of the PSII light-harvesting antenna, resulting in altered pigment interactions within protein subunits and caused by the change in external environment and the binding of protons (because of the increase in transthylakoid DpH) (Horton et al., 1991; Horton and Ruban, 1992). Of course, the notion that protein structure and function can be altered by pH change is well established and, extrapolating further, it is suggested that the oligomeric nature of the light-harvesting system may allow co-operativity in proton binding and, most importantly, allosteric regulation by other ligands (Ruban and Horton, 1995a; Ruban et al., 2001; Horton et al., 2000).

367

A conceptual model was formulated, comprising four conformational states of the light-harvesting system, with different fluorescence yields, differing in de-epoxidation and protonation state (Horton et al., 1991). The action of zeaxanthin was explained, it is an allosteric activator of quenching binding to a peripheral site on LHCII or another light-harvesting antenna protein. There is considerable experimental support for this hypothesis, although direct participation of zeaxanthin in the quenching mechanism cannot be excluded (Frank et al., 1994; Ma et al., 2003).

Paramount in this evidence is the startlingly different kinetic properties of qE observed in thylakoids isolated from leaves pretreated to have different de-epoxidation states (Ruban et al., 2001): qE forms more rapidly, at a lower DpH and with reduced co-operativity in the presence of zeaxanthin, all characteristics of it being an allosteric activator. Such properties also explain the kinetics of NPQ formation upon illumination of dark-adapted leaves, and the accelerated rate of formation upon a second illumination following a brief dark interval in which DpH, but not de-epoxidation is relaxed (Ruban and Horton, 1999). Recently it has been suggested that, rather than being a peripheral site on an antenna protein, the internal Lut 2 site discussed above provides the allosteric binding site for zeaxanthin (Formaggio et al., 2001). Indeed it has been shown that when the Lut 2 site in CP26 is occupied by zeaxanthin, there is a decrease in fluorescence lifetime, compared with when violaxanthin is bound (Moya et al., 2001; Crimi et al., 2001). In this model, reversible exchange of carotenoid binding to this site is proposed (Bassi and Caffarri, 2000), and in vitro this has been shown to be promoted at low pH (Morosinotto et al., 2002).

It is interesting, and rather unexpected, that there is no evidence for any one light-harvesting complex having an obligatory role in NPQ, at least in higher plants; wild-type levels of NPQ are found in Arabidopsis plants expressing antisense genes for CP29, CP26, and LHCII (Andersson et al., 2001, 2003). Similarly, chl b less mutants, which have only low levels of all Lhcb proteins show NPQ (Lokstein et al., 1993). This indicates that no single Lhcb protein can be the unique site of either energy dissipation or zeaxanthin activation. Furthermore, it suggests that NPQ could be ‘shared’ by all antenna proteins, including those of the PSII core, since the above principle of control of concentration quenching can apply to them all. It remains to be shown what the minimum structural unit for NPQ is.

In vitro studies also show how each antenna subunit has only a small capacity for zeaxanthin-induced quenching (Wentworth et al., 2000; Crimi et al., 2001), but that this is amplified when subunit interactions are enabled (Ruban and Horton, 1992; Moya et al., 2001). Indeed, this suggests that the intra- and inter-subunit conformational changes act in a concerted manner, for example, a quenching locus in LHCII or CP26 could be controlled by the Lut 2 domain, which is in turn affected by the xanthophyll bound there,

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

368 Horton and Ruban

the binding of xanthophylls at the more peripheral neoxanthin (N1) and violaxanthin (V1) sites, and interactions with neighbouring protein subunits.

The role of PsbS and zeaxanthin activation Arabidopsis plants with mutation in the PsbS gene (npq4 mutants) show none of the rapidly reversible type of NPQ known at qE, despite having wild-type levels of zeaxanthin and DpH, and unaltered levels of light-harvesting and reaction centre complexes (Li et al., 2000). It was suggested that PsbS provides the site of NPQ, by binding both protons and zeaxanthin, and interacting with a chlorophyllcontaining complex to introduce a new quenching pathway.

This hypothesis for the role of PsbS has been subject to intensive investigation. Evidence for proton binding has come from the identification of binding sites for DCCD (Dominici et al., 2002), and from the inhibition of qE by site-directed mutagenesis of potential proton binding residues (Li et al., 2002a). The pigment-binding properties of PsbS are unclear; purified PsbS appears to contain no chlorophyll or carotenoid.

Finding the location of zeaxanthin binding sites has been an important goal in understanding NPQ. It has been shown that isolated LHCII and the minor complexes CP26 and CP29 contain violaxanthin (Ruban et al., 1994, 1999; Bassi and Caffarri, 2000). Only some of this is available for de-epoxidation, shown to include that loosely bound to the complex (Ruban et al., 1999; Caffarri et al., 2001), and there is evidence that violaxanthin bound to the Lut 2 site on CP26 and CP24 could exchange for zeaxanthin (Morosinotto et al., 2002). However, such studies of zeaxanthin binding sites could be misleading, since it is likely that only a small proportion of the zeaxanthin pool is involved in NPQ (Gilmore, 2001). This subpopulation of zeaxanthin has been studied using a spectroscopic approach. It is well known that qE is very strongly correlated with an absorption change at 535 nm (Bilger et al., 1988; Ruban et al., 1993), whilst it was considered to arise from selective light scattering, its proximity to the carotenoid absorption bands, suggested it may be electronic in origin.

Indeed, the DA535 spectrum could be modelled as a red shift in 1–2 carotenoid molecules (Ruban et al., 2002).

Resonance Raman spectroscopy has been used to identify the absorption bands from each of the xanthophylls (Ruban et al., 2000). In leaves, induction of qE was associated with the appearance of a zeaxanthin species selectively excited at 528 nm. This signal was absent in the npq4 mutant. These data not only proved the electronic nature of DA535, but also showed that it most likely arose from a small population of red-shifted zeaxanthin molecules (Ruban et al., 2002). It is suggested that this was indicating the activation of zeaxanthin, required to fulfil its role in qE. The Raman signal also indicated that this species was found in a rather specific configuration, indicating binding to a protein com-

ponent of the membrane. In order to test whether this site could be on PsbS, zeaxanthin was reconstituted with a purified sample of this protein; a red-shifted absorption band was created with an identical Raman signature to DA535 (Aspinall-O’Dea et al., 2002).

Although specific binding to another protein site is not excluded, these data suggest that the activator role of zeaxanthin in qE requires that it be bound to PsbS. This binding, in vivo, requires DpH, most likely because of proton binding to PsbS, and confers upon zeaxanthin very new configurational and excited state features. How these properties are related to its mechanism in qE is currently unknown. On the one hand, these features may be needed for zeaxanthin to quench chlorophyll excited states on a neighbouring antenna complex directly (Ma et al., 2003).

On the other hand, protonated PsbS may ‘deliver’ zeaxanthin to its active site in the antenna: in this case it again may be a direct quencher, or as suggested above, it may fulfil an allosteric activating role. Finally, a protonated zea/ PsbS complex may have an allosteric role. In the latter two cases, PsbS would be regarded as the regulatory subunit of the light-harvesting antenna. Further work is needed to distinguish these possibilities.

Comparative biochemistry and physiology of NPQ NPQ is observed in all higher plants; it has been detected in a wide range of species in many different plant types, in a wide range of habitats (Demmig-Adams, 1998; Johnson et al., 1993). It has also been found in lower plants (Campbell et al., 1998; El Bissati et al., 2000), green algae (Niyogi et al., 1997), and diatoms (Casper-Lindley and Bjo¨rkman, 1998; Arsalane et al., 1994). It seems reasonable to assume that the underlying molecular mechanisms are the same in all cases. Nevertheless a comparative study of NPQ in different species has proved to be rewarding, different aspects of NPQ appear to be enhanced in certain species, making them good models for further study.

Moreover, in such cases, the knowledge obtained, can give new approaches to gaining fundamental information on the mechanism of NPQ. NPQ has been described as being the sum of two different processes (Horton et al., 1996). In addition to the rapidly relaxing qE type discussed above, there is further NPQ that relaxes very slowly when excess light is removed. This latter type, called qI, is usually much smaller than qE, and is attributed to a more sustained effect of excess light. Therefore qI has been associated with ‘photoinhibition’ rather than ‘regulation’, and was thought to arise from quenching by photodamaged PSII reaction centres.

However, it was suggested that qI, like qE, may not be an effect of excess light stress, but an adaptive response to it (Horton et al., 1988, 1996; Ruban et al., 1993). Thus, perhaps quenching by a photoinhibited reaction centre is

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

Molecular design of the PSII light-harvesting antenna

a useful response that provides photoprotection of the antenna and thylakoid membrane (Chow et al., 2002).

Since strong evidence of the accumulation of damaged PSII under photoinhibitory conditions was absent, it was further suggested that qI occurs in the antenna, perhaps even by the same or similar mechanism to qE (Demmig and Winter, 1988; Ruban et al., 1993; Ruban and Horton, 1995b).

In order to determine the relationship between qE and qI, species that show particular NPQ characteristics have been sought. The epiphytic plant, Guzmania monostachia, exhibits a large NPQ capacity (Ruban et al., 1993). Both qE and qI are induced in excess light: successive periods of illumination progressively ‘pump’ qE to its maximum value, and this is activation is correlated with an increase in qI. Thus, qI is the signature of the light ‘activation’ described above, and is therefore attributed to the de-epoxidation of violaxanthin. The large extent of activation and qI in this species is also associated with a large xanthophyll cycle pool and a high de-epoxidation state. Spectroscopy carried out on Guzmania leaves upon the induction of qI showed the formation of red-shifted chlorophyll emission previously identified in aggregated LHCII. A very stable type of NPQ found in plants acclimated to low temperature also has this feature, the ‘cold hard band’, suggesting that evergreen plants which can withstand freezing conditions over winter exist in a permanently photoprotected qI state of aggregated LHCII (Gilmore and Ball, 2000). Biochemical data on this state suggests that factors such as LHCII phosphorylation ¨ quist and (Ebbert et al., 2001) and accumulation of PsbS (O Huner, 2003) may play a role in establishing and/or stabilizing this state of the PSII antenna.

NPQ found in some algae has different features compared with higher plants. In Chlamydomonas, NPQ is rather weak (about 50% of a typical value in higher plants), slow to form, and partially dependent upon zeaxanthin formation (Elrad et al., 2002). Various NPQ-deficient mutants have been isolated, including one with a mutation in a light-harvesting gene, not in PsbS (Elrad and Grossman, 2004). In many respects NPQ resembles the higher plant qI, rather than qE. In diatoms, NPQ is similar to that found in plants, except that it is large (sometimes about five times bigger) and it is very slow to relax; it forms like qE but relaxes like qI (Lavaud et al., 2002). Diatoms have a xanthophyll cycle, but it is different from that found in plants. It is a one-step deepoxidation of diadinoxanthin to diatoxanthin, equivalent to the second step of the plant xanthophyll cycle, antheraxanthin–zeaxanthin conversion. The light-harvesting complexes in diatoms are different from those found in higher plants; they bind different carotenoids and chlorophylls, and have very low sequence similarity. They are predicted 3-helix proteins. The xanthophyll cycle carotenoids account for 20–40% of the carotenoid pool, and it is estimated that each LHC binds 2–4 molecules of diadinoxanthin. PsbS is absent from these organisms, although it is interesting that some similar PsbS protein sequences are found in their LHC

369

proteins. The DA535 is absent, although a light-induced absorbance change around 522 nm is observed, and may be indicative of some diatoxanthin activation. The formation of NPQ is DpH-dependent and this appears to be not only due to stimulation of de-epoxidation at low lumen pH; collapse of the DpH by uncouplers reverses NPQ even though the level of diatoxanthin in unchanged. Interestingly, when NPQ has been completely formed, it is no longer uncoupler sensitive.

Previously an attempt was made to explain both qE and qI in terms of a 4-state model of the light-harvesting system (Horton and Ruban 1992); qI was considered to be a state in which de-epoxidation has taken place, but not yet protonated, whereas qE is always a protonated state that would normally also bind zeaxanthin. The pKa of the epoxidized and de-epoxidized states, the strength and number of xanthophyll cycle binding sites, the xanthophyll cycle pool size, the level of PsbS and the pH of the thylakoid lumen would determine the amounts of qE and qI observed in different plants. Extending this rationale to the green algae, perhaps the DpH is not large enough, and the zeaxanthin concentration not high enough to cause protonation of the qE active sites in the absence of PsbS. In diatoms, the large NPQ can be explained by either a stronger quencher, or by a larger number of quenching sites. There is currently insufficient data to distinguish these.

The kinetic features indicate a qI-type process, but one which is nevertheless dependent upon protonation, and formed rapidly. The fact that formation of diatoxanthin only requires one de-epoxidation, could explain the rapid activation of quenching. Since, PsbS is absent from diatoms, NPQ must occur by direct effects on the light-harvesting complexes.

The slow relaxation could arise from extreme stability of the protonated state of these complexes when diatoxanthin is present, so that reversal of quenching absolutely depends upon the rather slow kinetics of de-epoxidation. This stability may arise from particularly strong protein interactions within this antenna that ‘locks’ the quenched state. This blurring of the distinction between qE and qI leads to a new suggestion about the role of the PsbS protein in plants: perhaps it provides control over NPQ, enabling not only induction of NPQ, but also its rapid reversibility.

Comparative ecology of NPQ The variation in character of NPQ in different organisms suggests that facets of this process have undergone evolutionary adaptation to particular ecological niches. In land plants, a major requirement for the regulation of light harvesting is a rapid, dynamic response to changes in light levels. The activation by de-epoxidation, and triggering by protonation, is adapted to the more rapid changes associated with leaf movements and sunflecks, against a background of diurnal and daily changes. The relatively slow kinetics of de-epoxidation act as a molecular memory of ‘average’ light conditions, whereas protonation provides an

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

370 Horton and Ruban

‘on/off switch’ with a response time of just a few seconds.

This switch, apparently, requires a transition between two states of the light-harvesting system with different excitation lifetimes, setting the upper limit on the magnitude of NPQ. The amount of quenching depends upon the number of units in each state. Plants acclimated, and adapted to particular light environments, tune the dynamics of NPQ. Plants exposed consistently to large excesses of radiation are found to have the high NPQ capacity, driven by a high xanthophyll cycle pool and PsbS concentration, a low content of LHCII and perhaps an increased capacity of DpH generation (by cyclic electron transfer). In overwintering plants, these features appear to be accompanied by other mechanisms, which stabilize the quenched state even in the absence of DpH. Diatoms are able to achieve a very high NPQ capacity.

This appears to be an adaptation to environment, with large but relatively slow fluctuations in incident radiation, arising from ocean currents. The requirement for a rapid off switch is quite low in this environment. It seems likely in any case that it is not possible to have, at the same time, a very large NPQ, and to have it rapidly reversible.

Evolution of NPQ The diversity of NPQ characteristics suggests a strong evolutionary pressure upon the attainment of optimized photoprotection. This evolutionary process can be viewed within the context of the LHC gene super family (Green and Pichersky, 1994; Jansson, 1999). The cyanobacterial HLIP and plant Lil 2 genes encode single-helix proteins that resemble the progenitors of the Lhc proteins. The Lil3 gene then arose by duplication, to give a 2-helix protein. PsbS is then considered as a further duplication, since it is a 4-helix protein. ELIPS, which are 3-helix proteins, would then have evolved from a PsbS ancestor by deletion of a C-helix region. All of these proteins are associated with stress tolerance (Heddad and Adamska, 2002), and probably do not bind chlorophylls. Instead, they may have evolved as carotenoid protein complexes. For instance, the ELIP protein found in abundance in some algae, Cbr, may bind carotenoids (Levy et al., 1993). Accumulation of plant ELIPS have also been reported to correlate closely with the amount of xanthophyll pigments (Montane et al., 1998).

Light-harvesting proteins can then be considered to have evolved from these stress proteins, by binding chlorophylls, and excluding many carotenoids (Fig. 1). The trimeric state of plant LHCII is probably the most evolved complex for light harvesting (Wentworth et al., 2003). It is interesting to consider the LHCF protein of diatoms; this complex binds chlorophyll, but has retained a high amount of carotenoid binding. The level of this protein increases under conditions of high NPQ, and hence may be considered as truly bifunctional—operating equally in both photoprotection and light harvesting, its two modes depending upon deepoxidation. By contrast, in plants, photoprotection requires PsbS, a protein not involved in light harvesting. However, this protein is able to confer upon LHCII (or any PSII antenna complex) a photoprotective function, perhaps by providing delivery and removal of the ‘active’ zeaxanthin.

The presence of PsbS, therefore, allows the light-harvesting complex to function as an efficient light harvester, with a high absorption cross-section, its chlorophyll pigments not needing to be diluted by binding high levels of carotenoid.

The extra benefit from this mode of control is rapidity as reversal does not depend upon epoxidation of zeaxanthin.

Moreover, once the appropriate de-epoxidation state has been reached, NPQ can be turned on and off rapidly. In this context it is interesting to consider the effect of PsbS concentration on NPQ. Manipulation of the amount of PsbS in Arabidopsis shows that its stoichiometry correlates with changes in the amplitude of the qE component of NPQ (Li et al., 2002b). Mutants which lack PsbS altogether have only a slowly relaxing NPQ; in the absence of PsbS the unassisted process is revealed, one which relies upon the passive mass action effect of the xanthophyll cycle and in which qE and qI are indistinguishable. The evolution of qE in plants, driven by the need to respond rapidly to changing light intensity, resulted from the specification of an ancestral carotenoid– protein complex to a new role reversibly binding xanthophyll cycle carotenoids and protons, and interacting with the PSII light-harvesting antenna proteins.

Conclusions The principles that have guided the molecular design of the PSII light harvesting have been elucidated. Modifications to this design are found in different organisms’ adaptations to

Fig. 1. Evolutionary and functional relationships between proteins involved in light harvesting and photoprotection. LHCII is seen as the end-product of a process that maximizes light harvesting through high amounts of chlorophyll binding. These may have evolved from stress-responsive single-helix proteins such as HLIP and ELIPS, which bind only carotenoids. PsbS provides an essential function in plant photoprotection, with only minimal pigment binding. On the other hand, LHCF binds both chlorophyll and carotenoids in high amounts, functioning in both light harvesting and photoprotection.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

Molecular design of the PSII light-harvesting antenna

specific environments. An essential feature is the structural flexibility of the light-harvesting proteins, which allow control over the transfer of excitation energy to particular quenching species, associates of chlorophyll or chlorophyll and xanthophyll. The recent elucidation of the threedimensional structure of LHCII has revealed the possible molecular mechanisms by which this may occur (Liu et al., 2004). High resolution electron microscopy and image analysis of wild-type and genetically altered plants are showing how these structures are organized within the intact membrane (Yakushevska et al., 2003; Ruban et al., 2003) and new spectroscopic approaches to the investigation of intact systems are beginning to reveal details of how such mechanisms operate (Robert et al., 2004).

References Allen JF, Forsberg J. 2001. Molecular recognition in thylakoid structure and function. Trends in Plant Science 6, 317–326.

Andersson J, Ruban AV, Walters R, Howard C, Wentworth M, Horton P, Jansson S. 2003. Absence of Lhcb1 and Lhcb2 proteins of the light-harvesting complex of photosystem II: effects on photosynthesis, grana stacking and fitness. The Plant Journal 35, 350–361.

Andersson J, Walters RG, Horton P, Jansson S. 2001. Antisense inhibition of the photosynthetic antenna proteins CP29 and CP26: implications for the mechanism of protective energy dissipation.

The Plant Cell 13, 11930–1204.

Arsalane W, Rousseau B, Duval JC. 1994 Influence of the pool size of the xanthophyll cycle on the effects of light stress in a diatom: competition between photoprotection and photoinhibition. Photochemistry and Photobiology 60, 237–243.

Aspinall-O’Dea M, Wentworth M, Pascal A, Robert B, Ruban AV, Horton P. 2002. The PsbS subunit of photosystem II binds zeaxanthin and activates it for non-photochemical fluorescence quenching. Proceedings of the National Academy of Sciences, USA 99, 16331–16335.

Bassi R, Caffarri S. 2000. Lhc proteins and the regulation of photosynthetic light harvesting function by xanthophylls. Photosynthesis Research 64, 243–256.

Beddard GS, Porter G. 1976. Concentration quenching in chlorophyll. Nature 260, 366–367.

Bennett J. 1983. Regulation of photosynthesis by reversible phosphorylation of light-harvesting a/b protein. Biochemical Journal 212, 1–13.

Bilger W, Heber U, Schreiber U. 1988. Kinetic relationship between energy-dependent fluorescence quenching, light-scattering, chlorophyll luminescence and proton pumping in intact leaves.

Zeitschrift fu¨r Naturforschung 43, 877–887.

Boekema EJ, van Breemen JFL, van Roon H, Dekker JP. 2000.

Arrangement of photosystem II supercomplexes in crystalline macrodomains within the thylakoid membrane of green plant chloroplasts. Journal of Molecular Biology 301, 1123–1133.

Burke JJ, Ditto CL, Arntzen CJ. 1987. Involvement of the lightharvesting complex in cation regulation of excitation energy distribution in chloroplasts. Archives of Biochemistry and Biophysics 187, 252–263.

Caffarri S, Croce R, Breton J, Bassi R. 2001. The major antenna complex of photosystem II has a xanthophyll binding site not involved in light harvesting. Journal of Biological Chemistry 276, 35924–35933.

Caffarri S, Croce R, Cattivelli L, Bassi R. 2004. A look within LHCII: differential analysis of the Lhcb1–3 complexes building

371

the major trimeric antenna complex of higher plant photosynthesis.

Biochemistry 43, 9467–9476.

Campbell D, Hurry V, Clarke AK, Gustafsson P, Oquist G. 1998.

Chlorophyll fluorescence analysis of cyanobacterial photosynthesis and acclimation. Microbiology and Molecular Biology Reviews 62, 667–677.

Casper-Lindley C, Bjorkman O. 1998. Fluorescence quenching in four unicellular algae with different light-harvesting and xanthophyll cycle pigments. Photosynthesis Research 56, 277–289.

Chow WS, Lee HY, Park YI, Park YM, Hong YN, Anderson JM. 2002. The role of inactive photosystem-II-mediated quenching in a last-ditch community defence against high light stress in vivo.

Philosophical Transactions of the Royal Society of London, Series B 357, 1441–1449.

Crimi M, Dorra D, Bosinger CS, Giuffra E, Holzwarth AR, Bassi R. 2001. Time-resolved fluorescence analysis of the recombinant photosystem II antenna complex CP29: effects of zeaxanthin, pH and phosphorylation. European Journal of Biochemistry 268, 260–267.

Croce R, Breton J, Bassi R. 1996. Conformational changes induced by phosphorylation in the CP29 subunit of photosystem II. Biochemistry 35, 11142–11148.

Dekker JP, van Roon H, Boekema EJ. 1999. Heptameric association of light-harvesting complex II trimers in partially sollubilized photosystem II membranes. FEBS Letters 449, 211–214.

Demmig-Adams B. 1990. Carotenoids and photoprotection: a role for the xanthophyll zeaxanthin. Biochimica et Biophysica Acta 1020, 1–24.

Demmig-Adams B. 1998. Survey of thermal energy dissipation and pigment composition in sun and shade leaves. Plant and Cell Physiology 39, 474–482.

Demmig B, Winter K. 1988. Characterization of 3 components of nonphotochemical fluorescence quenching and their response to photoinhibition. Australian Journal of Plant Physiology 15, 163–177.

Dominici P, Caffarri S, Armenante F, Ceoldo S, Crimi M, Bassi R. 2001. Biochemical properties of the PsbS subunit of photosystem II either purified from chloroplast or recombinant. Journal of Biological Chemistry 277, 22750–22758.

Ebbert V, Demmig-Adams B, Adams WW, Mueh KE, Staehelin LA. 2001. Correlation between persistent forms of zeaxanthindependent energy dissipation and thylakoid protein phosphorylation. Photosynthesis Research 67, 63–78.

El Bissati K, Delphin E, Murata N, Etienne AL, Kirilovsky D. 2000. Photosystem II fluorescence quenching in the cyanobacterium Synechocystis PCC 6803: involvement of two different mechanisms. Biochimica et Biophysica Acta 1457, 229–242.

Elrad D, Grossman AR. 2004. A genome’s-eye view of the lightharvesting polypeptides of Chlamydomonas reinhardtii. Current Genetics 45, 61–75.

Elrad D, Niyogi KK, Grossman AR. 2002. A major light-harvesting polypeptide of photosystem II functions in thermal dissipation. The Plant Cell 14, 1801–1816.

Formaggio E, Cinque G, Bassi R. 2001 Functional architecture of the major light-harvesting complex from higher plants. Journal of Molecular Biology 314, 1157–1166.

Frank HA, Cua A, Chynwat V, Young A, Gosztola D, Wasielewski MR. 1994. Photophysics of the carotenoids associated with the xanthophyll cycle in photosynthesis. Photosynthesis Research 41, 389–395.

Garab G, Mustardy L. 1999. Role of LHCII-containing macrodomains in the structure, function and dynamics of grana.

Australian Journal of Plant Physiology 26, 649–658.

Gilmore AM. 2001. Xanthophyll cycle-dependent nonphotochemical quenching in photosystem II: mechanistic insights gained from Arabidopsis thaliana L. mutants that lack violaxanthin deepoxidase activity and/or lutein. Photosynthesis Research 67, 89–101.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

372 Horton and Ruban Gilmore AM, Ball MC. 2000. Protection and storage of chlorophyll in overwintering evergreens. Proceedings of the National Academy of Sciences, USA 97, 11098–11101.

Green BR, Pichersky E. 1994. Hypothesis for the evolution of 3-helix chl a/b and chl a/c light-harvesting antenna proteins from 2-helix and 4-helix ancestors. Photosynthesis Research 32, 149–162.

Hankamer B, Barber J, Boekema EJ. 1997. Structure and membrane organization of photosystem II in green plants. Annual Reviews in Plant Physiology and Plant Molecular Biology 48, 641–671 Heddad M, Adamska I. 2002. The evolution of light stress proteins in photosynthetic organisms. Comparative and Functional Genomics 3, 504–510.

Horton P. 1999. Are grana necessary for regulation of light harvesting? Australian Journal of Plant Physiology 26, 659–669.

Horton P, Black MT. 1980. Activation of adenosine 59 triphosphateinduced quenching of chlorophyll fluorescence by reduced plastoquinone. FEBS Letters 119, 141–144.

Horton P, Ruban AV. 1992. Regulation of photosystem II. Photosynthesis Research 34, 375–385.

Horton P, Ruban AV. 1994. The role of LHCII in energy quenching.

In: Baker NR, Bowyer JR, eds. Photoinhibition of photosynthesis— from molecular mechanisms to the field. Oxford: Bios Scientific Publishers, 111–128.

Horton P, Oxborough K, Rees D, Scholes JD. 1988. Regulation of the photochemical efficiency of photosystem II; consequenes for the light response of field photosynthesis. Plant Physiology and Biochemistry 26, 453–460.

Horton P, Ruban AV, Rees D, Pascal A, Noctor GD, Young A. 1991. Control of the light-harvesting function of chloroplast membranes by the proton concentration in the thylakoid lumen: aggregation states of the LHCII complex and the role of zeaxanthin. FEBS Letters 292, 1–4.

Horton P, Ruban AV, Walters RG. 1996. Regulation of light harvesting in green plants. Annual Reviews of Plant Physiology and Plant Molecular Biology 47, 655–684.

Horton P, Ruban AV, Wentworth M. 2000. Allosteric regulation of the light-harvesting system of photosystem II. Philosophical Transactions of The Royal Society London B 355, 1361–1370.

Horton P, Ruban AV, Young AJ. 1999. Regulation of the structure and function of the light-harvesting complexes of photosystem II by the xanthophyll cycle. In: Frank HA, Young AJ, Cogdell JC, eds. The photochemistry of carotenoids: applications in biology.

Kluwer Academic Publishers, 271–291.

Jansson S. 1999. A guide to the Lhc genes and their relatives in Arabidopsis. Trends in Plant Science 4, 236–240.

Johnson GN, Young AJ, Scholes JD, Horton P. 1993. The dissipation of excess excitation-energy in British plant-species.

Plant, Cell and Environment 16, 673–679.

Kuhlbrandt W, Wang DN, Fujiyoshi Y. 1994. Atomic model of plant light-harvesting complex by electron crystallography. Nature 367, 211–214.

Lavaud J, Rousseau B, van Gorkom HJ, Etienne A-L. 2002.

Influence of the diadinoxanthin pool size on the photoprotection in marine planktonic diatom Phaeodactylum tricornutum. Plant Physiology 129, 1398–1406.

Levy H, Tal T, Shaish A, Zamir A. 1993. Cbr, an algal homolog of plant early light-inducible proteins, is a putative zeaxanthin binding protein. Journal of Biological Chemistry 268, 20892–20896.

Li XP, Bjorkman O, Shih C, Grossman AR, Rosenquist M, Jansson S, Niyogi KK. 2000. A pigment-binding protein essential for regulation of photosynthetic light harvesting. Nature 403, 391–395.

Li XP, Muller-Moule P, Gilmore AM, Niyogi KK. 2002b. PsbSdependent enhancement of feedback de-excitation protects photo-

system II from photoinhibition. Proceedings of the National Academy of Sciences, USA 99, 15222–15227.

Li XP, Phippard A, Pasari J, Niyogi KK. 2002a. Structure– function analysis of photosystem II subunit S (PsbS) in vivo.

Functional Plant Biology 29, 1131–1139.

Liu Z, Yan H, Wang K, Kuang T, Zhang J, Gui L, An X, Chang W. 2004. Crystal structure of spinach major light-harvesting ˚ resolution. Nature 428, 287–292. complex at 2.72 A Lokstein H, Hartel H, Hoffmann P, Renger G. 1993. Comparison of chlorophyll fluorescence quenching in leaves of wild type with a chlorophyll b-less mutant of barley. Journal of Photochemistry and Photobiology 19, 217–225.

Ma YZ, Holt NE, Li XP, Niyogi KK, Fleming GR. 2003. Evidence for direct carotenoid involvement in the regulation of photosynthetic light harvesting. Proceedings of the National Academy of Sciences, USA 100, 4377–4382.

Montane MH, Tardy F, Kloppstech K, Havaux M. 1998. Differential control of xanthophylls and light-induced stress proteins, as opposed to light-harvesting chlorophyll a/b proteins, during photosynthetic acclimation of barley leaves to light irradiance.

Plant Physiology 118, 227–235.

Morosinotto T, Baronio R, Bassi R. 2002. Dynamics of chormophore binding to Lhc proteins in vivo and in vitro during operation of the xanthophyll cycle. Journal of Biological Chemistry 277, 36913–36920.

Moya I, Silvestri M, Vallon O, Cinque G, Bassi R. 2001. Timeresolved fluorescence analysis of the photosystem II antenna proteins in detergent micelles and liposomes. Biochemistry 40, 12552–12561.

Mullet JE, Arntzen CJ. 1980. Simulation of grana stacking in a model system. Mediation by a purified light-harvesting pigment– protein complex from chloroplasts. Biochimica et Biophysica Acta 589, 100–117.

Niyogi KK, Bjorkman O, Grossman AR. 1997. Chlamydomonas xanthophyll cycle mutants identified by video imaging of chlorophyll fluorescence quenching. The Plant Cell 9, 1369–1380.

Oquist G, Huner NPA. 2003. Photosynthesis of overwintering evergreen plants. Annual Review of Plant Biology 54, 329–355.

Remy R, Tremolieres, A, Duval JC, Ambardbretteville F, Dubacq JP. 1982. Study of the supramolecular organization of light-harvesting chlorophyll protein (LHCP): conversion of the oligomeric form into the monomeric one by phospholipase-a2 and reconstitution with liposomes. FEBS Letters 137, 271–275.

Robert B, Pascal AA, Horton P, Ruban AV. 2004. Insights into the molecular dynamics of the plant light-harvesting proteins in vivo.

Trends in Plant Science 9, 385–390.

Ruban AV, Horton P. 1992. Mechanism of DpH-dependent dissipation of absorbed excitation energy by photosynthetic membranes. I. Spectroscopic analysis of isolated light-harvesting complexes. Biochimica et Biophysica Acta 1102, 30–38.

Ruban AV, Horton P. 1995a. Regulation of nonphotochemical quenching of chlorophyll fluorescence in plants. Australian Journal of Plant Physiology 22, 221–230.

Ruban AV, Horton P. 1995b. An investigation of the sustained component of nonphotochemical quenching of chlorophyll fluorescence in isolated chloroplasts and leaves of spinach. Plant Physiology 108, 721–726.

Ruban AV, Horton P. 1999. The xanthophyll cycle modulates the kinetics of nonphotochemical energy dissipation in isolated lightharvesting complexes, intact chloroplasts and leaves. Plant Physiology 119, 531–542.

Ruban AV, Lee PJ, Wentworth M, Young, AJ, Horton P. 1999.

Determination of the stoichiometry and strength of binding of xanthophylls to the photosystem II light-harvesting complexes.

Journal of Biological Chemistry 274, 10458–10465.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

Molecular design of the PSII light-harvesting antenna Ruban AV, Pascal A, Robert B. 2000. Xanthophylls of the major photosynthetic light-harvesting complex of plants: identification, conformation and dynamics. FEBS Letters 477, 181–185.

Ruban AV, Pascal AA, Robert B, Horton P. 2002. Activation of zeaxanthin is an obligatory event in the regulation of photosynthetic light harvesting. Journal of Biological Chemistry 277, 7785–7789.

Ruban AV, Rees D, Pascal AA, Horton P. 1992. Mechanism of DpH-dependent dissipation of absorbed excitation energy by photosynthesis II. The relationships between LHCII aggregation in vitro and qE in isolated thylakoids. Biochimica et Biophysica Acta 1102, 39–44.

Ruban AV, Wentworth M, Yakushevska, AE, Andersson J, Lee PJ, Keegstra W, Dekker JP, Boekema EJ, Jansson S, Horton, P. 2003. Plants lacking the main light-harvesting complex retain photosystem II macro-organization. Nature 421, 648–652.

Ruban AV, Wentworth M, Horton P. 2001. Kinetic analysis of nonphotochemical quenching of chlorophyll fluorescence. I. Isolated chloroplasts. Biochemistry 40, 9896–9901.

Ruban AV, Young A, Pascal A, Horton P. 1994. The effects of illumination on the xanthophyll composition of the photosystem II light-harvesting complexes of spinach thylakoid membranes. Plant Physiology 104, 227–234.

373

Ruban AV, Young A, Horton P. 1993. Induction of nonphotochemical energy dissipation and absorbance changes in leaves; evidence for changes in the state of the light-harvesting system of photosystem II in vivo. Plant Physiology 102, 741–750.

Simidjiev I, Stoylova S, Amenitsch H, Javorfi T, Mustardy L, Laggner P, Holzenburg A, Garab G. 2000. Proceedings of the National Academy of Sciences, USA 97, 1473–1476. van Amerongen H, van Grondelle R. 2001. Understanding the energy transfer function of LHCII, the major light-harvesting complex of green plants Journal of Physical Chemistry B 105, 604–617.

Wentworth M, Ruban AV, Horton P. 2000. Chlorophyll fluorescence quenching in isolated light-harvesting complexes induced by zeaxanthin. FEBS Letters 471, 71–74.

Wentworth M, Ruban AV, Horton P. 2003. Thermodynamic investigation into the mechanism of the chlorophyll fluorescence quenching in isolated photosystem II light-harvesting complexes.

Journal of Biological Chemistry 278, 21845–21850.

Yakushevska AE, Keegstra W, Boekema EJ, Dekker JP, Jansson S, Ruban AV, Horton P. 2003. The structure of photosystem II in Arabidopsis: localization of the CP26 and CP29 antenna complexes. Biochemistry 42, 608–613.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009