Folding, assembly, and stability of the major light-harvesting complex of higher plants, LHCII, in the presence of native lipids.

The influence of thylakoid lipids on the association kinetics and thermal stability of the major light-harvesting complex of photosytem II (LHCII) has been studied in vitro. The apoprotein, light-harvesting chlorophyll a/b-binding protein (Lhcb1), can be refolded and complexed with pigments in detergent solution even in the absence of lipids. Two thylakoid lipids, phosphatidyl glycerol and digalactosyl diacylglycerol, are known to interact specifically with LHCII in vivo. Here we show that both of these lipids, as well as monogalactosyl diacylglycerol, stabilize reconstituted LHCII toward thermal denaturation. Two slow kinetic phases are connected with the establishment of energy transfer between chlorophyll b and chlorophyll a and, thus, are thought to reflect the formation of the pigment-protein complex with tightly coupled chlorophylls. The lipids studied here all have the same effect on the rate of complex assembly in vitro and slow these two kinetic phases by the same degree. Both kinetic phases also slow when reactant concentrations are decreased, suggesting that the corresponding reaction step(s) involve(s) pigment binding.

Calcium binding to the photosystem II subunit CP29.

We have identified a Ca(2+)-binding site of the 29-kDa chlorophyll a/b-binding protein CP29, a light harvesting protein of photosystem II most likely involved in photoregulation. (45)Ca(2+) binding studies and dot blot analyses of CP29 demonstrate that CP29 is a Ca(2+)-binding protein. The primary sequence of CP29 does not exhibit an obvious Ca(2+)-binding site therefore we have used Yb(3+) replacement to analyze this site. Near-infrared Yb(3+) vibronic side band fluorescence spectroscopy (Roselli, C., Boussac, A., and Mattioli, T. A. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 12897-12901) of Yb(3+)-reconstituted CP29 indicated a single population of Yb(3+)-binding sites rich in carboxylic acids, characteristic of Ca(2+)-binding sites. A structural model of CP29 presents two purported extra-membranar loops which are relatively rich in carboxylic acids, one on the stromae side and one on the lumenal side. The loop on the lumenal side is adjacent to glutamic acid 166 in helix C of CP29, which is known to be the binding site for dicyclohexylcarbodiimide (Pesaresi, P., Sandonà, D., Giuffra, E. , and Bassi, R. (1997) FEBS Lett. 402, 151-156). Dicyclohexylcarbodiimide binding prevented Ca(2+) binding, therefore we propose that the Ca(2+) in CP29 is bound in the domain including the lumenal loop between helices B and C.

Effects of copper and zinc ions on photosystem II studied by EPR spectroscopy.

The effect of Zn(2+) or Cu(2+) ions on Mn-depleted photosystem II (PS II) has been investigated using EPR spectroscopy. In Zn(2+)-treated and Cu(2+)-treated PS II, chemical reduction with sodium dithionite gives rise to a signal attributed to the plastosemiquinone, Q(A)(*)(-), the usual interaction with the non-heme iron being lost. The signal was identified by Q-band EPR spectroscopy which partially resolves the typical g-anisotropy of the semiquinone anion radical. Illumination at 200 K of the unreduced samples gives rise to a single organic free radical in Cu(2+)-treated PS II, and this is assigned to a monomeric chlorophyll cation radical, Chl a(*)(+), based on its (1)H-ENDOR spectrum. The Zn(2+)-treated PS II under the same conditions gives rise to two radical signals present in equal amounts and attributed to the Chl a(*)(+) and the Q(A)(*)(-) formed by light-induced charge separation. When the Cu(2+)-treated PS II is reduced by sodium ascorbate, at >/=77 K electron donation eliminates the donor-side radical leaving the Q(A)(*)(-) EPR signal. The data are explained as follows: (1) Cu(2+) and Zn(2+) have similar effects on PS II (although higher concentrations of Zn(2+) are required) causing the displacement of the non-heme Fe(2+). (2) In both cases chlorophyll is the electron donor at 200 K. It is proposed that the lack of a light-induced Q(A)(*)(-) signal in the unreduced Cu(2+)-treated sample is due to Cu(2+) acting as an electron acceptor from Q(A)(*)(-) at low temperature, forming the Cu(+) state and leaving the electron donor radical Chl a(*)(+) detectable by EPR. (3) The Cu(2+) in PS II is chemically reducible by ascorbate prior to illumination, and the metal can therefore no longer act as an electron acceptor; thus Q(A)(*)(-) is generated by illumination in such samples. (4) With dithionite, both the Cu(2+) and the quinone are reduced resulting in the presence of Q(A)(*)(-) in the dark. The suggested high redox potential of Cu(2+) when in the Fe(2+) site in PS II is in contrast to the situation in the bacterial reaction center where it has been shown in earlier work that the Cu(2+) is unreduced by dithionite. It cannot be ruled out however that Q(A)-Cu(2+) is formed and a magnetic interaction is responsible for the lack of the Q(A)(-) signal when no exogenous reductant is present. With this alternative possibility, the effects of reductants would be explained as the loss of Cu(2+) (due to formation of Cu(+)) leading to loss of the Cu(2+) from the Fe(2+) site due to the binding equilibrium. The quite different binding and redox behavior of the metal in the iron site in PS II compared to that of the bacterial reaction center is presumably a further reflection of the differences in the coordination of the iron in the two systems.

Thermoluminescence measurements on chloride-depleted and calcium-depleted photosystem II

Photosystem II (PSII) in which O2 evolution was inhibited by depletion of either chloride or calcium ions was studied by thermoluminescence (TL) and luminescence (L) measurements in the presence and absence of 3-(3',4'-dichlorophenyl)-1,1-dimethyl urea (DCMU). Cl--depleted PSII gives rise to TL and L signals which are similar to those in untreated controls i.e., DCMU shifts the TL band from 30 degreesC to 8 degreesC and suppresses the L component with t1/2=10-15 s. In Ca2+-depleted PSII a TL-band at around 50 degreesC and a slow luminescence decay (t1/2=60 s) is observed. Under these conditions, DCMU does not lead to a downshift of the peak temperature of the TL-band nor does it accelerate the decay kinetics of the luminescence. This indicates that in Ca2+-depleted PSII the QA/QB electron transfer is inhibited prior to the addition of DCMU while in Cl--depleted PSII QA/QB electron transfer seems unaffected. These results are consistent with previous fluorescence measurements which showed that the midpoint potential of the redox couple QA/QA- is unchanged in Cl--depleted PSII compared to the control while in Ca2+-depleted PSII it is shifted towards a more positive value [A. Krieger, A.W. Rutherford, Biochim. Biophys. Acta, 1319 (1997) 91-98]. In the literature there are several conflicting reports concerning the TL in Ca2+ and Cl--depleted material so we attempted to understand the origin of some of these discrepancies. We find that in the absence of cryoprotectants, excitation of TL at low temperatures leads to an upshift of TL-bands in Cl--depleted PSII, both in the presence and absence of DCMU, while the peak temperature of TL-bands in control and Ca2+-depleted PSII are downshifted. When TL is excited at 20 degreesC or at low temperature in the presence of a cryoprotectant then there was no shift of the peak temperature of TL-bands. These unexpected results suggest that the formation of the charge pair triggers modifications in its environment and that the exact nature of these modifications differs depending on the temperature of excitation. It seems that once these modifications have occurred at a given temperature they remain 'locked in' being unaffected by subsequent temperature changes until charge recombination has occurred. Copyright 1998 Elsevier Science B.V.

Spectroscopic characterization of intermediate steps involved in donor-side-induced photoinhibition of photosystem II.

The reaction center protein D1 in photosystem II shows a high turnover during illumination. The degradation of the D1 protein is preceded by photoinhibition of the electron transport in photosystem II. There are two distinct mechanisms for this: acceptor-side- and donor-side-induced photoinhibition. Here, donor-side-induced photoinhibition was studied in photosystem II membranes after Cl- depletion or washing with tris(hydroxymethyl)aminomethane (Tris) which destroys water oxidation, reversibly or irreversibly, respectively. Photoinhibition after these treatments leads to fast degradation of the D1 protein, and the mechanism behind this was investigated. Illumination of Cl- depleted photosystem II membranes resulted in a rapid and simultaneous inhibition of Cl(-)-reconstitutable oxygen evolution, loss of 2 Mn ions per photosystem II center, increase in the electron transfer between the electron donor diphenylcarbazide and electron acceptor 2,6-dichlorophenolindophenol, and an increase in the EPR signal IIfast from tyrosine-Zox. The destruction of the Mn cluster leads to the loss of oxygen evolution and to an increased accessibility for diphenylcarbazide to donate electrons to Tyr-Zox. The increase in the EPR signal from Tyr-Zox can be explained by slower reduction kinetics of Tyr-Zox due to the Mn release. On a longer photoinhibition time scale, a decrease in the amplitude of Tyr-Zox and inhibition of the electron transport from diphenylcarbazide to 2,6-dichlorophenolindophenol occurred simultaneously in both Cl(-)-depleted and Tris-washed photosystem II membranes. These slower photoinhibition reactions were then studied in detail in Tris-washed photosystem II membranes. Compared to photoinhibition of Tyr-Zox, the EPR signal from tyrosine-Dox decreased much slower. Tyr-Dox was photoinhibited in parallel with the EPRsignals from reduced QA, reduced pheophytin, and an oxidized chlorophyll radical (chlorophyllz). This shows that the acceptor side components and the primary charge separation reaction (P680+ pheophytin-) were operational although Tyr-Z was inactivated. The amount of the D1 protein also declined in parallel with Tyr-Dox, which shows that the D1 protein is not damaged until long after the Mn complex and Tyr-Z have become inactivated. Instead, it is likely that the strongly oxidizing P680+ is responsible for the damage to the D1 protein.

Copper(II) inhibition of electron transfer through photosystem II studied by EPR spectroscopy.

EPR spectroscopy was applied to investigate the inhibition of electron transport in photosystem II by Cu2+ ions. Our results show that Cu2+ has inhibitory effects on both the donor and the acceptor side of photosystem II. In the presence of Cu2+, neither EPR signal IIvery fast nor signal IIfast, which both reflect oxidation of tyrosinez, could be induced by illumination. This shows that Cu2+ inhibits electron transfer from tyrosinez to the oxidized primary donor P680+. Instead of tyrosinez oxidation, illumination results in the formation of a new radical with g = 2.0028 +/- 0.0002 and a spectral width of 9.5 +/- 0.3 G. At room temperature, this radical amounts to one spin per PS II reaction center. Incubation of photosystem II membranes with cupric ions also results in release of the 16 kDa extrinsic subunit and conversion of cytochrome b559 to the low-potential form. On the acceptor side, QA can still be reduced by illumination or chemical reduction with dithionite. However, incubation with Cu2+ results in loss of the normal EPR signal from QA- which is coupled to the non-heme Fe2+ on the acceptor side (the QA(-)-Fe2+ EPR signal). Instead, reduction of QA results in the formation of a free radical spectrum which is 9.5 G wide and centered at g = 2.0044. This signal is attributed to QA- which is magnetically decoupled from the non-heme iron. This suggests that Cu2+ displaces the Fe2+ or severely alters its binding properties. The inhibition of tyrosinez is reversible upon removal of the copper ions with EDTA while the modification of QA was found to be irreversible.

Fast oxygen-independent degradation of the D1 reaction center protein in photosystem II.

The D1 reaction center protein in photosystem II is rapidly degraded during illumination of chloride-depleted or Tris-washed thylakoids. The degradation is independent of oxygen and occurs under anaerobic conditions provided that electrons can flow through the acceptor-side of photosystem II. This shows that oxygen-derived reactive species are not necessarily involved in the light-dependent damage of the D1 protein. Instead the illumination of chloride-depleted or Tris-washed thylakoids induces long-lived, strongly oxidizing radicals on the donor-side of photosystem II which are suggested to be the damaging species for the D1 protein.

Light-dependent degradation of the D1 protein in photosystem II is accelerated after inhibition of the water splitting reaction.

Strong illumination of oxygen-evolving organisms inhibits the electron transport through photosystem II (photoinhibition). In addition the illumination leads to a rapid turnover of the D1 protein in the reaction center of photosystem II. In this study the light-dependent degradation of the D1 reaction center protein and the light-dependent inhibition of electron-transport reactions have been studied in thylakoid membranes in which the oxygen evolution has been reversibly inhibited by Cl- depletion. The results show that Cl(-)-depleted thylakoid membranes are very vulnerable to damage induced by illumination. Both the D1 protein and the inhibition of the oxygen evolution are 15-20 times more sensitive to illumination than in control thylakoid membranes. The presence, during the illumination, of the herbicide 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) prevented both the light-dependent degradation of the D1 protein and the inhibition of the electron transport. The protection exerted by DCMU is seen only in Cl(-)-depleted thylakoid membranes. These observations lead to the proposal that continuous illumination of Cl(-)-depleted thylakoid membranes generates anomalously long-lived, highly oxidizing radicals on the oxidizing side of photosystem II, which are responsible for the light-induced protein damage and inhibition. The presence of DCMU during the illumination prevents the formation of these radicals, which explains the protective effects of the herbicide. It is also observed that in Cl(-)-depleted thylakoid membranes, oxygen evolution (measured after the readdition of Cl-) is inhibited before electron transfer from diphenylcarbazide to dichlorophenolindophenol.(ABSTRACT TRUNCATED AT 250 WORDS)