• Plants often regulate the amount and size of light-harvesting antenna (LHCII) to maximize photosynthesis at low light and avoid photodamage at high light.
• Gas exchange, 77 K chlorophyll fluorescence, photosystem II (PSII) electron transport as well as LHCII protein were measured in leaves irradiated at different light intensities.
• After irradiance transition from saturating to limiting one leaf photosynthetic rate in some species such as soybean and rice declined first to a low level, then increased slowly to a stable value (V pattern), while in other species such as wheat and pumpkin it dropped immediately to a stable value (L pattern). Saturating pre-irradiation led to significant declines of both 77 K fluorescence parameter F685/F735 and light-limited PSII electron transport rate in soybean but not in wheat leaves, indicating that some LHCIIs dissociate from PSII in soybean but not in wheat leaves.
• The L pattern of LHCII-decreased rice mutant and the V pattern of its wild type demonstrate that the V pattern is linked to dissociation/reassociation of some LHCIIs from/to PSII.
The light-harvesting complex of photosystem II (LHCII) is not only a primary site of the thermal dissipation (Elrad et al., 2002) but also an important regulator of light harvesting for photosynthesis in plants. Decreasing of LHCII abundance may be a long-term acclimatizational response to high light (Anderson et al., 1995; Teramoto et al., 2002) while dissociating of some LHCIIs from photosystem II (PSII) core complexes is probably a regulatory way dealing with short-term high light. Based on a reversible dissociation of LHCII from PSII at temperature above 35°C, Sundby & Andersson (1985) assumed that LHCII dissociation from the PSII reaction center complexes could be a short-term regulatory strategy for avoiding over-excitation and destruction of PSII at high light. However, there has been no direct experimental evidence for the assumption. Using the analyses of 77 K chlorophyll fluorescence and sucrose density gradient centrifugation, Hong & Xu (1999a) demonstrated that saturating irradiation led to the dissociation of some LHCIIs from the PSII complex in soybean leaves. Furthermore, through the protein kinase inhibitor experiments with soybean leaves and thylakoids, Zhang and Xu (2003) confirmed that the LHCII dissociation could effectively protect PSII reaction centers against photodamage by reducing the amount of photons transferred to the centers under saturating irradiance. By the difference in seed output between Arabidopsis thaliana mutants and their wild type Kühlheim et al. (2002) demonstrated elegantly that the feedback de-excitation conferred a strong fitness advantage in the field with fluctuating light. Nevertheless, the mechanism for the rapid regulation of light harvesting is not yet clear.
It is well known that light energy used by the reaction centers of the photosynthetic apparatus during photosynthesis is largely derived from light-harvesting complexes or antennae such as LHCII and LHCI. Also, the magnitude of photosynthetic rate is dependent on irradiance and antenna size at limiting irradiance. Therefore, it is reasonable to predict that a decrease in the size of light-harvesting antenna must lead to a significant decline in photosynthetic rate measured at limiting irradiance. The aims of present study were: (1) testing the prediction; (2) obtaining the evidence from leaf gas exchange for the LHCII dissociation induced by saturating irradiation; and (3) examining the universality of the LHCII dissociation as a photoprotective strategy in plants. For these aims the responses of leaf photosynthetic rate to irradiance transition from saturating to limiting one were observed in more than 10 plant species and two different response patterns were found. The analyses of 77 K chlorophyll fluorescence and PSII electron transport and the comparison of the responses between a rice mutant with decreased amount of LHCII and its wild type were made in order to verify the relationship between the response pattern and the LHCII dissociation.
Materials and Methods
Soybean (Glycine max cv. Baimangjie), wheat (Triticum aestivum cv. Gaoyuan 602), cotton (Gossypium hirsutum), spinach (Spinacea oleracea), maize (Zea mays), rice (Oryza sativa), Arabidopsis thaliana cv. Columbia and tobacco (Nicotiana tabacum) plants were grown in a phytotron at a photosynthetic photon flux density (PPFD) of c. 300 µmol m−2 s−1 with a 12 h light/12 h dark cycle. Sweet viburnum (Viburnum odoratissimum), ginkgo (Ginkgo biloba), castorbean (Ricinus communis), fig (Ficus carica), mulberry (Morus alba), bamboo (Bambusa multiplex) and pumpkin (Cucurbita pepo) plants were grown in the field. Experiments were performed using fully expanded sun leaves.
Leaf irradiation and photosynthesis measurement
The plant leaves were first irradiated at limiting irradiance (LI: 300 µmol m−2 s−1) until net photosynthetic rate (Anet) reached a steady state. The limiting irradiance was then replaced by saturating irradiance (SI: 750 and 1000 µmol m−2 s−1, respectively, for plants grown in the phytotron and in the field, LI-SI) for photosynthesis. After Anet was raised gradually to a new steady state the irradiance was reduced to a limiting one (LI-SI-LI) until Anet again reached a steady state. In general, Anet reaches its steady state value after c. 15 min and 30 min of irradiation at limiting and saturating irradiance, respectively.
Anet was measured at 350 µmol CO2 mol−1 and c. 30°C in situ by a portable photosynthetic gas analysis system with an LED light source (LI-COR 6400; LI-COR Inc., Lincoln, NE, USA). For low temperature (77 K) chlorophyll fluorescence analysis, PSII electron transport and LHCII protein measurements, the other leaves were irradiated, according to the procedure of irradiance changes mentioned above, by a metal halogen lamp (1000 W) at room temperature. After irradiation (LI, LI-SI and LI-SI-LI) the leaves were immediately dropped into liquid nitrogen.
Low-temperature fluorescence was measured at 77 K with a 44 W-fluorescence spectrofluorimeter built in our laboratory. F685, F735 and F685/F735 were measured and calculated as previously described (Hong & Xu, 1999a).
Measurement of the PSII electron transport rate
The PSII electron transport rate of thylakoids was measured at strong light (1200 µmol m−2 s−1) or weak light (100 µmol m−2 s−1) by a Clark-type oxygen electrode (assembled by Shanghai Institute of Plant Physiology), using 0.5 mm 1,4-benzoquinone as electron acceptor and 20 mm NH4Cl as uncoupler. Thylakoids were isolated from leaves as previously described (Hong & Xu, 1999b). Considering the difference among repeated experiments, the electron transport rate was given as percentage of that of those thylakoids from LI pre-irradiated leaves in this study.
Measurement of chlorophyll content
Chlorophyll was extracted from fresh leaf segments with 80% acetone and determined according to Arnon (1949).
Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting
Thylakoids were isolated from soybean and wheat leaves, respectively, according to the method described by Hong & Xu (1999b) with all solutions containing 1 mm phenylmethane-sulfonyl fluoride (PMSF; Merck, Darmstadt, Germany) to inhibit protease activity. Thylakoid membrane proteins were resolved by SDS-PAGE according to Laemmli (1970) using 15% acrylamide and 0.5% bisacrylamide in the presence of 6 m urea. Each loaded sample contained 2 µg chlorophyll.
The methods of thylakoid isolation and thylakoid membrane protein extraction as well as SDS-PAGE for rice leaves were basically the same as those for soybean and wheat leaves mentioned earlier. Because there were substantial differences in chlorophyll and some protein contents between rice mutant and its wild-type leaves, the amount of loaded samples was determined according to leaf area rather than chlorophyll or protein amount when SDS-PAGE of thylakoid membrane proteins from rice leaves was made. Therefore, for rice mutant and its wild type the same area leaves and the same volume solutions were used in all isolation and extraction processes. Then, 10 µl of obtained thylakoid membrane protein solutions was loaded on each line of 15% acrylamide SDS-PAGE gel in the presence of 6 m urea. After electrophoresis the gel was stained with Coomassie Brilliant Blue.
The immunodetection was performed according to Jansson et al., 1997) with some modifications. The polypeptides were electrophoretically transferred from the SDS-PAGE gel to nitrocellulose membrane (Amersham Pharmacia, Milton Keynes, UK) within a semi-dry transfer cell (Amersham Pharmacia). After blocking for 2 h in Tris-buffered saline (TBS: 20 mm Tris-HCl, pH 7.6, 150 mm NaCl) buffer supplied with 5% nonfat dried milk powder, the membranes were incubated with anti-LHCII b antibody (a kind gift from Prof. Li-Xin, Zhang) in antibody buffer (TBS, 3% dried milk powder and 0.05% Tween-20) for 1.5 h. The membranes were then washed four times in wash buffer (TBS and 0.2% Tween-20) and incubated with goat antirabbit IgG conjugated with horseradish peroxidase (KPL, Gaithersburg, MD, USA) in antibody buffer for 1.5 h. Following four rinses in wash buffer, the thylakoid LHCIIb proteins were immunodetected using an ECL assay Kit (Amersham Pharmacia) according to the manufacturer's instructions.
Response of net photosynthetic rate to change in irradiance
When irradiance was changed from saturating (750 µmol m−2 s−1) to limiting one (300 µmol m−2 s−1), Anet in soybean leaves declined immediately to a value much lower than that at limiting irradiance before saturating irradiation. Then, Anet rose slowly to a stable value near to that at limiting irradiance before saturating irradiation (Fig. 1a). Moreover, when Anet dropped to a lowest value, stomatal conductance (gs) fell a little compared with that at saturating irradiance, while intercellular CO2 concentration (Ci) jumped to a highest value (data not shown), indicating that the reduction of Anet is not caused by the decline in gs. Using soybean plants grown in the field similar responses were observed (Fig. 1c). After saturating irradiation and subsequent limiting irradiation, the LHCII protein amount of soybean leaves remained roughly the same as that before saturating irradiation (Fig. 1b), indicating that the saturating irradiation does not cause the degradation or net loss of LHCII proteins in these SI-irradiated leaves.
Interestingly, the responses of Anet in wheat leaves to the irradiance transition were significantly different from those in soybean leaves. After irradiance was changed from saturating to limiting one, Anet in wheat leaves decreased immediately to a stable value similar to that before saturating irradiation (i.e. no slow rise followed the sharp drop in Anet; Fig. 2a). Nevertheless, as in the situation in soybean leaves, saturating irradiation did not induce a significant change in LHCII protein amount of wheat leaves (Fig. 2b).
Such a difference in leaf photosynthetic response to the irradiance transition between soybean and wheat exists extensively in other plant species. We found that rice, tobacco, spinach, sweet viburnum, Arabidopsis, fig and mulberry showed the photosynthetic responses similar to soybean, while maize, cotton, pumpkin, ginkgo, castorbean and bamboo showed the photosynthetic responses similar to wheat (Fig. 3). These results indicate that in these species examined there are two different patterns in leaf photosynthetic responses to the irradiance transition from saturating to limiting one. According to the shapes of the response curves, they were defined as V and L patterns, respectively.
In order to demonstrate a speculated relationship between the response pattern and LHCII, we compared leaf photosynthetic responses of a rice mutant with yellow-green leaves (Zh249-M) and its wild-type Zhenhui 249 (Zh249-W) to the irradiance transition. It was found that the photosynthetic response curve of the mutant was the L pattern, while the curve of its wild type was the V pattern. Also, at limiting light the leaf Anet in the mutant was markedly lower than that in its wild type, but it was almost identical with that of its wild type at saturating light (Fig. 4a). By contrast, the mutant had substantially decreased chlorophyll content (Fig. 4b) and some membrane protein (Fig. 4c), especially LHCII protein (Fig. 4d). It appears that the change from the V pattern of Zh249-W to the L pattern of Zh249-M in leaf photosynthetic response to irradiance transition is closely related to the decrease in the LHCII content.
Effects of saturating irradiation on 77 K chlorophyll fluorescence parameters
The fluorescence parameters F685 and F685/F735 declined significantly in SI-irradiated soybean leaves (Fig. 5a,c) but not in SI-irradiated wheat leaves (Figs 5b,d), compared with those of LI-irradiated leaves. and after 1 h of subsequent limiting irradiation these decreased parameters recovered to the levels observed before saturating irradiation. Nevertheless, no significant change in F735 during irradiance transitions was observed in soybean and wheat leaves (Fig. 5).
Effects of saturating pre-irradiation on PSII electron transport rate
The PSII electron transport rate measured at saturating irradiance changed little in thylakoids from the SI pre-irradiated soybean leaves, compared with that in thylakoids from LI pre-irradiated soybean leaves. However, a significant decrease in the rate was observed when measured at limiting irradiance (Fig. 6a).
There was a remarkable difference in the results of PSII electron transport rate measurements between wheat and soybean. The rate measured at limiting irradiance did not decline significantly in thylakoids from SI pre-irradiated wheat leaves compared with that in thylakoids from LI pre-irradiated wheat leaves (Fig. 6b).
The substantial decline in Anet of soybean leaves after transferring from saturating to limiting irradiance is mainly due to a lowered photosynthetic activity of mesophyll cells rather than a lowered stomatal conductance, because the lowest Anet is accompanied by a highest but not lowest intercellular CO2 concentration (data not shown). The lowered photosynthetic activity may be explained simply by a dissociation of some LHCIIs from PSII complexes rather than degradation of LHCII proteins caused by saturating irradiation since no net loss of LHCII proteins occurs in SI-irradiated leaves (Fig. 1b). The dissociation leads to a decrease in the PSII antenna size and thereby a decrease in excitation energy transferred to the PSII reaction centers. At saturating irradiance excess photons can offset the effect of the decreased antenna size on Anet. However, at limiting irradiance the decreased antenna must lead to a large drop of Anet. Thereafter, as the dissociated LHCIIs returned gradually to PSII at limiting irradiance, Anet returned gradually to the level before saturating irradiation.
The explanation for reversible dissociation of some LHCIIs is consistent with the results of chlorophyll fluorescence analysis. At 77 K the chlorophyll fluorescence emissions peaked at 685 nm (F685) and 735 nm (F735) stem from PSII and PSI antennae, respectively. Although F685 comes from the core antenna of PSII (Bassi et al., 1990; Krause & Weis, 1991), the peripheral antenna LHCII also contributes to F685 because photons absorbed by LHCII can be transferred to the core antenna when they are linked to each other. A change in F685 therefore can reflect the change in the status of association of LHCII and PSII core complex. The decreases in F685 and F685/F735 after saturating irradiation (Fig. 5a,c) indicate the dissociation rather than damage or degradation of some LHCIIs, because the possibility of the damage or degradation at saturating irradiance can be ruled out by no change in LHCII amount after saturating irradiation (Fig. 1b) and by subsequent rapid recovery of these parameters at limiting irradiance (Fig. 5c).
The explanation about reversible dissociation of some LHCIIs is also supported by the results of PSII electron transport rate measurement. As well known, light energy required for photosynthesis is mainly from light-harvesting antennae. Photons absorbed by the antennae are transferred to the reaction centers and lead to the charge separation of the centers and subsequent electron transport within two photosystems. The magnitude of the electron transport rate depends on irradiance and antenna size at limiting irradiance. Therefore, a decrease in the antenna size must result in a decline in the electron transport rate at limiting irradiance. That is the case in our experiments. The rate in thylakoids from SI pre-irradiated soybean leaves measured at saturating irradiance changed little while it decreased significantly when measured at limiting irradiance (Fig. 6a). This is because the saturating pre-irradiation induces dissociation of some LHCIIs, leading to the smaller antennae of some PSII core complexes.
Our experimental results obtained with wheat and rice mutant further confirm the link of the pattern of leaf photosynthetic response to irradiance transition with LHCII dissociation. Compared with that before saturating irradiation, there was no decline in Anet of wheat leaves after transition from saturating to limiting irradiance (Fig. 2a), showing that its response curve was the L pattern. Moreover, F685/F735 in SI-irradiated wheat leaves (Fig. 5b,d) and PSII electron transport rate measured at limiting irradiance of thylakoids from SI pre-irradiated wheat leaves (Fig. 6b) did not decline significantly, indicating that no dissociation of LHCII occurs in wheat leaves at saturating irradiance. It appears that the L pattern of leaf photosynthetic response is not accompanied by dissociation of LHCII while the V pattern of the response is linked with dissociation of some LHCIIs, as shown by wheat and soybean, respectively. If the conclusion is correct, the pattern of leaf photosynthetic response to irradiance transition in rice mutant with decreased LHCII content should be the L pattern but not the V pattern. The experimental results of the rice mutant with decreased LHCII content (Fig. 4) have demonstrated the deduction. Because such mutants seldom encounter the danger of photodamage the dissociation of some LHCIIs is not necessary at high irradiance. Of course, from Fig. 4c it is also seen that in addition to the LHCII protein other membrane proteins such as c. 22 kDa and 20 kDa proteins are decreased to a certain extent in the mutant compared with its wild type. It appears that the substantial decrease in LHCII content, as shown by substantially decreased chlorophyll and LHCII protein contents (Fig. 4b,d), is not the sole reason for the L pattern in the rice mutant. Nevertheless, whether proteins such as the 22 kDa and 20 kDa proteins can dissociate reversibly from PSII at saturating irradiance is an open question. If the answer to the question is negative, the decreases in these protein contents are unlikely to be a cause for the change from the V pattern of the wild type to the L pattern of the mutant.
The reversible dissociation of some LHCIIs from PSII complex that we observed in SI-irradiated soybean leaves is unlikely to be a phenomenon of state transitions. State transitions discovered by Murata (1969) and Bonaventura & Myers (1969) provide a mechanism balancing the energy distribution and excitation of the two photosystems for optimizing photosynthesis, and involve a reversible association of some LHCIIs with either PSII or PSI. During the transition from state 1 to state 2 some LHCIIs become phosphorylated, migrate, and attach to PSI (Bassi et al., 1988; Vallon et al., 1991; Samson & Bruce, 1995; Tan et al., 1998; Lunde et al., 2000; Snyders & Kohorn, 2001), while these LHCIIs are dephosphorylated and reattach to PSII during the transition from state 2 to state 1. The protein kinase (Stt7) required for LHCII phosphorylation and state transition has been identified (Depege et al., 2003) and it is inactivated at high irradiance (Rintamäki et al., 2000). Although the association of LHCII with PSI in state 2 had been a controversial question, the data for an Arabidopsis mutant lacking the PSI-H subunit presented by Lunde et al. (2000) have excellently demonstrated that not only does LHCII functionally connect to PSI in state 2, but this connection is also essential for state transition. Obviously, the state transition from state 1 to state 2 is characterized not only by the dissociating of some LHCIIs, but also by the migrating and attaching of the dissociated LHCIIs to PSI. In soybean leaves, however, the LHCII dissociation was not followed by the attaching to PSI, as shown by the unchanged fluorescence parameter F735 (Fig. 5a,c). Moreover, the transition from state 1 to state 2 occurred, as shown by the decreased F685/F735 and increased F735, in weak light-irradiated (Hong & Xu, 1999a) but not in SI-irradiated wheat leaves (Fig. 5b,d), indicating that the transition from state 1 to state 2 is induced by low light rather than high light (Pesaresi et al., 2002). Therefore, the reversible LHCII dissociation occurring in SI-irradiated soybean leaves is not state transition. In fact, it is different from state transition in many respects such as light induction conditions, mechanism and physiological function. The reversible LHCII dissociation is a protective strategy from overexcitation and photodamage of the PSII reaction centers, while state transitions serve to balance energy distribution of the two photosystems.
To our knowledge this is the first report indicating that reversible dissociation of some LHCIIs leads to a decrease in leaf photosynthetic rate at limiting irradiance. The reversible dissociation of some LHCIIs is very important for survival, growth and development of plants under natural conditions. In nature most plants live in a light fluctuating environment. Sunlight incident on surface of plant leaves changes every day in a range of 0 to c. 2000 µmol m−2 s−1 because of changes in sunlight incident angle, cloud cover and shading by plants. Regulation of light harvesting by the reversible dissociation of some LHCIIs can balance the absorption and utilization of light energy to maximize light utilization of PS II reaction centers at low light and avoid photodamage of the centers at high light. From the course of photosynthetic response to irradiance transition in soybean leaves (Fig. 1a) it is estimated that the rejoining process for dissociated LHCII may be performed within c. 15 min after irradiance transition. Obviously, the regulation of light harvesting is fast, favoring the fitness of plant to light fluctuating environment.
Using Arabidopsis mutants Kühlheim et al. (2002) demonstrated that nonphotochemical quenching of excitation energy confers a strong fitness advantage and suggested that this advantage results from the increase in plant tolerance to irradiance variation rather than high light itself. Our finding indicates that some plants such as soybean and rice can tolerate high light itself by the reversible dissociation of some LHCIIs. The mechanism by which the dissociated LHCIIs dissipate the absorbed light energy as heat remains unclear. It has been proved that the PsbS protein plays a key role in nonphotochemical quenching (Li et al., 2000). The relationship between thermal dissipation of dissociated LHCII and the PsbS protein is worthy of further study.
This study was supported by the National Basic Research Program of China (Project no. 2005CB121106). We thank Prof. Rong-Xian, Zhang for the seeds of the rice mutant with yellow-green leaves and Prof. Li-Xin, Zhang for the LHCIIb antibody.