The light harvesting proteins of higher plants play a vital role in the regulation of photosynthesis. These proteins show striking structural and functional flexibility. This dynamic behaviour provides a key physiological function to plants, which adapts them to continuously changing environmental conditions. In particular, dynamic properties of the light harvesting proteins of photosystem II promote the efficient collection of sunlight when the intensity is limiting photosynthesis and effective photoprotection when the intensity is excess.
Photosynthetic light harvesting
Photosystem II (PSII) is the multisubunit chloroplast membrane-associated pigment-protein complex that uses the energy of sunlight to drive the oxidation of water, evolving oxygen, donating electrons into the photosynthetic electron transfer chain and depositing protons into the thylakoid lumen. Along with the Photosystem I (PSI) it forms the electron-proton transfer chain, which drives the synthesis of ATP and NADPH. Vitally important components of both photosystems are the light-harvesting antennae, light-collecting units (mainly LHCI and LHCII for PSI and PSII respectively) which ensure high rates of energy input into the photosynthetic reaction centres (RCI and RCII respectively) by intercepting large numbers of light quanta of various energies/colours. The analogy of LHCs as funnels feeding the RCs with light energy is most appropriate.
Structure and function of LHCII
The light-harvesting antenna of PSII consists of several proteins which together bind around 300 chlorophyll molecules. Associated tightly with the D1/D2 reaction centre are the core antenna complexes CP47 and CP43. The remainder of the antenna consists of the Lhcb proteins, Lhcb1-6. These bind chlorophyll a, chlorophyll b and xanthophylls to form several different complexes – LHCII, CP24, CP26, and CP29. LHCII is the main complex and contains about 40% of the PSII chlorophyll – it is the most abundant chlorophyll protein in nature.
Figure 1. Structural model of LHCII as determined by Lui et al (Nature 428, 287-292, 2004). A. Trimer viewed from its stromal surface B, monomeric subunit. Chlorphyll a (green), chlorophyll b (blue), xanthophylls (yellow).
The single monomeric unit of LHCII is a relatively small protein, approx. 25-28 kDa, containing 3 transmembrane alpha-helical structures and binding 13 molecules of chlorophyll (7 chlorophyll a and 6 chlorophyll b) (Figure 1). Three LHCII monomers are associated together into heterotrimers containing Lhcb1, Lhcb2 and Lhcb3. Additional pigments, carotenoids, are also present in LHCII. Three xanthophylls, luteins, are associated with the helices A and B. A third carotenoid, neoxanthin, is associated with the helix C and the trimer also binds peripherally the carotenoids violaxanthin or zeaxanthin. The xanthophylls play several important roles: a) ensuring correct assembly of LHCII during biosynthesis; b) protection of chlorophylls against photo-oxidation by quenching chlorophyll triplet states and scavenging oxygen radicals; c) being structurally flexible molecules they allow dynamic behaviour of LHCII leading to alteration in chlorophyll interaction and subsequent change in light collection efficiency.
Figure 2 . The macrostructure of photosystem II. A. Results of single particle analysis of solubilised thylakoid membranes from Arabidopsis as reported in Yakushevska et al (Biochemistry 42, 608-613, 2003). The averaged projection of a C2S2M2 supercomplex showing the position of the PSII cores (C), the S and M-types of LHCII trimers (yellow) and minor antennae complexes (green). B. Fitting of the structure of the RC and light harvesting complexes (from Dekker and Boekema, Biochimica et Biophysica Acta 1706, 12-39, 2005).
 | In vivo, two reaction centre complexes build the dimeric PSII core complex, which binds four trimeric LHCII and six monomeric CP24, CP26 and CP29, forming PSII megacomplexes (Figure 2). In the photosynthetic membrane these PSII units are often seen as ordered arrays (Figure 3). Here the LHCII antenna forms an entire network, or macrodomain, of monomeric and trimeric subunits, associated with each other and the reaction center complex. Associated with this macrostructure is the stacking of the complexes together in the characteristic grana membranes. Figure 3. Organisation of PSII in the grana membranes. A, Thin section electron micrograph of Arabidopsis leaf showing the chloroplast thylakoid membranes and the characteristic granal organization; B, Electron micrograph of negatively stained paired grana membrane fragments prepared from Arabidopsis thylakoid membranes; C, the sum of 450 images obtained from the semicrystalline regions of B. The unit cell or repeating motif of the crystal is indicated; D, contour version of C with the position of the S-type and M-type LHCII trimers, and CP29, CP26 and CP24 indicated. |
Three major parameters determine the efficiency of light harvesting: a) the number of pigments and their ability to intercept light quanta with the broad range of energy); b) the time during which energy of light can be kept in antenna so it can be funneled into the reaction center; c) rate of the funneling. Thus, a large absorption cross-section, long excitation lifetime and high rate of energy transfer to the reaction centre are the attributes of an efficient antenna and productive photosynthetic unit. Hence, the efficiency of light harvesting in PSII is determined by the number of LHCII subunits, the pigment order within them, the interaction between subunits and their closeness to the reaction center complex.
The need for control of the light reactions of photosynthesis
 | The light and dark reactions of photosynthesis are tightly coupled, the ATP and NADPH provided by the electron-proton transfer system driving the fixation of CO2 into carbohydrate. During a day, the light spectrum changes dramatically due to filtering by the atmosphere, clouds and sometimes by other plants. Since LHCI and LHCII have different capacities to absorb light of different colours – the former preferring more red light than the latter – there will be a frequent imbalance in light energy input into reaction centres. This imbalance will decrease the electron transfer efficiency, hence the quantum yield of photosynthesis. The light intensity is also a very changeable parameter, sometimes varying 1000-fold. Whenever light input exceeds the capacity of electron transport and CO2 assimilation, there is an excess of light, which may cause photoinhibition - the destabilisation (by over-reduction and over-energisation) and even damage (by photo-oxidation and generation of reactive oxygen species) to photosynthetic components, mainly in PSII. The effects of changes in light delivery can be amplified by alterations in photosynthetic capacity brought about by changes in other environmental factors, or by changes in metabolic demand, which result in variation in the rate of turnover and the stoichiometry of ATP and NADPH. |
| Figure 4. Regulatory interaction between light harvesting, electron transport and carbon assimilation. Red arrows describe the identified feedback and feedforward mechanisms that regulate photosynthetic energy flow. 1. NPQ 2. pH control of electron transport; 3, state transition; 4, redox control over PSI cyclic electron transport; 5, redox control of ATP synthase; 6, redox control of Calvin cycle enzymes; 7, DpH control of Calvin cycle enzymes; 8, ATP control of Calvin cycle enzymes; 9, Pi control of ATP synthesis. (Modified from Horton, P., Photosynthetic Mechanisms and the Environment, J. Barber and N.R. Baker, eds., vol. 6, pp 135-187, Topics in Photosynthesis, Elsevier, 1985). |
| In order to achieve balance and stability of the photosynthetic process, regulatory mechanisms are necessary. The plant cell first has to assemble a chloroplast with the “correct” composition, and then various parts of the photosynthetic process need to be able to adjust their activities in response to internal and external information. Thus, the electron transport and carbon metabolism can be considered to be linked by a feed-back and feed-forward control network (Figure 4). There is a compromise between maximizing the collection and utilization of light, and the avoidance of instability when light is in excess. The most appropriate way of looking at the regulatory mechanisms is, given a fixed composition and a fluctuating environment, that they extend the range of conditions over which photosynthesis can remain in balance – they provide homeostasis of excitation energy level, redox state and DpH – and therefore minimize potential losses in photosynthesis (Figure 5). |  |
| Figure 5. Regulation of light harvesting. Plants acclimate to their light environment by assembling a photosynthetic apparatus that functions optimally in the average conditions experienced. When light levels are less than this, reductions in quantum efficiency may results and when they are in excess of this, there is photoinhibition (blue). Regulation of light harvesting moderate these losses (red) – state transitions maintain high quantum yield under limiting light by ensuring balanced excitiation of PSII and PSI and NPQ dissipates excess excitation energy so preventing photoinhibition. These two processes extend the range of light conditions over which optimal photosynthesis can be achieved, and result ni net photosynthetic gains (yellow). (Modified from Horton, P., Progress in Photosynthesis Research, J. Biggins ed., vol. 2, pp681-688 Martinus-Nijhoff, Dordrecht, 1987). |
To optimise light harvesting function, two distinct regulatory strategies are required. The first aims to balance the input of light energy to PSI and PSII reaction centres in order to optimize electron transfer rate – this is called the State Transitions. The second should regulate the amount of light energy directed to PSII – this is called non-photochemical energy dissipation, or quenching (NPQ). As a result of evolution, both regulatory mechanisms have developed in the LHCII system and rely on its abilities to sense the “state” of the light energy balance and structurally respond, being capable of dynamic behaviour.
State transitions
The Lhcb1 and Lhcb2 polypeptides of LHCII are reversibly phosphorylated by a redox-regulated thylakoid-associated protein kinase. Phosphorylation at threonine residues near the N terminus of LHCII weakens the association between PSII and LHCII, as a result of electrostatic and conformational changes. Phospho-LHCII partitions in favour of PSI, this association depending the presence of the PSI-H subunit – in mutants without PSI-H is phospho-LHCII remains associated with PSII. The kinase is activated when the plastoquinone pool is reduced, this being sensed by the quinone oxidation site on the cytochrome b6f complex. The kinase therefore monitors the relative rates of delivery of excitation energy to PSII and PSI (Figure 6); over-excitation of PSII (State 1) activates the kinase, decreases delivery of excitation energy to PSII and increases transfer to PSI (State 2). Levels of LHCII phosphorylation are lower at high light compared to low light, showing that the state transitions optimise photosynthesis in limiting light. This includes not only optimising linear electron transport, but also the provision of optimal DpH and ATP/NADPH ratios by controlling the proportion of cyclic electron transfer around PSI.
 | Figure 6. Control of the State Transitions by reversible phosphorylation of LHCII. A sub-population of LHCII can be phosphorylated by a thylakoid protein kinase, which is activated when the PQ pool is reduced. Phosphorylated LHCII is released from tight association with PSII and can after lateral diffusion from the grana become associated with PSI. Kinase activity detects the balance of excitation of PSII and PSI, Therefore, the relative absorption cross sections are continuously adjusted to offset any imbalances (Modified from Horton, P., FEBS Letters, 152, 47-52, 1983). |
Nonphotochemical Energy Dissipation
Energy dependent quenching, qE
Dissipation, or quenching, of excess excitation energy absorbed by LHCII occurs by two processes, distinguished by the speeds with which they are induced upon exposure to illumination, their relaxation times in darkness and the light intensity thresh-holds. The major process under “normal” conditions is called qE, since it depends upon the energisation of the thylakoid membrane as a result of DpH formation. Thus, the progressive increase in DpH that occurs as light saturation of photosynthesis is reached is the trigger for energy dissipation. It can therefore be referred to as feedback de-excitation. The formation of qE occurs within seconds-minutes of exposure to excess light, and relaxes with a similar rate in darkness. In contrast, the sustained qI-type of quenching may take several minutes or even hours to appear and relax, and under some conditions (such as low temperature) may be stable for days.
The Site of Quenching
Application of mathematical models for PSII energy transfer showed that quenching associated with qE occurred in the light harvesting antenna. Refinements of the analysis showed that qE is best explained by the transition between two states of the antenna, with different rate constants for energy dissipation. Subsequently, analysis of the chlorophyll fluorescence lifetimes provided direct support for this suggestion. Spectrocopic data and biochemical analysis confirmed that the PSII antenna was the site of quenching. A powerful way to determine which proteins are involved in qE is to investigate plants deficient in specific components. Plants in which Lhcb1&2, Lhcb4, Lhcb5 and Lhcb6 have been separately reduced to less than 5% wild type levels of protein by antisense technology still have qE, although in some cases it is reduced. Thus, it has not proved possible to invoke a particular antenna complex as having an obligatory role. A major landmark in qE research was the identification of the npq4 mutant, a mutant lacking qE and which is deficient in the Lhc-related protein PsbS. This four-helix protein is located in PSII, and contains protonatable amino acid residues which are thought to sense the thylakoid lumen pH.
The Xanthophyll Cycle
Figure 7. The xanthophyll cycle. Violaxanthin is an epoxy-carotenoid bound to the light harvesting complexes. In excess light, violaxanthin de-epoxidase is activated by the DpH and converts violaxanthin into zeaxanthin by removable of the epoxides (red) via the mono-epoxy carotenoid antherxanthin. In low light, zeaxanthin is epoxidised.
Under conditions of excess light, violaxanthin is de-epoxidised to zeaxanthin via the mono-epoxide antheraxanthin due to the activation of violaxanthin de-epoxidase, a thylakoid lumen enzyme (Figure 7). Another enzyme, zeaxanthin epoxidase reverses the reaction in low light. A vast amount of data, from a wide variety of plant species under many different environments showed correlations between the extent of qE and the de-epoxidation state of the xanthophyll cycle pool. Zeaxanthin is an activator of qE in that the apparent pKa of the protonation reaction inducing qE shifts from near 4.5 to well over 5.0 in the presence of zeaxanthin. Since the DpH in vivo does not fall below 5.5, qE will often depend completely on de-epoxidation of violaxanthin. Therefore, qE is an allosteric process, under the control of the interacting effects of protonation and zeaxanthin binding, similar to a regulated enzyme (Figure 8). This mode of regulation explained how the chloroplast could have a DpH high enough in limiting light to allow ATP synthesis without qE, yet in saturating light, how to have maximum electron transport rates and high qE simultaneously.
| Figure 8. The LHCII conformation model for NPQ. The model depicts the way in which de-epoxidation and protonation control the conformation of the PSII antenna, and the extent of energy dissipation. The proximity between the inner rectangles represents the extent of conformational change (e.g. the changes in configuration of chl and xanthophyll in LHC that cause energy dissipation), which governs the efficiency of quenching (thickness of red arrows). States I-IV refer to the difference quenching states – I is the dark-adapted unprotonated unquenched state binding violaxanthin (yellow); state II is the unprotonated zeaxanthin bound state (orange); protonation of state I and II results in formation of states III and IV, respectively but with different pK. Hence zeaxanthin is an allosterically activator qE. (Modified from Horton et al., Annual Reviews of Plant Physiology and Plant Molecular Biology, 47, 655-684, 1996) |  |
However, so far it has proved difficult to distinguish between the proposal that zeaxanthin was directly involved in quenching chlorophyll excited states by chlorophyll/zeaxanthin energy transfer, and the notion that zeaxanthin was working indirectly, modifying or inducing a quenching process intrinsic to the PSII antenna.
The nature of the quenched state of LHCII
Changes in the absorption spectra of chlorophyll and carotenoid occur upon formation of qE. A band with a maximum at 535 nm (ΔA535) has attracted most attention, since its appearance is perfectly correlated to the amount of “quencher”. Resonance Raman spectroscopy of leaves and chloroplasts has shown that ΔA535 is electronic in origin, and arises from a pool of one or two zeaxanthin molecules per PSII, which undergo a strong red shift in the presence of ΔpH. A similar red shift occurs when zeaxanthin binds to PsbS in vitro, and it is concluded that ΔA535 arises from this interaction.
Analysis of fluorescence lifetimes shows that the qE state of the PSII antenna has a lifetime of 0.4 ns, and the unquenched state 1.6 ns. Experiments with isolated light harvesting complexes has shown how such a change can arise. Detergent solubilised LHCII has a lifetime of 4 ns, but reduction of the detergent concentration causes formation of oligomers with an average lifetime of 0.3 ns. This quenching has many features that are similar or identical to qE. The spectral changes in chlorophyll accompanying this in vitro quenching, indicate that a particular LHCII domain is involved. The in vitro quenched state can also be investigated by reconstitution of light harvesting complexes with altered carotenoid content. Substitution of lutein with zeaxanthin in LHCII causes partial chlorophyll fluorescence quenching.
In general, the ease with which LHCII in vitro can adopt a quenched state reflects the high density of pigment within it. Its design as a light harvesting complex means it maximises the amount of chlorophyll per unit volume. Specific features of protein structure are needed to control the interactions between pigments, to promote energy transfer (see II) and to prevent formation of quenchers – dimers or excimers of chlorophylls and/or chlorophyll-xanthophyll associates. Therefore, only rather small changes in structure are needed to allow a quencher to form. Indeed the activation energy for the formation of a quencher in LHCII is only 6 Kcal/mole, indicative of the breakage of just a couple of H bonds. Thus, LHCII is uniquely poised to allow a switch between an unquenched state functioning in photosynthetic energy capture and a quenched state, dissipating excess energy as heat.
Signal transduction and dynamics
Figure 9. Dynamic aspects of NPQ. Upon illumination of a dark adapted leaf, NPQ forms with two phases. The first is rapid and coincides with the formation of DpH. The second slower phase correlates with accumulation of zeaxanthin. A brief dark period causes collapse of DpH but very little zeaxanthin epoxidation; a portion of NPQ also relaxes, the qE component. Upon re-illumination NPQ forms rapidly with almost no slower phase. This quenching relaxes with two dominant phases – the rapidly relaxing qE and the more slowly relaxing qI, which correlates with disappearance of zeaxanthin. These characteristic kinetics are explained by the model in Figure 8; the first illumination results in the initial formation of state III, which is gradually replaced by state IV. Relaxation of NPQ is then mostly a conversion to the state II. The second illumination reflects the conversion directly from state II to state IV. The different pK for these transitions explains why for the same DpH the level of qE is larger and forms more quickly in the second illumination when qE has been “activated”.
Sudden increase to saturating light intensity results in an increase in DpH across the thylakoid membrane and the pH of the lumen drop to around 5.5. This exerts two effects. Firstly, the violaxanthin de-epoxidase is activated. During the following 30 minutes the de-epoxidation of the xanthophyll cycle pool is progressively increased. Secondly, protonation of thylakoid membrane proteins, particular PsbS occurs, resulting in a rapid burst of induction of qE within the first 30-60 seconds. Thereafter, further increase in qE occurs as the level of zeaxanthin increases. During this period until maximum qE is reached, about 15-30 minutes, the quenching becomes progressively more stable, a feature refereed to as qI. Hence if the light is interrupted by a dark period, qE relaxes completely in a few minutes, but some quenching (qI) remains. If the sample, is re-illuminated after a short period in darkness (e.g 10 minutes) there is little epoxidation, and qE develops rapidly and completely within 30-60 seconds (Figure 9).
Biological diversity of qE
The amount of qE in different plants is highly variable. High light grown plants may have 2-3 times more qE capacity and plant species adapted to growing in stressful environments have a much higher qE capacity than those inhabiting milder conditions. Genetic manipulation to increase the level of the PsbS protein results in an increase in qE, suggesting that the level of this protein is an important determinant of qE capacity.
Algae exhibit different qE properties to higher plants. In Chlamydomonas, qE is smaller and forms more slowly, and mutation of an LHCII subunit decreases the capacity of qE, suggesting that here a less efficient process is controlled directly by the antenna. In diatoms, which have light harvesting proteins of a different type, qE can be much larger than in higher plants, and its formation and relaxation totally depend upon the de-epoxidation of the violaxanthin analogue, diadinoxanthin. It seems that some features of the molecular mechanism for the regulation of light harvesting have diverged during evolution of classes of organisms.
Other responses of LHCII to excess light
The sustained qI-type of quenching appears to also result, in part at least, from alteration in the antenna of PSII. At low temperature there may be strong quenching which has some features diagnostic of a increase in aggregation state of LHCII – formation of a red-shifted fluorescence emission band at 77 K. This state of LHCII seems to be semi-permanent in some evergreen plants during winter. Quenching may also be induced by light. The mechanism by which such quenching is formed is currently unknown, although one possibility is that it arises from a direct effect of light on LHCII. Such an effect is well known for LHCII in vitro, and it appears to be triggered by a light-induced dissociation of trimers into monomers. At elevated temperature increased light induced quenching arises from a different process - LHCII dissociates from PSII to be quenched by PSI, in a process that resembles the state transition, and in fact phosphorylation of LHCII and elevated temperature act synergistically.
When plants grown under low light are exposed to a sustained increase in light intensity, LHCII degradation is induced. Only monomers can be the substrates for proteolysis, and so this process must be preceded by a light-induced breakdown of trimers. Little is known about the biochemistry or regulation of this proteolytic process. Under extreme conditions, not only LHCII but complete photosystems are degraded as the plants seek to minimize oxidative damage.
The content of LHCII varies significantly depending on the light intensity under which the plants are grown. Typically, a low light grown plant may have 4-5 trimers per PSII core complex, whereas in high light this reduces to 1-2. The extra LHCII in low light plants appears to be present in membrane domains deficient in PSII core complexes. These LHCII domains can still be involved in efficient energy transfer, even transferring energy to pigments on opposite membranes in the grana.
The adjustment of the composition of photosystem II upon growth in high light compared to low light maintains the redox potential and DpH at the correct level, optimising photosynthesis and avoiding oxidative stress. At higher growth irradiance, when LHCII can not decrease any more, there are increases in the capacity for photoprotection (e.g. through an increase in xanthophyll cycle activity or PsbS content) and, eventually, decrease in the chlorophyll content of the leaf in order to lower light absorption. Even macroscopic events such chloroplast movements or changes in leaf orientation can be similarly viewed as contributing to the optimisation of light harvesting and counter-acting photodamage. Thus, plants do whatever they can, by a multitude of mechanisms, to limit the level of excitation energy in the PSII antenna, giving balance with the demands of photosynthesis, and optimising the redox state and ΔpH.