Plastoquinones are effectively reduced by ferredoxin:NADP+ oxidoreductase in the presence of sodium cholate micelles. Significance for cyclic electron transport and chlororespiration

The effect of sodium cholate and other detergents (Triton X-100, sodium dodecyl sulphate, octyl glucoside, myristyltrimethylammonium bromide) on the reduction of plastoquinones (PQ) with a different length of the side-chain by spinach ferredoxin:NADP(+) oxidoreductase (FNR) in the presence of NADPH has been studied. Both NADPH oxidation and oxygen uptake due to plastosemiquinone autoxidation were highly stimulated only in the presence of sodium cholate among the used detergents. Sodium cholate at the concentration of 20 mM was found to be the most effective on both PQ-4 and PQ-9-mediated oxygen uptake. The FNR-dependent reduction of plastoquinones incorporated into sodium cholate micelles was stimulated by spinach ferredoxin but inhibited by Mg(2+) ions. It was concluded that the structure of sodium cholate micelles facilitates contact of plastoquinone molecules with the enzyme and creates favourable conditions for the reaction similar to those found in thylakoid membranes for PQ-9 reduction. The obtained results were discussed in terms of the function of FNR as a ferredoxin:plastoquinone reductase both in cyclic electron transport and chlororespiration.     

Ferredoxin-Dependent Photosynthetic Electron Transport and Coupled Processes of Energy Storage in Chloroplasts of Wheat Plants Grown under Different Levels of Nitrogen Supply

The rates of electron transfer in the presence of natural cofactors, ferredoxin and NADP, which were added in the amounts catalyzing noncyclic or cyclic electron transfer, were studied in thylakoids isolated from 17-day-old wheat seedlings. Upon excitation of both photosystems (PS) of photosynthesis, the potential rate of NADP reduction in thylakoids isolated from plants grown on nitrogen-free nutrient solution did not differ from that in thylakoids from the control plants. However, the P/2e ratio was significantly lower in thylakoids isolated from nitrogen-deficient plants. On the contrary, in the presence of DCMU, the rate of PSI-driven electron transfer from an artificial donor to NADP was considerably higher in these than in the control thylakoids. In the presence of ferredoxin under anaerobic conditions, the rate of phosphorylation coupled to cyclic electron transport was also significantly higher in thylakoids isolated from nitrogen-deficient plants, than in thylakoids isolated from control plants. Our data show that PSI-driven electron transport and cyclic photophosphorylation are activated in nitrogen-starved wheat plants, at least at the initial stages of starvation.     

Glycolytic Flux and Hexokinase Activities in Anoxic Maize Root Tips Acclimated by Hypoxic Pretreatment

In vitro cyclic electron transport around PSI was studied in thylakoids isolated from barley (Hordeum vulgare L.). Redox poising was obtained by using anaerobic conditions, preillumination, and the addition of 3-(3,4-dichlorophenyl)-1,1-dimethylurea. Postillumination rates of P700+ re-reduction of 1 to 5 electrons s-1 were observed, depending on the conditions. The thylakoids supported two parallel paths of cyclic electron transport that were distinguishable by differences in antimycin sensitivity, saturation characteristics, and substrate specificity. The pathway most sensitive to antimycin was not saturated at ferredoxin concentrations up to 50 [mu]M, whereas the more insensitive pathway was saturated at 5 [mu]M ferredoxin. At the lower concentration of reduced ferredoxin, the antimycin-sensitive rate of P700+ re-reduction was lower than the antimycin-insensitive rate. The lower range of reduced ferredoxin concentrations are closer to in vivo conditions. Flavodoxin is shown to mediate cyclic electron transport. Flavodoxin was less efficient in mediating the antimycin-sensitive pathway but mediated the antimycin-insensitive pathway as efficiently as ferredoxin. Antibodies raised against ferredoxin:NADP+ oxidoreductase had no effect on either pathway for re-reduction of P700+. However, the ferredoxin: NADP+ oxidoreductase inhibitor 2[prime]-monophosphoadenosine-5[prime]-diphosphoribose was able to inhibit the antimycin-sensitive as well as the antimycin-insensitive pathway.     

Participation of ferredoxin in oxygen reduction in a photosynthetic electron-transport chain

Oxygen reduction in a photosynthetic electron-transport chain (PETC) was studied in isolated pea thylakoids in the presence of either ferredoxin, or ferredoxin + NADP+, or cytochrome c. The contribution of the electron flow through ferredoxin to the total oxygen reduction was evaluated by comparing the rate of oxygen reduction and the rate of oxidation of reduced ferredoxin in the light. It was found that at ferredoxin concentrations optimal for NADP+ reduction, 30-50% of electrons transferred to oxygen went through ferredoxin both in the absence and presence of NADP+. However, the absolute rate of oxygen reduction by membrane components of PETC in the presence of NADP+ was 3-4 times less than that in the presence of ferredoxin alone and close to the rate of oxygen reduction in the presence of cytochrome c. It was assumed that a Photosystem I component, whose role in this process depends on the rate of electron outflow from terminal acceptors of this photosystem, participates in oxygen reduction, and this component is phylloquinone.     

Pterin-protein complex and ferredoxin as possible components of a molecular complex in photosystem I]

Properties are studied of pterin-protein complex (PPC), possible precursor of ferredoxin in electrone transport chain PS-I. PPC is shown to form chelate complex with Fe ions which are capable to activate the process of light reduction of NADP+ by chloroplasts in the absence of ferrodoxin. A possibility of dark reduction of NADP+ with the participation of light pre-activated PPC is found. Two-step scheme of electrone transport from P700 to ferredoxin is supposed. The data obtained indicate that one of sites of electrone transport into interior part of tilakoid is located on the reduction side of PS-I. An additional possibility of NADP+ reduction with the participation of PPC-Fe in vivo is suggested.     

Ferredoxin-NADP+ reductase. Kinetics of electron transfer, transient intermediates, and catalytic activities studied by flash-absorption spectroscopy with isolated photosystem I and ferredoxin

The electron transfer cascade from photosystem I to NADP+ was studied at physiological pH by flash-absorption spectroscopy in a Synechocystis PCC6803 reconstituted system comprised of purified photosystem I, ferredoxin, and ferredoxin-NADP+ reductase. Experiments were conducted with a 34-kDa ferredoxin-NADP+ reductase homologous to the chloroplast enzyme and a 38-kDa N-terminal extended form. Small differences in kinetic and catalytic properties were found for these two forms, although the largest one has a 3-fold decreased affinity for ferredoxin. The dissociation rate of reduced ferredoxin from photosystem I (800 s(-1)) and the redox potential of the first reduction of ferredoxin-NADP+ reductase (-380 mV) were determined. In the absence of NADP+, differential absorption spectra support the existence of a high affinity complex between oxidized ferredoxin and semireduced ferredoxin-NADP+ reductase. An effective rate of 140-170 s(-1) was also measured for the second reduction of ferredoxin-NADP+ reductase, this process having a rate constant similar to that of the first reduction. In the presence of NADP+, the second-order rate constant for the first reduction of ferredoxin-NADP+ reductase was 20% slower than in its absence, in line with the existence of ternary complexes (ferredoxin-NADP+ reductase)-NADP+-ferredoxin. A single catalytic turnover was monitored, with 50% NADP+ being reduced in 8-10 ms using 1.6 microM photosystem I. In conditions of multiple turnover, we determined initial rates of 360-410 electrons per s and per ferredox-in-NADP+ reductase for the reoxidation of 3.5 microM photoreduced ferredoxin. Identical rates were found with photosystem I lacking the PsaE subunit and wild type photosystem I. This suggests that, in contrast with previous proposals, the PsaE subunit is not involved in NADP+ photoreduction.     

Generation of superoxide anion and hydrogen peroxide during redox cycling of 5-(4-nitrophenyl)-penta-2,4-dienal by mammalian microsomes and enzymes

5-(4-Nitrophenyl)penta-2,4-dienal (NPPD) stimulated NADPH-supported oxygen consumption by rat liver microsomes in a concentration-dependent manner. The NPPD stimulation of O2 uptake was not inhibited by metyrapone and was decreased in the presence of NADP+ and p-hydroxymercuribenzoate. These observations suggest that the NPPD initial reduction step is mediated by NADPH-cytochrome P-450 reductase and not by cytochrome P-450. Spin-trapping studies using 5,5-dimethyl-1-pyrroline N-oxide (DMPO) revealed the formation of superoxide anion upon incubation of NPPD, NADPH, DMPO and rat liver microsomes. Hydrogen peroxide generation was also detected in these incubations, thus confirming redox cycling of NPPD under aerobic conditions. NPPD stimulated oxygen consumption, superoxide anion formation and hydrogen peroxide generation by rat kidney, testes and brain microsomes. Other enzymes capable of nitroreduction (NADH dehydrogenase, xanthine oxidase, glutathione reductase, and NADP+ ferredoxin oxidoreductase) were also found to stimulate redox cycling of NPPD. The ability of NPPD to induce superoxide anion and hydrogen peroxide formation might play a role in its reported mutagenicity.     

Molecular structure and regulation of phototropin kinase by blue light

The ferredoxin:NADP+ oxidoreductase (FNR) is a plant enzyme, catalyzing the last step of photosynthetic linear electron transport, and involved also in cyclic electron transport around photosystem I. In this study we present the first evidence of FNR (isolated from spinach and from wheat) interaction directly with a model membrane without the mediation of any additional protein. The monomolecular layer technique measurements showed a significant increase in surface pressure after the injection of enzyme solution beneath a monolayer consisting of chloroplast lipids: monogalactosyldiacylglycerol or digalactosyldiacylglycerol. An ATR FTIR study revealed also the presence of FNR in a bilayer composed of these lipids. The secondary structure of the protein was significantly impaired by lipids, as with a pH-induced shift. The stabilization of FNR in the presence of lipids leads to an increase in the rate of NADPH-dependent reduction of dibromothymoquinone catalyzed by the enzyme. The biological significance of FNR-membrane interaction is discussed.     

Proton to electron stoichiometry in electron transport of spinach thylakoids

According to the concept of the Q-cycle, the H+/e- ratio of the electron transport chain of thylakoids can be raised from 2 to 3 by means of the rereduction of plastoquinone across the cytochrome b6f complex. In order to investigate the H+/e- ratio we compared stationary rates of electron transport and proton translocation in spinach thylakoids both in the presence of the artificial electron acceptor ferricyanide and in the presence of the natural acceptor system ferredoxin+NADP. The results may be summarised as follows: (1) a variability of the H+/e- ratio occurs with either acceptor. H+/e- ratios of 3 (or even higher in the case of the natural acceptor system, see below) are decreased towards 2 if strong light intensity and low membrane permeability are employed. Mechanistically this could be explained by proton channels connecting the plastoquinol binding site alternatively to the lumenal or stromal side of the cytochrome b6f complex, giving rise to a proton slip reaction at high transmembrane DeltapH. In this slip reaction protons are deposited on the stromal instead of the lumenal side. In addition to the pH effect there seems to be a contribution of the redox state of the plastoquinone pool to the control of proton translocation; switching over to stromal proton deposition is favoured when the reduced state of plastoquinone becomes dominant. (2) In the presence of NADP a competition of both NADP and oxygen for the electrons supplied by photosystem I takes place, inducing a general increase of the H+/e- ratios above the values obtained with ferricyanide. The implications with respect to the adjustment of a proper ATP/NADPH ratio for CO2 reduction are discussed.     

Control of energy dissipation and photochemical activity in photosystem I by NADP-dependent reversible conformational changes

Addition of NADP(+) to thylakoid membranes or isolated photosystem I (PSI) submembrane fractions quenched chlorophyll fluorescence by up to 40% at low or room temperature. This quenching was reversed by NADPH. Similar quenching was also observed with the addition of heparin or thenoyltrifluoroacetone (TTFA), inhibitors that bind ferredoxin:NADP(+) reductase (FNR) and prevent reduction of NADP(+). The NADP(+)-induced quenching coincided with a reversible conformational change of the secondary protein structure in the PSI submembrane fractions where 20% of the alpha-helix conformations were transformed mainly into beta-sheet-like structures. Further, P700 photooxidation was retarded due to this conformational change, and about 25% of the centers could not be photooxidized, these changes being also reversible with addition of NADPH. The above modifications in the presence of NADP(+) also increased photodamage processes under strong illumination, and NADPH protected it. Conformational modification of FNR upon binding of NADP(+) or NADPH is proposed to trigger the macromolecular changes in a larger part of the protein complex of PSI. The conformational changes must increase the intermolecular distances and change the mutual orientation between the various cofactors in the PSI complex. This new control mechanism of energy dissipation and photochemical activity by NADP(+)/NADPH is proposed to increase the turnover rate of PSI under conditions when both linear and cyclic electron transport activities must be supported.     

Electron transfer by ferredoxin:NADP+ reductase. Rapid-reaction evidence for participation of a ternary complex

Rapid reaction studies presented herein show that ferredoxin:NADP+ oxidoreductase (FNR, EC 1.18.1.2) catalyzes electron transfer from spinach ferredoxin (Fd) to NADP+ via a ternary complex, Fd X FNR X NADP+. In the absence of NADP+, reduction of ferredoxin:NADP+ reductase by Fd was much slower than the catalytic rate: 37-80 s-1 versus at least 445 e-s-1; dissociation of oxidized spinach ferredoxin (Fdox) from one-electron reduced ferredoxin:NADP+ reductase (FNRsq) limited the reduction of FNR. This confirms the steady-state kinetic analysis of Masaki et al. (Masaki, R., Yoshikaya, S., and Matsubara, H. (1982) Biochim. Biophys. Acta 700, 101-109). Occupation of the NADP+ binding site of FNR by NADP+ or by 2',5'-ADP (a nonreducible NADP+ analogue) greatly increased the rate of electron transfer from Fd to FNR, releiving inhibition by Fdox. NADP+ (and 2',5'-ADP) probably facilitate the dissociation of Fdox; equilibrium studies have shown that nucleotide binding decreases the association of Fd with FNR (Batie, C. J. (1983) Ph.D. dissertation, Duke University; Batie, C. J., and Kamin, H. (1982) in Flavins and Flavoproteins VII (Massey, V., and Williams, C. H., Jr., eds) pp. 679-683, Elsevier, New York; Batie, C.J., and Kamin, H. (1982) Fed. Proc. 41, 888; and Batie, C.J., and Kamin, H. (1984) J. Biol. Chem. 259, 8832-8839). Premixing Fd with FNR was found to inhibit the reaction of the flavoprotein with NADP+ and with NADPH; thus, substrate binding may be ordered, NADP+ first, then Fd. FNRred and NADP+ very rapidly formed an FNRred X NADP+ complex with flavin to nicotinamide charge transfer bands. The Fdred X NADP+ complex then relaxed to an equilibrium species; the spectrum indicated a predominance of FNRox X NADPH charge-transfer complex. However, charge-transfer species were not observed during turnover; thus, their participation in catalysis of electron transfer from Fd to NADP+ remains uncertain. The catalytic rate of Fd to NADP+ electron transfer, as well as the rates of electron transfer from Fd to FNR, and from FNR to NADP+ were decreased when the reactants were in D2O; diaphorase activity was unaffected by solvent. On the basis of the data presented, a scheme for the catalytic mechanism of catalysis by FNR is presented.     

Characterization of factors affecting the activity of photosystem I cyclic electron transport in chloroplasts

PSI cyclic electron transport is essential for photosynthesis and photoprotection. In higher plants, the antimycin A-sensitive pathway is the main route of electrons in PSI cyclic electron transport. Although a small thylakoid protein, PGR5 (PROTON GRADIENT REGULATION 5), is essential for this pathway, its function is still unclear, and there are numerous debates on the rate of electron transport in vivo and its regulation. To assess how PGR5-dependent PSI cyclic electron transport is regulated in vivo, we characterized its activity in ruptured chloroplasts isolated from Arabidopsis thaliana. The activity of ferredoxin (Fd)-dependent plastoquinone (PQ) reduction in the dark is impaired in the pgr5 mutant. Alkalinization of the reaction medium enhanced the activity of Fd-dependent PQ reduction in the wild type. Even weak actinic light (AL) illumination also markedly activated PGR5-dependent PSI cyclic electron transport in ruptured chloroplasts. Even in the presence of linear electron transport [11 mumol O2 (mg Chl)(-1) h(-1)], PGR5-dependent PSI electron transport was detected as a difference in Chl fluorescence levels in ruptured chloroplasts. In the wild type, PGR5-dependent PSI cyclic electron transport competed with NADP+ photoreduction. These results suggest that the rate of PGR5-dependent PSI cyclic electron transport is high enough to balance the production ratio of ATP and NADPH during steady-state photosynthesis, consistently with the pgr5 mutant phenotype. Our results also suggest that the activity of PGR5-dependent PSI cyclic electron transport is regulated by the redox state of the NADPH pool.     

Participation of ferredoxin in oxygen reduction by the photosynthetic electron transport chain

Oxygen reduction by isolated pea thylakoids was studied in the presence of ferredoxin (Fd), Fd + NADP, and cytochrome c. At Fd concentrations optimal for NADP reduction, it contributed 30–50% of the reducing equivalents (as deduced by comparing the rates of oxygen reduction and light oxidation of reduced Fd). The oxygen reduction rate in the presence of Fd + NADP was 3–4 times lower than with Fd alone, and comparable to that with cyt c. It is supposed that the process involves a photosystem I component whose reaction with oxygen depends on the rate of electron efflux from the PS I terminal acceptors, and that this component is phylloquinone.     

Regulation of Cyclic Photophosphorylation during Ferredoxin-Mediated Electron Transport : Effect of DCMU and the NADPH/NADP Ratio

Addition of ferredoxin to isolated thylakoid membranes reconstitutes electron transport from water to NADP and to O₂ (the Mehler reaction). This electron flow is coupled to ATP synthesis, and both cyclic and noncyclic electron transport drive photophosphorylation. Under conditions where the NADPH/NADP⁺ ratio is varied, the amount of ATP synthesis due to cyclic activity is also varied, as is the amount of cyclic activity which is sensitive to antimycin A. Partial inhibition of photosystem II activity with DCMU (which affects reduction of electron carriers of the interphotosystem chain) also affects the level of cyclic activity. The results of these experiments indicate that two modes of cyclic electron transfer activity, which differ in their antimycin A sensitivity, can operate in the thylakoid membrane. Regulation of these activities can occur at the level of ferredoxin and is governed by the NADPH/NADP ratio.     

The enzymatic one-electron reduction of porphyrins to their anion free radicals

The anaerobic enzymatic one-electron reduction of uroporphyrin I (in the absence of light) by the ferredoxin/ferredoxin:NADP+ oxidoreductase system was investigated using NADPH as the source of reducing equivalents. The porphyrin anion free radical metabolite formed by one-electron reduction of the parent molecule was detected with ESR spectroscopy. The ESR spectrum exhibited a singlet (g = 2.0021) with a 5.4-G peak-to-peak linewidth. The reduction process was also investigated under aerobic conditions. The reduction of molecular oxygen to superoxide anion radical by the porphyrin anion radical was demonstrated by using the ESR technique of spin trapping. The ESR spectra of the spin-trapped oxygen-derived radicals were superoxide dismutase-sensitive and catalase-insensitive, supporting the assignment of the trapped radical to the superoxide anion radical. These aerobic experiments demonstrate electron transfer from the porphyrin anion radical to molecular oxygen. The anaerobic reduction of Photofrin II by hepatic microsomes and the ferredoxin/ferredoxin:NADP+ oxidoreductase system to a porphyrin anion radical was also investigated. Free radical formation by ferredoxin: NADP+ oxidoreductase is totally dependent upon ferredoxin. The ESR spectrum of this porphyrin free radical also exhibited a singlet (g = 2.0026) with a 15-G peak-to-peak linewidth.     

Differential effects of DBMIB and DNP-INT on the kinetic phases of P-700 reduction

In the presence of ferredoxin and NADP, DBMIB abolishes the fast-relaxing portion of P-700 together with the reduction of NADP. The slow-relaxing portion is inhibited at much higher concentrations. Qualitatively similar results have been observed with DNP-INT. However, its action appears to be a light-dependent process. The slow, cyclic turnover of P-700 in the presence of DCMU, ferredoxin and NADPH is inhibited by DBMIB but only slightly by DNP-INT. The data suggest that the inhibitors act at different sites of the electron-transport system.     

The transition metal-mediated formation of the hydroxyl free radical during the reduction of molecular oxygen by ferredoxin-ferredoxin:NADP+ oxidoreductase

The NADPH-supported enzymatic reduction of molecular oxygen by ferredoxin-ferredoxin:NADP+ oxidoreductase was investigated. The ESR spin trapping technique was employed to identify the free radical metabolites of oxygen. The spin trap 5,5-dimethyl-1-pyrroline N-oxide (DMPO) was used to trap and identify the oxygen-derived free radicals. [17O]Oxygen was employed to demonstrate that the oxygen-centered radicals arose from molecular oxygen. From the data, the following scheme is proposed: (Formula:see text). The formation of the free hydroxyl radical during the reduction of oxygen was demonstrated with quantitative competition experiments. The hydroxyl radical abstracted hydrogen from ethanol or formate, and the resulting scavenger-derived free radical was trapped with known rate constants. If H2O2 was added to the enzymatic reaction, a stimulation of the production of the hydroxyl radical was obtained. This stimulation was manifested in both the concentration and the rate of formation of the DMPO/hydroxyl radical adduct. Catalase was shown to inhibit formation of the hydroxyl radical adduct, further supporting the formation of hydrogen peroxide as an intermediate during the reduction of oxygen. All three components, ferredoxin, ferredoxin:NADP+ oxidoreductase, and NADPH, were required for reduction. Ferredoxin:NADP+ oxidoreductase reduces ferredoxin, which in turn is responsible for the reduction of oxygen to hydrogen peroxide and ultimately the hydroxyl radical. The effect of transition metal chelators on the DMPO/hydroxyl radical adduct concentration suggests that the reduction of chelated iron by ferredoxin is responsible for the reduction of hydrogen peroxide to the hydroxyl radical via Fenton-type chemistry.     

Photochemical system for regenerating NADPH from NADP with use of immobilized chloroplasts

Spinach chloroplasts were immobilized in 2% agar gel. Crude ferredoxin and NADP–ferredoxin oxidoreductase isolated from spinach were used as electron carriers. The activity of the NADP reduction by immobilized chloroplasts increased with increasing ferredoxin concentration and the maximum activity was obtained at 8μM ferredoxin. The saturation of NADP reduction was observed at a light intensity of over 1000 lx. The optimum pH and temperature of NADP reduction were 8 and 25°C, respectively. The reduced NADP in a reaction medium increased linearly with increasing reaction time under illumination. NADP was continuously reduced for 2 hr with a hollow-fiber reactor containing immobilized chloroplasts. NADPH and NADP were separated with a hollow-fiber dialyzer from ferredoxin and NADP–ferredoxin oxidoreductase, which were reused. The conversion ratio of NADP to NADPH was from 40 to 80%.     

Hydrogen peroxide synthesis in isolated spinach chloroplast lamellae : an analysis of the mehler reaction in the presence of NADP reduction and ATP formation

Light-dependent O₂ reduction concomitant with O₂ evolution, ATP formation, and NADP reduction were determined in isolated spinach (Spinacia oleracea L. var. America) chloroplast lamellae fortified with NADP and ferredoxin. These reactions were investigated in the presence or absence of catalase, providing a tool to estimate the reduction of O₂ to H₂O₂ (Mehler reaction) concomitant with NADP reduction. In the presence of 250 micromolar O₂, O₂ photoreduction, simultaneous with NADP photoreduction, was dependent upon light intensity, ferredoxin, Mn²⁺, NADP, and the extent of coupling of phosphorylation to electron flow.

In the presence of an uncoupling concentration of NH₄⁺, saturating light intensity (>500 watts/square meter), saturating ferredoxin (10 micromolarity) rate-limiting to saturating NADP (0.2-0.9 millimolarity), and Mn²⁺ (50-1000 micromolarity), the maxium rates of O₂ reduction were 13-25 micromoles/milligram chlorophyll per hour, while concomitant rates of O₂ evolution and NADP reduction were 69 to 96 and 134 to 192 micromoles/milligram chlorophyll per hour, respectively. Catalase did not affect the rate of NADPH or ATP formation but decreased the NADPH:O₂ ratios from 2.3-2.8 to 1.9-2.1 in the presence of rate-limiting as well as saturating concentrations of NADP.

Photosynthetic electron flow at a rate of 31 micromoles O₂ evolved/milligram chlorophyll per hour was coupled to the synthesis of 91 micromoles ATP/milligram chlorophyll per hour, while the concomitant rate of O₂ reduction was 0.6 micromoles/milligram chlorophyll per hour and was calculated to be associated with an apparent ATP formation of only 2 micromoles/milligram chlorophyll per hour. Thus, electron flow from H₂O to O₂ did not result in ATP formation significantly above that produced during NADP reduction.