Ferredoxin-activated Fructose Diphosphatase of Spinach Chloroplasts

The hydrolytic cleavage of Pi from fructose diphosphate by the fructose diphosphatase (FDPase) system of spinach chloroplasts was found to require (a) an alkaline FDPase component insensitive to inhibition by AMP or fructose diphosphate; (b) reduced ferredoxin; (c) a protein factor; and (d) Mg++. Reduced spinach ferredoxin could be replaced by reduced methyl viologen or dithiothreitol but not by reduced bacterial ferredoxin. Activation of the alkaline FDPase component by the nonphysiological substitutes required the protein factor. The alkaline FDPase component in isolated intact chloroplasts was also activated by light. The protein factor component of the FDPase system has been partly purified and found to have a molecular weight of 40,000. The alkaline FDPase component, purified to homogeneity, had a molecular weight of 145,000 and contained 2 % carbohydrate. Except for a remarkably high half-cystine content (16% by weight as compared to 2 % for mammalian FDPases) which reflects an apparent abundance of cystine, the amino acid composition of the chloroplast alkaline FDPase component resembled that of the mammalian FDPases. In the absence of reduced ferredoxin and protein factor, the chloroplast alkaline FDPase component can be activated by high concentrations of Mg+f. Like liver alkaline FDPase, the chloroplast enzyme component is also activated by cystamine or 5,5’-dithiobis(Z-nitrobenozic acid) (DTNB). Ferredoxin-activated FDPase was inhibited completely by 5 x 10V4 M EDTA but was insensitive to KCN, cuprizone, and diethyldithiocarbamate (all at 1 X lop3 M). The activity of chloroplast FDPase during photosynthesis appears to be controlled photochemically through reduced ferredoxin. What, if any, physiological role can be ascribed to the activation of the enzyme by high concentrations of Mg++ and disulfide reagents, independently of reduced ferredoxin, is not clear.


SUMMARY
The hydrolytic cleavage of Pi from fructose diphosphate by the fructose diphosphatase (FDPase) system of spinach chloroplasts was found to require (a) an alkaline FDPase component insensitive to inhibition by AMP or fructose diphosphate; (b) reduced ferredoxin; (c) a protein factor; and (d) Mg++.
Reduced spinach ferredoxin could be replaced by reduced methyl viologen or dithiothreitol but not by reduced bacterial ferredoxin.
Activation of the alkaline FDPase component by the nonphysiological substitutes required the protein factor.
The alkaline FDPase component in isolated intact chloroplasts was also activated by light.
The protein factor component of the FDPase system has been partly purified and found to have a molecular weight of 40,000.
The alkaline FDPase component, purified to homogeneity, had a molecular weight of 145,000 and contained 2 % carbohydrate.
Except for a remarkably high half-cystine content (16% by weight as compared to 2 % for mammalian FDPases) which reflects an apparent abundance of cystine, the amino acid composition of the chloroplast alkaline FDPase component resembled that of the mammalian FDPases. In the absence of reduced ferredoxin and protein factor, the chloroplast alkaline FDPase component can be activated by high concentrations of Mg+f. Like liver alkaline FDPase, the chloroplast enzyme component is also activated by cystamine or 5,5'-dithiobis(Z-nitrobenozic acid) (DTNB). Ferredoxin-activated FDPase was inhibited completely by 5 x 10V4 M EDTA but was insensitive to KCN, cuprizone, and diethyldithiocarbamate (all at 1 X lop3 M). The activity of chloroplast FDPase during photosynthesis appears to be controlled photochemically through reduced ferredoxin.
What, if any, physiological role can be ascribed to the activation of the enzyme by high concentrations of Mg++ and disulfide reagents, independently of reduced ferredoxin, is not clear.
* This work was aided by Grant GB-23796 from the National Science Foundation.
Ferredoxins are iron-sulfur proteins noted for t,heir strongly electronegative oxidation-reduction potentials.
Ferredoxins are widely distributed in nature and function as electron carriers in all photosynthetic cells and in certain anaerobic nonphotosynthetic bacteria (1). Work with isolated chloroplasts has shown that ferredoxin is the endogenous catalyst of cyclic and noncyclic photophosphorylation-the two processes that provide the assimilatory power, made up of ATP and NADPI-I, needed for carbon dioxide assimilation (1,2).
An additional role of ferredoxin came to light when Buchanan,Kalberer,and Arnon (3) found that photochemically reduced ferredoxin activated the hydrolytic cleavage of fructose 1,6diphosphate to fructose 6-phosphate and Pi by a fructose diphosphatase present in an aqueous extract of chloroplasts (Equation 1). reduced ferredoxin Fructose 1,6-diphosphate + H20 > chloroplast extract (1) fructose g-phosphate + Pi Since chloroplasts reduce ferredoxin only in the light, Buchanan et al. (3) suggested that a requirement of reduced ferredoxin for activation of the chloroplast FDPaser (4-7)-a key enzyme of the photosynthetic reductive pentose phosphate cycle @)-could serve as a light-actuated mechanism for regulation of carbon dioxide assimilation in photosynthesis.
Kinetic evidence for a light activation of FDPase in whole algal cells was reported by Bassham,Kirk,and Jensen (9) and Pedersen,Kirk,and Bassham (10).
The fructose 1,6-diphosphatase reaction in chloroplasts was previously considered to require a single enzyme, alkaline FDPase, analogous to mammalian alkaline FDPase-a "pacemaker" enzyme in gluconeogenesis (II). The present experiments show that the chloroplast FDPase is, however, more complex. It consists not only of the alkaline FDPase component but also of a protein factor and ferredoxin.
Ferredoxin, in reduced form, and the protein factor are needed to activat,e the alkaline FDPa.se component (Equation 2)  The alkaline FDPase component was earlier extensively purified from Fpinach leaves by Racker and Schroeder (12) and Preiss,Biggs,and Greenberg (13); its crystallization has recently been reported by El-Badry and Bassham (14).
The chloroplast alkaline FDPase component resembles the alkaline FDPase from mammalian cells (15) in showing stimulation by cyst,amine and DTNB.
However, unlike the mammalian FDPase (11, 16-al), the alkaline FDPase component of t,he chloroplast FDPase system is not inhibited by AMP or fructose diphosphate (13).
A preliminary report of some of these findings has been published (22).

METHODS
Rabbit muscle aldolase, liver catalase, pig heart malic dehydrogennsc, horse heart cytochrome c, ovalbumin, and bovine serum albumin were purchased from Sigma. Homogeneous alkaline FDl'ase from rabbit liver and plastocyanin from spinach leaves were gifts of Dr. 13. Horecker (Albert I<Znstcin College of Medicine) and Mr. R. Chain of this Department., respectively.
A sample of homogeneous alkaline FDPase from spinach leaves was an earlier gift of Dr. J. Preiss (University of California at Davis). Protoil was estimated by a modified phenol reagent procedure (23). Chlorophyll was determined as described by Arnon (24). The carbohydrate content of chloroplast FDl'ase was determined by a naphthalene procedure (25). Zinc was determined by atomic absorption spectrophotomet.rg.
Tryptophan was determined spectrophotometrically (29) and by the thioglgcolate procedure (30). The amino acid content of FDPxsc was based on an average of two analyses; the composition shown was calculated on the basis of 12 histidine residues per mole of enzyme.
Fcrredoxin with A.izs:A276 of 0.46 (or higher) was isolated from spinach leaves by the method of Tagawa and Arnon (as given in the review by Buchanan and rirnon (31)). Intact chloroplasts were isolated in isotonic sucrose (not sorbitol which interfered with l'i determination) as described by Kalberer, Buchanan,and Arnon (32). Chloroplast fragments (l'&) were prepared from whole chloroplast,s (isolated either in isotonic NaCl as described by Whatley and Arnon (33) or in isotonic sucrose (32)) by suspending in a 1 to 10 dilution of the preparative solut,ion, collecting by centrifugation (5 min, 35,000 x g), and resuspending in the dilute preparative solut.ion.

Assay and PwiJkation
of Alkaline FDI'ase Component of Chloroplast FDPase System For routine assay of the alkaline FDl':~sc component, the reaction mixture contained (in micromoles) aside from the en-zyme: Tris-HCl buffer, ~1- 1 8.5,100;MgClz,16; sodium EDTA, 0.1; and sodium fructose 1 ,6-diphosphate, 6, in a final volume of 1.0 ml. The assay, carried out in test tubes at room temperature for 10 min, was started by adding fructose diphosphate and stopped by 0.5 ml of 10% trichloroacetic acid. The Pi released was determined calorimetrically by a modified Fiske-SubbaRow procedure (34). One unit of enzyme activity is defined as that amount of enzyme which catalyzes the release of 1 pmole of l'i per min under the assay conditions.
Unless indicated otherwise, all purification steps of the FDPase component given below were carried out ai. 4".
The frozen leaves (in l-kg batches) were blended in 1,400 ml of water for 3 min in a Waring Blendor (gallon size, model CR-5). Ten milliliters of 1 11 K&IP04 (sufficient to give a final pH of 7) were added prior to blending.
The homogenate was filtered through four layers of cheesecloth and the residue was discarded.
The remaining particulate material was sedimented by centrifugation at 13,000 x g for 5 min and was also discarded. pH 4.5 Precipitation-The green supernatant fraction (10,700 ml) was adjusted to pH 4.5 with 1 N formic acid and centrifuged (5 min, 12,000 x g). The yellow supernatant fraction obtained in this step contained the bulk of the protein factor needed for activation of alkaline FDPase by reduced ferredoxin; the fraction was neutralized with 1 M Tris and fractionated as described below. The green precipitate containing the alkaline FDPase component was suspended in a HEPES-EDTA solution (0.025 hI HEPES-NaOH buffer, pH 7.6, containing 2.5 X lop4 M EDTA-Na) and then brought to a volume of 1,000 ml. The pH was adjusted to 6.5 with 0.1 N NH,OH.
The suspension was centrifuged for 10 min at 35,000 x g and the residue was discarded. Final volume of the supernat,ant fraction containing the FDPase was 700 ml. ~lmmonium Sulfate I+actionation-Solid ammonium sulfate was added to the redissolved ~1-1 4.5 precipitate fraction to give 50% saturat'ion.
The solution was centrifuged (10 min, 13,000 x g) and the precipit,ate was discarded.
Ammonium sulfate was added to the supernatant fraction for 90% saturation.
The solution was centrifuged as above, the supernatant fraction was discarded, and the precipitate (containing the bulk of the FDPase activity) was dissolved in 0.03 v Tricine buffer, pH 8.0, to a final volume of 80 ml. This slightly turbid solution was centrifuged (15 min, 35,000 x g) to clarify.
Xephadex G-100 Chromatography-The 50 to 90% ammonium sulfate fraction was applied to a Sephadex G-100 column (5 x 90 cm) equilibrated beforehand and developed with 0.03 RI Tricine, pI-I 8.0. Fractions of 10 ml were collected with a fraction collector.
FDl'ase chromatographed just behind the excluded volume and was eluted before the bulk of applied protein.
The slightly yellow fractions containing FDPase were pooled; and sufficient HEPES-NaOH buffer, pH 7.6, and NaCl were added to give a final concentration of 0.05 &I HEPES, pH 7.6, and 0.2 RI NaCl.
DEAE-Cellulose Chromatography-The combined FDPase fractions were applied to a DEAE-cellulose column ( pH 7.6, and 0.5 M N&I. FDPase was eluted just behind the solvent front; the active fractions were pooled and concentrated by vacuum dialysis. The FDPase component purified by this procedure was homogeneous. The enzyme was relatively stable and could be stored in HEPES-EDTA buffer at -20" for several weeks with little loss of activity. Repeated freezing and thawing caused a slow loss of activity.

Puri$cation of Protein Factor Component of Chloroplast FDPase System
The activity of the protein factor was determined by the increase of Pi release from the fructose diphosphate in the presence of photoreduced ferredoxin and homogeneous FDPase component (cf. Table IV).
The neutralized pH 4.5 supernatant fraction from Step 2 of the alkaline FDPase procedure was used as a source of the protein factor. Acetone, cooled to -2O", was added with constant stirring to a final concentration of 750/,; the solution was left 1 hour at -20" to allow the precipitate to settle. The supernatant fraction was decanted and discarded; the precipitate (containing the protein factor) was collected by centrifugation (3 min, 1000 x g) and suspended in 0.03 M Tris-HCl buffer, pH 8.0. The turbid solution was then dialyzed 24 hours against 0.03 M Tris-HCl buffer, pH 8.0.
Denatured protein was centrifuged off (15 min, 13,000 x g), and solid ammonium sulfate was added to the supernatant fraction to 50% saturation. The heavy precipitate was sedimented by centrifugation as before and discarded. The ammonium sulfate concentration in the supernatant solution was increased to 90% saturation and the precipitate (containing the protein factor) was collected by centrifugation as above and dissolved in 100 ml of 0.02 M Tris-HCl buffer, pH 8.0. The slightly turbid solution was centrifuged (5 min, 35,000 x g) to clarify.
The 50 to 90% ammonium sulfate fraction was applied to a Sephadex G-100 column (5 x 90 cm) equilibrated beforehand and developed with 0.03 M Tris-HCl buffer, pH 8.0. Fractions (13 ml) containing protein factor activity were pooled and stored at -20". The protein factor was stable for several months at -20".

Other Methods
Ultracentrifugation was carried out in a Spinco model E ultracentrifuge. Sedimentation velocity experiments were done in a 12-mm aluminum, l-sector cell with quartz windows. The high speed sedimentation equilibrium experiments were done according to Yphantis (35) with 3-mm liquid columns in a g-channel cell. The time required to reach equilibrium (24 hours) was determined by measuring fringe displacement until it became constant. Fringe and schlieren boundary displacements were measured with a Nikon a-coordinate comparator. Partial specific volume was determined from sedimentation equilibrium centrifugation in DzO and Hz0 as described by Edelstein and Schachman (36) and from amino acid composition as described by Schachman (37).
The molecular weight of chloroplast alkaline FDPase was estimated by gel filtration as described by Andrews (38). A Sephadex G-200 column (1.5 x 80 cm, equilibrated beforehand with a solution containing 0.1 M potassium phosphate buffer, pH 7.3, and 0.1 M NaCl) was calibrated with liver catalase, pig heart malic dehydrogenase, horse heart cytochrome c, ovalbumin, bovine serum albumin, and plastocyanin as standards. Chloroplast alkaline FDPase was applied to the calibrated column and was eluted between aldolase and bovine serum albumin, corresponding to a molecular weight of approximately 130,000.
The procedure of Davis (39) was used in acrylamide gel electrophoresis with 7.5% gels and Tris-glycine, pH 8.3, as electrode buffer. A current of 2 to 3 ma per gel was applied for 2 hours at room temperature. Gels were stained in Coomassie blue (0.5% solution in 12.5% trichloroacetic acid) (40).
The reaction of FDPase with p-mercuribenzoate was followed spectrophotometrically as described by Boyer (41). p-Mercuribenzoate (10u3 M) was added in 5+1 amounts to a cuvette containing 0.3 mg of FDPase in 3 ml of 0.1 M Tris-HCl buffer, pH 8.0. After 5 min, absorbance at 250 nm was measured against a cuvette containing FDPase but no p-mercuribenzoate. Additional incubation did not increase the amount of p-mercuribenzoate reacting with the FDPase.
The reaction of FDPase with DTNB was carried out spectrophotometrically as described by Ellman (42). A cuvette contained 0.3 mg of FDPase in 3 ml of 0.1 M Tris-HCl buffer, pH 7.3 (in the presence or absence of 8 M urea) and 20 ~1 of a 0.4% DTNB solution. Absorbance at 412 nm was read immediately after mixing and again after 12 hours at room temperature, when the reaction was complete.

AND DISCUSSION
Chemical and Physical Properties of Alkaline FDPase Component of Chloroplast FDPase Xystem Alkaline FDPase, protein factor, and ferredoxin are the three protein components of the FDPase system of chloroplasts. Early in this investigation the FDPase and protein factor components were obtained from the aqueous extract of isolated chloroplasts. However, for preparation on a large scale, procedures were devised for purification of the FDPase and protein factor from spinach leaves. Table I shows a summary of the purification and yield of the alkaline FDPase component (from leaves) in each of the purification steps described under "Methods." The freshly purified FDPase component appeared to be homogeneous. The purified protein showed a single peak in the ultracentrifuge (Fig. 1,lcwer patterns in A and B) and in Sephadex G-200 chromatography. The FDPase component traveled in polyacrylamide gel as one main band trailed by diffuse protein material (Fig. 2). This slow moving material, present in all preparations, was probably due to change in the homogeneous enzyme during electrophoresis-an interpretation supported by the observation that under certain conditions (particularly after Issue of October 10, 1971 B. B. Buchanan, P. Schiimzann, and P. P. Kalberer the enzyme was concentrated by vacuum or pressure dialysis) the FDPase component dissociated into subunits.
The subunits formed during concentration dialysis were separated from the parent alkaline FDPase component of the FDPase system by Sephadex G-200 chromatography.
The subunit fraction showed a single peak in the ultracentrifuge (Fig. 1, upper pelteras in A and B) and, based on both gel filtration and ultracentrifugation (see below), had a molecular weight half that of the original enzyme. This finding indicates that the active alkaline FDPase component is composed of two subunits of about equal molecular weight; the isolated subunit fraction had no FDPase activity.
The sedimentation velocity pattern (determined with schlieren optics) of the fresh alkaline FDPase component (lower patterns) and the isolated subunits (upper patlerns) in Fig. 1 were meas- The sediment,ation equilibrium technique showed a pattern typical of a homogeneous preparation and gave a molecular weight of 145,000 for the active undissociated FDPase component (Fig. 3) based on an average partial specific volume of 0.700. (The partial specific volume was determined experiment.ally by sedimentation equilibrium in DzO and Hz0 and found to be 0.697; a value of 0.706 was calculated from amino acid composition.) The molecular weight of the subunit fraction would therefore be 73,000. Values of 130,000 and 139,000 for the active FDPase component were obtained by the Sephadex G-200 chromatography technique and by amino acid analysis. A molecular weight of 145,000 is substantially lower than 195,000 obtained by Preiss (quoted by Preiss and Kosuge (43)) for his alkaline FDPase preparation. Table II shows the amino acid composition of the alkaline FDPase component from chloroplasts in comparison with the enzyme from rabbit muscle and liver. Each of the three proteins is characterized by a high content of glutamate, aspartate, and glycine, a low content of histidine, and the absence of tryptophan. A striking feature of the chloroplast component (which distinguishes it from the mammalian enzymes) is the unusually high half-cystine content (210 residues per mole of enzyme, accounting for 16% of the enzyme by weight as compared to 2% for the mammalian enzymes). This feature is especially noteworthy when considered in the light of sulfhydryl group titration data. The chloroplast alkaline FDPase component contained only 10 p-mercuribenzoate-reactive -SH groups per mole (a value not increased by treating the enzyme with 8 M urea) and three to four DTNB-reactive -SH groups per mole (a value increased to 10 by treating the enzyme with 8 M urea). The data point to an unusually large cystine content of the alkaline   The reaction was carried out in Warburg vessels fitted with double side arms and center wells. The complete system contained (in one side arm) 0.25 ml of intact chloroplasts equivalent to one mg of chlorophyll; (in the second side arm) sodium fructose 1,6-diphosphate, 6  mammalian FDPase-a finding that has led to the conclusion (11,(19)(20)(21) that the activity of FDPase in gluconeogenesis is regulated by the level of AMP.
Such a role for AMP in sugar formation in photosynthesis can, however, be excluded. Fig. 4 shows that the chloroplast alkaline FDPase component, unlike the liver enzyme, is not inhibited by AMP.
Insensitivity of the chloroplast alkaline FDPase component to AMP was also observed by Preiss et al. (13).
Apart from differential sensitivity to AMP, plant and mammalian (or microbial) FDPases differ in the level of Mg* required for maximal activity.
The FDPases from rat liver (46), rabbit liver (47), and nonphotosynthetic microorganisms (48, 49) require, depending on the source, 0.7 to 5.0 mM Mg* for maximal activity, whereas the enzyme from photosynthetic cells requires about a lo-fold greater concentration of Mg*. Spinach leaf alkaline FDPase, dependent on pH, was reported to require 5 to 50 mM Mg* (12,13), and the EugZena grucilis enzyme was reported to require 50 mM Mg* (50). It is not known whether chloroplasts can maintain the high concentrations of Mgft reported to be required by the alkaline FDPase component.
A possible explanation is that light induces an accumulation in chloroplasts of Mg* sufficient to activate the alkaline FDPase component.
Such a dependence on light might also explain the photoactivation of the FDPase component observed in isolated intact chloroplasts (Table III); activity of the alkaline FDPase component was nearly doubled by illuminating the chloroplasts for 10 min prior to breaking osmotically and assaying the enzyme. The flux of Mg* required for activation would, however, be in a direction opposite to that observed experimentally: recent studies from several laboratories have shown that Mg* is expelled from chloroplasts on illumination and is reabsorbed in the dark (51-53).
We searched, therefore, for a mechanism for activation of the chloroplast alkaline FDPase other than a light-induced enhancement of the intrachloroplastic level of Mg++.
This search was made at a low level of Mg* (0.67 mM)-which cannot itself activate the FDPase component. fragments, PI&, equivalent to 0.1 mg of chlorophyll; and (in pmoles) Tris-HCl buffer, pH 8.0, 100; neutralized reduced glutathione, 5; sodium fructose 1,6-diphosphate, 6; sodium ascorbate, 10; 2,6-dichlorophenol indophenol, 0.1; EDTA-Na, 0.1; and MgC12, 1. Final volume, 1.5 ml; gas phase, argon. The reaction (carried out at 20" in Warburg flasks) was started by adding fructose diphosphate and chloroplast fragments from the side arm and was continued for 30 min under illumination (10,000 lux). The reaction was stopped and (after centrifuging off the precipitate) Pi was estimated-as in Table III (Table IV) constituted the first evidence that such a mechanism might be valid. Ferredoxin, photoreduced by chloroplasts, enhanced the release of Pi from fructose diphosphate 40-fold. The release of Pi required, in addition to reduced ferredoxin, MgCl,, fructose diphosphate, the alkaline FDPase component, and a new component, the protein factor. The reaction was stimulated by an -SH reagent (reduced glutathione or 2-mercaptoethanol) supplied in addition to reduced ferredoxin. Native spinach ferredoxin could not be replaced by bacterial ferredoxin (from Clostridium pasteurianum) but could be partially replaced by the nonphysiological dye methyl viologen (Table IV). The FDPase component was shown earlier to be activated in the dark when ferredoxin was reduced with hydrogen gas in the presence of hydrogenase (3). Reduced NADP or NAD could not replace light or hydrogen gas.
The protein factor, a component of the soluble protein fraction of chloroplasts, has been partly purified. It is heat-sensitive (5 min, 80") and, based on gel filtration, has a molecular weight of 40,000. Other properties are unknown.
A homogeneous chloroplast alkaline FDPase preparation (sup-fructose diphosphofe, mhf FIG. 5. Effect of fructose diphosphate concentration on ferredoxin-activated FDPase. Except for varying the fructose diphosphate concentration, conditions were as described in Table  IV. plied by Dr. J. Preiss) behaved like our own preparation in the ferredoxin-dependent release of Pi from fructose diphosphate. A homogeneous preparation of the enzyme from rabbit liver, however, showed no response to reduced ferredoxin under the conditions in Table IV centrations of Mgft (5 mu or greater), the extent of the ferredoxin-linked activation is strictly dependent on the Mgft concentration (Fig. 6). A consistent stimulation in Pi release was observed at all levels of MgClz tested but the effect of reduced ferredoxin was most pronounced at 0.67 to 2 mru MgClz.

Activation of Chloroplast Alkaline FDPase Component
by Dithiothreitol and --s--X--

Reagents
The nonphysiological -SH reagent dithiothreitol (55) could replace reduced ferredoxin in activation of the chloroplast alkaline FDPase component (Table V). Pi release from fructose diphosphate by the dithiothreitol-activated FDPase component occurred in the dark without chloroplasts and required MgCla and the protein factor. The requirement for the protein factor in activation of the FDPase component by dithiothreitol was specific-an equivalent amount of bovine serum albumin or of various protein fractions obtained in purification of the alkaline FIG. 7. Effect of cystamine on chloroplast alkaline FDPase component.
Final volume, 1.5 ml. Gas phase, argon. The reaction (carried out at 20" in Warburg flasks) was started by adding fructose diphosphate from the side arm and was stopped by adding 0.5 ml of 10% trichloroacetic acid. Pi released was estimated.
FDPase component was without effect. Cysteine, reduced glutathione, 2-mercaptoethanol, and sodium dithionite did not replace dithiothreitol in activation of the alkaline FDPase component.
Cystamine and DTNB, S-S compounds found by Pontremoli et al. (15) to activate liver FDPase, also activated the homogeneous chloroplast alkaline FDPase component. Fig. 7 shows that cystamine enhanced 50-fold the release of Pi from fructose diphosphate; the protein factor had no effect on the FDPase component under these conditions.
When cystamine was replaced by an equivalent amount of DTNB, a lo-fold increase in Pi release was observed.
Activation by S-S reagents (via disulfide exchange) provides a model system for regulation of the FDPase reaction, first for mammalian cells (15) and now for chloroplasts.
The physiological significance of this type of activation is, however, an open question.
In sum, the present experiments provide evidence that the FDPase system of chloroplasts is activated by light.
Ferredoxin, a component of the system, is reduced photochemically and, in collaboration with another component, the protein factor, reduced ferredoxin activates a third component, the alkaline FDPase.
Once activated, alkaline FDPase catalyzes the formation of fructose 6-phosphate (and Pi) from fructose diphosphate-a key reaction in the reductive pentose phosphate cycle of photosynthesis.
The mechanism of activation of the alkaline FDPase component is not known, but it may involve a reduction by reduced ferredoxin of specific S-S groups that cannot be reduced by more electropositive agents such as reduced glutathione or 2mercaptoethanol (Equations 4 and 5).

HS-)
The abundance of cysteine residues iu the alkaline FDPase and the replaceability of reduced ferredoxin by dithiothreitol or reduced methyl viologen are in accord with such a mechanism.