Calcium modulates the photoassembly of photosystem II (Mn)4-clusters by preventing ligation of nonfunctional high-valency states of manganese.

The requirement for Ca2+ in the Mn(2+)-dependent photoactivation of oxygen evolution was re-evaluated using 17 kDa/24 kDa-less photosystem II (PSII) membranes depleted of (Mn)4-clusters by NH2OH extraction. At optimum conditions (1 mM Mn2+/10 microM 2,6-dichlorphenolindophenol (DCIP)/20 mM Ca2+), the light-induced increase of oxygen-evolution activity, the increase of membrane-bound Mn, and the B-band thermoluminescence emission intensity occurred in parallel. The extent of recovery of the oxygen-evolution activity was equivalent to 88% and 66% of the activity shown by parent NaCl-extracted PSII membranes and by PSII membranes, respectively. Neither photodamage of primary electron transport nor photoligation of nonfunctional Mn > or = 3+ occurred. Analyses of the Ca2+ concentration dependence for the maximum recovery of oxygen evolution activity gave evidence for Ca(2+)-binding site(s) having Km values of approximately 38 and approximately 1300 microM. Illumination of membranes in the strict absence of Ca2+ resulted in large increases (up to 18 Mn/200 chlorophyll) of EDTA nonextractable, EPR silent, nonfunctional membrane-bound Mn > or = 3+ and small increases of oxygen-evolution capability, dependent on pH and concentrations of Mn2+ and DCIP. No photodamage of primary electron transport and only approximately 17% decrease of AT-band thermoluminescence occurred during the photoligation of the Mn > or = 3%. In the strict absence of Ca2+, significant recovery of oxygen-evolution activity was obtained under a limited set of conditions permitting photoligation of a limited abundance of the nonfunctional Mn > or = 3+. Small (NH2-OH, H2O2) as well as bulky external reductants readily reduced and dissociated the Mn > or = 3+ from the membranes. Reillumination of these membranes under optimal conditions for photoactivation (plus Ca2+) gave a high yield of (Mn)4-clusters and oxygen-evolution capability. Similarly, simple addition of Ca2+ to membranes containing nonfunctional Mn > or = 3+ followed by reillumination resulted in the conversion of Mn > or = 3+ to (Mn)4-clusters. It is argued that Ca2+ promotes the conformational change involved in the conversion of the Mn2+ mononuclear intermediate to the Mn(3+)-Mn2+ binuclear intermediate in the photoactivation mechanism, thereby permitting photoassembly of (Mn)4-clusters and preventing photo-inactivation by Mn > or = 3+ ions.

Superoxide contributes to the rapid inactivation of specific secondary donors of the photosystem II reaction center during photodamage of manganese-depleted photosystem II membranes.

The role of superoxide in the mechanism of photoinactivation of the secondary donors of the reaction center of photosystem II membranes depleted of Mn by extraction with NH2OH plus EDTA (NH2OH/EDTA-PSII) was assessed. EPR analyses (g = 2 region) in continuous light, optical kinetic spectrophotometric analyses of P680+ and Car+, and AT-band emission measurements were made after various durations of weak and strong light treatment of NH2OH/EDTA-PSII in the presence and absence of superoxide dismutase, or of PSII electron acceptors to suppress superoxide formation. Additionally, flash-induced variable fluorescence of chlorophyll a and the capabilities of the membranes of photooxidize Mn2+ (in the presence of H2O2) via a high-affinity site (Km approximately 180 nM) and to carry out the photoactivation of the Mn-cluster were determined. In the absence of any additions to the NH2OH/EDTA-PSII membranes which were highly depleted of Mn, weak light treatment caused rapid (t1/2 approximately 20 s) and parallel losses of (a) the approximately 10 microseconds phase of P680+ reduction, which reflects the TyrZ-->P680+ reaction, (b) the amplitude of chlorophyll a variable fluorescence, (c) the capability to accumulate the TyrZ(+)-radical in continuous light, and (d) the capability to photooxidize Mn2+/H2O2 in continuous light. As reported previously [Blubaugh et al. (1991) Biochemistry 30, 7586-7597], a dark-stable 12-G-wide featureless EPR signal centered at g = 2.004 was formed rapidly during illumination. This signal previously was tentatively identified as a Car+ radical and was suggested to contribute to the quenching of chlorophyll a variable fluorescence and to the slowing of the TyrZ-->P680+ reaction. However, we failed to detect Car+ formation by sensitive optical spectrophotometry and obtained no definable evidence for either a quencher of fluorescence other than P680+ itself or a slowing of the TyrZ-->P680+ reaction. Addition of a saturating concentration (96 units/mL) of superoxide dismutase diminished the rate of photodamage(s) by approximately 30-fold, but did not abolish it completely. Superoxide dismutase similarly suppressed strong light-induced photodamages, causing the loss of capability to photooxidize Mn2+/H2O2, to carry out photoactivation, and to generate the AT-band emission as well as TyrZ+ EPR signal. In contrast to others, we found no evidence that the initial target(s) of photodamage is (are) different in weak versus strong light treatment. The totality of the results suggests that the initial event in either weak light or strong light photodamage of NH2OH/EDTA-PSII is a decoupling of the redox activity of TyrZ from P680.(ABSTRACT TRUNCATED AT 400 WORDS)

A recollection of the development of the Kok-Joliot model for photosynthetic oxygen evolution.

Twenty-five years of period-four O2-flash yield oscillation are celebrated with a personal recollection of the development of the Kok-Joliot model for photosynthetic oxygen evolution.

Photoinhibition of hydroxylamine-extracted photosystem II membranes: studies of the mechanism.

The effects of photosystem II (PSII) exogenous electron donors and acceptors on the kinetics of weak light photoinhibition of NH2OH/EDTA-extracted spinach PSII membranes were examined. Under aerobic conditions, Mn2+ (approximately 1 Mn/reaction center; Km approximately 400 nM) inhibited photoinactivation and approximately 1 Mn/reaction center plus 100 microM NH2NH2 gave almost complete protection. In the absence of electron donors, strict anaerobiosis greatly inhibited photoinactivation even in the presence of an electron acceptor. Under aerobic conditions, the addition of electron acceptors (FeCN, DCIP), oxyradical scavengers, or superoxide dismutase strongly suppressed rates of photodamages. Increase in the concentrations of superoxide above those produced by illuminated NH2OH/EDTA-photosystem II membranes increased the rates of damage in the light but gave no damage in the dark. Scavengers of hydroxyl radicals and singlet oxygen did not suppress the rates of aerobic photoinhibition. These findings, along with others, lead us to conclude that photodamage of the secondary donors of the PSII reaction center occurs by two mechanisms: (1) a rapid superoxide and tyrosine YZ+ dependent process and (2) a slower process in which P680+/Chl+ catalyze the damages.

Photoinhibition of hydroxylamine-extracted photosystem II membranes: identification of the sites of photodamage.

Electron paramagnetic resonance (EPR) analyses (g = 2 region) and optical spectrophotometric analyses of P680+ were made of NH2OH-extracted photosystem II (PSII) membranes after various durations of weak-light photoinhibition, in order to identify the sites of damage responsible for the observed kinetic components of the loss of electron transport [Blubaugh, D.J., & Cheniae, G.M. (1990) Biochemistry 29, 5109-5118]. The EPR spectra, recorded in the presence of K3Fe(CN)6, gave evidence for rapid (t1/2 = 2-3 min) and slow (t1/2 = 3-4) losses of formation of the tyrosyl radicals YZ+ and YD+, respectively, and the rapid appearance (t1/2 = 0.8 min) of a 12-G-wide signal, centered at g = 2.004, which persisted at 4 degrees C in subsequent darkness in rather constant abundance (approximately 1/2 spin per PSII). This latter EPR signal is correlated with quenching of the variable chlorophyll a fluorescence yield and is tentatively attributed to a carotenoid (Car) cation. Exogenous reductants (NH2OH greater than or equal to NH2NH2 greater than DPC much greater than Mn2+) were observed to reduce the quencher, but did not reverse other photoinhibition effects. An additional 10-G-wide signal, tentatively attributed to a chlorophyll (Chl) cation, is observed during illumination of photoinhibited membranes and rapidly decays following illumination. The amplitude of formation of the oxidized primary electron donor, P680+, was unaffected throughout 120 min of photoinhibition, indicating no impairment of charge separation from P680, via pheophytin (Pheo), to the first stable electron acceptor, QA. However, a 4-microsecond decay of P680+, reflecting YZ----P680+, was rapidly (t1/2 = 0.8 min) replaced by an 80-140 microsecond decay, presumably reflecting QA-/P680+ back-reaction. Photoinhibition caused no discernible decoupling of the antenna chlorophyll from the reaction center complex. We conclude that the order of susceptibility of PSII components to photodamage when O2 evolution is impaired is Chl/Car greater than YZ greater than YD much greater than P680, Pheo, QA.

Kinetics of photoinhibition in hydroxylamine-extracted photosystem II membranes: relevance to photoactivation and sites of electron donation.

Kinetic analyses were made of the effects of weak-light photoinhibition on the capacity of NH2OH-extracted photosystem II membranes to photooxidize the exogenous electron donors Mn2+, diphenylcarbazide, and I- or to assemble functional water-oxidizing complexes during photoactivation. The loss of capacity for photooxidation of the donors showed two first-order components (half-times of 2-3 min and 1-4 h) with relative amplitudes dependent on the donor, suggesting two photodamageable sites of electron donation (sites 1 and 2, respectively), a conclusion confirmed by analyses of velocity curves of electron donation by each donor. All of the donors appear to be oxidized preferentially by site 1 both at saturating and at limiting light intensity; however, the contribution by site 2 was nearly comparable in saturating light. Loss of photoactivation also exhibited biphasic kinetics, with components having half-times of approximately 0.8 and 3.2 min. The major component (t1/2 = 3.2 min) corresponded to loss of site 1; essentially no photoactivation was observed after its loss. From these and other analyses, we conclude (1) the relative contributions of site 1 and site 2 to the photooxidation of various exogenous electron donors is determined largely by the rates of equilibration of the donors with the two sites, and (2) only site 1 contributes to photoactivation of the water-oxidizing complex. Site 1 is attributed to tyrosine Z of the reaction center's D1 polypeptide. The molecular identity of site 2 is unknown but may be tyrosine D of the D2 polypeptide.

Early Inhibition of Photosynthesis during Development of Mn Toxicity in Tobacco.

Early physiological effects of developing Mn toxicity in young leaves of burley tobacco (Nicotiana tabacum L. cv KY 14) were examined in glass-house/water cultured plants grown at high (summer) and low (winter) photon flux. Following transfer of plants to solutions containing 1 millimolar Mn(2+), sequential samplings were made at various times for the following 9 days, during which Mn accumulation by leaves increased rapidly from approximately 70 on day 0 to approximately 1700 and approximately 5000 microgram per gram dry matter after 1 and 9 days, respectively. In plants grown at high photon flux, net photosynthesis declined by approximately 20 and approximately 60% after 1 and 9 days, respectively, and the onset of this decline preceded appearance (after 3 to 4 days) of visible foliar symptoms of Mn toxicity. Intercellular CO(2) concentrations and rates of transpiration were not significantly affected; moreover, the activity of the Hill and photosystem I and II partial reactions of chloroplasts remained constant despite ultimate development of severe necrosis. Though the activity of latent or activated polyphenol oxidase increased in parallel with Mn accumulation, neither leaf respiration nor the activity of catalase [EC 1.11.1.6] and peroxidase [EC 1.10.1.7] were greatly affected. These effects from Mn toxicity could not be explained by any changes in protein or chlorophyll abundance. Additionally, they were not a consequence of Mn induced Fe deficiency. Therefore, inhibition of net photosynthesis and enhancement of polyphenol oxidase activity are early indicators of excess Mn accumulation in tobacco leaves. These changes, as well as leaf visual symptoms of Mn toxicity, were less severe in plants cultured and treated at low photon flux even though the rates of leaf Mn accumulation at high and low photon flux were essentially equivalent.

Evidence for Effects on the in Vivo Activity of Ribulose-Bisphosphate Carboxylase/Oxygenase during Development of Mn Toxicity in Tobacco.

The progressive decrease in net photosynthesis accompanying development of Mn toxicity in young leaves of burley tobacco (Nicotiana tabacum L. cv KY 14) is a result of effects on in vivo activity of ribulose-1,5-bisphosphate (RuBP) carboxylase/oxygenase (rubisco, EC 4.1.1.39). This conclusion is supported by: (a) decrease in rates of CO(2) depletion during measurements of CO(2) compensation, (b) increase in leaf RuBP concentrations, (c) progressive decreases in rate-constants of RuBP loss (light to dark transition analyses) with progressive increases of leaf Mn concentrations, and (d) restoration of diminished rates of net photosynthesis to control rates by elevated CO(2) (5%). Moreover, elevated CO(2) (1100 microliters per liter) during culture of Mn-treated plants decreased elevated RuBP concentrations to control levels and alleviated foliar symptoms of Mn toxicity. These effects of Mn toxicity on in vivo activity of rubisco were not expressed by in vitro kinetic analyses of rubisco prepared under conditions to sequester Mn or to adsorb polyphenols or their oxidation products. Similarly, the in vitro activity of fructose bisphosphatase (EC 3.1.3.11) was unaffected by Mn toxicity.

Studies on 17,24 kD Depleted Photosystem II Membranes : I. Evidences for High and Low Affinity Calcium Sites in 17,24 kD Depleted PSII Membranes from Wheat versus Spinach.

Analyses were made of the effects of extraction of the 17,24 kilodalton extrinsic proteins from spinach versus wheat photosystem II (PSII) membranes on Ca abundance and O(2) evolution capacity determined in the absence and presence of either Cl(-) or Ca(2+). Extraction of these proteins from spinach PSII routinely diminished steady state O(2) evolution by about 70% when assayed in the presence of sufficient Cl(-). Additionally, O(2) evolution of 17,24 kilodalton-less spinach PSII membranes showed about 2-fold more enhancement by Ca(2+) than by Cl(-) during assay. When the same extraction and assay procedures were applied to wheat PSII membranes, we observed, in contrast to 17,24 kilodalton-less spinach PSII, only about 50% inhibition of O(2) evolution and about 2-fold greater enhancement by Cl(-) than by Ca(2+). Irrespective of differences in the magnitude of enhancement of O(2) evolution by Ca(2+)versus Cl(-) in spinach versus wheat, the K(m) values for Cl(-) (about 1.7 millimolar) and Ca(2+) (about 1.5 millimolar) were similar for both type preparations. The abundance of Ca specifically associated with fully functional PSII (about 2 and about 3 Ca/200 chlorophyll for spinach and wheat, respectively) was diminished to about 1 per 200 chlorophyll upon 17.24 kilodalton protein depletion. Further treatment of wheat 17,24 kilodalton-less PSII in darkness with 2 molar NaCl/1 millimolar ethyleneglycol-bis(beta-aminoethyl ether)-N,N'-tetraacetic acid/20 micromolar A23187(2) made O(2) evolution highly dependent on Ca(2+) addition, much like the 17,24 kilodalton-less spinach PSII. Analyses of this Ca(2+) effect on O(2) evolution revealed both high (K(m) about 65 micromolar) and low (K(m) about 1.5 millimolar) affinity Ca(2+) sites in wheat 17,24 kilodalton-less PSII. The results suggest that during 17,24 kilodalton extraction by NaCl, spinach PSII is more susceptible than wheat PSII to loss of high affinity Ca and irreversible inhibition of O(2) evolution.

Studies on the Photoactivation of the Water-Oxidizing Enzyme: II. Characterization of Weak Light Photoinhibition of PSII and Its Light-Induced Recovery.

Inactivation of the water splitting enzyme complex in leaves or isolated chloroplasts results in increased susceptibility of photosystem II (PSII) to damage by light. Photoinhibition under this condition occurs in very weak light. The site of damage is exclusive of the water splitting complex yet still on the oxidizing side of PSII, as the Q(B) locus is unaffected while photoreduction of silicomolybdate is inhibited. The kinetics of loss in PSII activity are more complex than apparent first-order, and the quantum efficiency is low. We observe no evidence of deletion from thylakoid membranes of any PSII polypeptide as a consequence of photoinhibition, although recovery from the photoinhibition is dependent upon both light and 70S protein synthesis. Enhanced synthesis of two proteins occurs during recovery, only one of which (D2) appears to be causally related to the recovery. We present a model which describes the relationship of weak light photoinhibition and its recovery to photoactivation of the S-state water oxidizing complex.

Studies on the photoactivation of the water-oxidizing enzyme : I. Processes limiting photoactivation in hydroxylamine-extracted leaf segments.

In weak yet optimal light intensity, complete photoactivation of the water-oxidizing enzyme in NH(2)OH-extracted wheat (Triticum aestivum, var Oasis) leaf segments could be obtained only after long dark preincubation. Photoactivation was not affected by ethylenediaminetetraacetate or inhibitors of photophosphorylation and protein synthesis, but was partially inhibited by a divalent cation ionophore. Complete photoactivation required ligation of approximately 4 Mn by the water oxidizing enzyme.WITHOUT DARK PREINCUBATION, PHOTOSYSTEM II (PSII) WAS SUSCEPTIBLE TO WEAK LIGHT PHOTOINHIBITION RESULTING IN: (a) 50% maximum decrease in photooxidation of artificial electron donors by PSII: (b) increased times for the variable fluorescence rise (with 3-(3,4-dichlorophenyl)-1,1-dimethyl urea): (c) abolishment of photoactivation: and (d) the imposition of sensitivity to inhibitors of photophosphorylation and 70S but not 80S protein synthesis on subsequent light-dependent recovery from photoinhibition and recovery of O(2) evolution. Decrease in susceptibility to photoinhibition and increase in rates of photoactivation resulting from dark preincubations proved closely correlated. Neither protein synthesis nor increases in abundances of thylakoid Mn(2+) and Ca(2+) were required for escape from photoinhibition. However, photoactivation of the wateroxidizing enzyme in NH(2)OH-extracted Chlamydomonas occurred in absence of dark preincubation and protein synthesis. Results are discussed in the context of disassembly/reassembly/resynthesis of specific PSII polypeptides.

Studies on the reconstitution of o(2)-evolution of chloroplasts.

Extraction of spinach (Spinacia oleracea L.) chloroplasts with cholate-asolectin in the absence of Mg(2+) results in the rapid and selective inactivation of O(2) evolution and a partial (30 to 40%) loss of photosystem II (PSII) donor activity without extraction of thylakoid bound Mn ( approximately 5 to 6 Mn per 400 Chlorophyll). Inclusion of ethylene glycol in the extractions inhibits loss of O(2) evolution and results in quantitative and qualitative differences in proteins solubilized but does not significantly inhibit the partial loss of PSII donor activity. Similarly, in two stage experiments (extraction followed by addition of organic solvent and solubilized thylakoid protein), O(2) evolution (V and V(max)) of extracted chloroplasts is enhanced approximately 2.5- to 8-fold. However, PSII donor activity remains unaffected. This reversal of cholate inactivation of O(2) evolution can be induced by solvents including ethanol, methanol, 2-propanol, and dimethyl sulfoxide. Such enhancements of O(2) evolution specifically required cholate-solubilized proteins, which are insensitive to NH(2)OH and are only moderately heat-labile. NH(2)OH extraction of chloroplasts prior to cholate-asolectin extraction abolishes reconstitutability of O(2) evolution. Thus, the protein(s) affecting reconstitution is unlike those of the O(2).Mn enzyme. The specific activity of the protein fraction effecting reconstitution of O(2) evolution is greatest in fractions depleted of the reported Mn-containing, 65-kilodalton, and the Fe-heme, 232-kilodalton (58-kilodalton monomer), proteins. Divalent ( approximately 3 millimolar) and monovalent ( approximately 30 millimolar) cations do not affect reconstitution of PSII donor activity but do affect reconstitution of O(2) evolution by decreasing the protein(s) concentration required for reconstitution of O(2) evolution in nonfractionated, cholate-asolectin extractions. The data indicate a reconstitution of the PSII segment linking the PSII secondary donor(s) to O(2)-evolving centers.

Flash Inactivation of Oxygen Evolution: IDENTIFICATION OF S(2) AS THE TARGET OF INACTIVATION BY TRIS.

Brief saturating light flashes were used to probe the mechanism of inactivation of O(2) evolution by Tris in chloroplasts. Maximum inactivation with a single flash and an oscillation with period of four on subsequent flashes was observed. Analyses of the oscillations suggested that only the charge-collecting O(2)-evolving catalyst of photosystem II (S(2)-state) was a target of inactivation by Tris. This conclusion was supported by the following observations: (a) hydroxylamine preequilibration caused a three-flash delay in the inactivation pattern; (b) the lifetimes of the Tris-inactivable and S(2)-states were similar; and (c) reagents accelerating S(2) deactivation decreased the lifetime of the inactivable state. Inactivation proved to be moderated by F, the precursor of Signal II(s), as shown by a one flash delay with chloroplasts having high abundance of F. Evidence was obtained for cooperativity effects in inactivation and NH(3) was shown to be a competitive inhibitor of the Tris-induced inactivation. S(2)-dependent inactivation was inhibited by glutaraldehyde fixation of chloroplasts, possibly suggesting that inactivation proceeds via conformational changes of the S(2)-state.

Studies on the mechanism of Tris-induced inactivation of oxygen evolution.

A study was made of the inactivation by Tris of O2 evolution in chloroplasts and the subsequent reactivation of O2 evolution. We conclude: 1. At concentrations of Tris sufficient to inhibit O2 evolution directly, a slow rate (t 1/2 approximately 20--25 min) of inactivation occurs; 2. Inactivation is accelerated (t 1/2 approximately 2 min) by weak light absorbed by system II and is rate limited by a dark step with a half-time of about 200 s; 3. Minimally one quantum event within System II is sufficient to inactive 50--70% of the O2 evolving centers; 4. This process is 3-(3,4-dichlorophenyl)-1,1-dimethylurea insensitive but is inhibited by reduced dichlorophenol indophenol and phenazine methosulfate, carbonylcyanide-p-trifluoromethoxyphenylhydrazone, 2-(3-chloro-4-trifluoromethyl)-anilino-3,5-dinitrothiophene and tetraphenylboron; 5. Partial reactivation of inactive O2 evolving centers is affected by the use of the same reagents inhibiting the light induced inactivations; 6. The life-time (t 1/2 approximately 1 to 3 h) of the activable state is correlated with diffusion across thylakoids of the larger manganese pool released from binding sites and remaining in thylakoids following inactivation of O2 evolution.

Effects of Hydroxylamine on Photosystem II: II. Photoreversal of the NH(2)OH Destruction of O(2) Evolution.

The inactivation of O(2)-evolving centers by NH(2)OH extraction was shown to be reversible. This reversal required light and manganese. This light-induced restoration of active O(2)-evolving centers was analyzed using three green algae and the blue-green alga, Anacystis nidulans. The following results were obtained: [List: see text].

Effects of Hydroxylamine on Photosystem II: I. Factors Affecting the Decay of O(2) Evolution.

Illumination of chloroplasts in the presence of NH(2)OH (2 mm) leads to the destruction of all system II activities without affecting system I activity. The system II primary charge separation remains intact when incubated with this agent in the dark with release of one of the system II Mn pools and simultaneous destruction of O(2) evolving capacity. The size of the Mn pool associated with the O(2) evolving center is calculated to be 4 Mn/O(2)-evolving center.We observed the following properties of the hydroxylamine-induced destruction of O(2) centers in darkness: [List: see text]Chloroplasts from summer greenhouse spinach (4-5 Mn/400 chl(total) or 50 to 60% of the Mn pool of the O(2)-evolving center) showed high quantum requirement for O(2) evolution (5-6 hv/equivalent) yet photooxidized NH(2)OH with low quantum requirement ( approximately 2 hv/equivalent).

Photoreaction of manganese catalyst in photosynthetic oxygen evolution.

The formation of active O(2) evolving centers following addition of Mn(2+) to Mn deficient Anacystis nidulans cells yielded an estimate of 6 to 12 Mn atoms associated with each O(2) evolving reaction center. Restoration of activity upon addition of Mn ions is affected in 3 ways: (1) Stimulation of the uptake of exogenous Mn into the cells-this uptake occurs in darkness, but is enhanced 5 to 10 fold by light; a high concentration of DCMU (1 x 10(-5)m) decreases this light enhanced influx no more than 50 to 75%; (2) Photoreactivation of the O(2) evolving centers, after excess Mn has been accumulated in the cells essentially no increase in Hill activity is observed unless the cells are illuminated. This photoreactivation is fully inhibited by 10(-6)m DCMU and partially by benzoquinone. The Q(10) of photoreactivation proper is close to 1; (3) Photoinhibition of the activation-photoreactivation occurs most effectively in weak intensities (< one-fiftieth photosynthetic saturation in normal cells). Apparently at higher intensities an inhibitory photoprocess is overriding. This inhibition proved reversible. The photoreactivation leads to new stable O(2) evolving centers as evidenced by an increase in the rate at saturating intensity, quantum yield, and the O(2) gush.