Regulatory seryl-phosphorylation of C4 phosphoenolpyruvate carboxylase by a soluble protein kinase from maize leaves

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 269, No. 2, March, pp. 526-535,1989

Regulatory Seryl-Phosphorylation of C4 Phosphoenolpyruvate by a Soluble Protein Kinase from Maize Leaves’ JIN-AN Department

of Biochemistry,

JIAO AND RAYMOND

University

of Nebraska-Lincoln,

Carboxylase

CHOLLET’ East Campus, Lincoln, Nebraska 68583-0718

Received September 8,1988, and in revised form November

8,1988

A reconstituted system composed of purified phosphoenolpyruvate carboxylase (PEPCase) and a soluble protein kinase (PK) from green maize leaves was developed to critically assess the effects of in vitro protein phosphorylation on the catalytic and regulatory (malate sensitivity) properties of the target enzyme. The PK was partially purified from light-adapted leaf tissue by ammonium sulfate fractionation (O-69% saturation fraction) of a crude extract and blue dextran-agarose affinity chromatography. The resulting preparation was free of PEPCase. This partially purified protein kinase activated PEPCase from dark-adapted green maize leaves in an ATP-, Mgs’-, time-, and temperaturedependent fashion. Concomitant with these changes in PEPCase activity was a marked decrease in the target enzyme’s sensitivity to feedback inhibition by L-malate. The PKmediated incorporation of “P from [y-32P]ATP into the protein substrate was directly correlated with these changes in PEPCase activity and malate sensitivity. The maximal molar 32P-incorporation value was about 0.25 per lOO-kDa PEPCase subunit (i.e., 1 per holoenzyme). Phosphoamino acid analysis of the 32P-labeled target enzyme by two-dimensional thin-layer electrophoresis revealed the exclusive presence of phosphoserine. These in vitro results, together with our recent studies on the light-induced changes in phosphorylation status of green maize leaf PEPCase in vivo (J.-A. Jiao and R. Chollet (1988) Arch. Biochem. Biophys. 261,409-41’7), collectively provide the first unequivocal evidence that the seryl-phosphorylation of the dark-form enzyme by a soluble protein kinase is responsible for the changes in catalytic activity and malate sensitivity of C, PEPCase observed in vivo during dark/light transitions of the parent leaf tissue. o 1989 AcademicPress,Inc.

and Pi in the mesophyll-cell cytoplasm during &-photosynthesis by plants such as maize (1). In recent years, substantial progress has been made in the areas of PEPCase structure, mechanism of catalysis, and modes of regulation (2). For example, the amino acid sequence of the lOOkDa maize leaf PEPCase monomer has been deduced from sequence analysis of cDNA (3). In addition, studies using groupselective chemical modifiers and PEP analogs have provided considerable insight into the catalytic reaction mechanism and

Phosphoenolpyruvate carboxylase (PEPCase,3 EC 4.1.1.31) catalyzes the conversion of PEP and bicarbonate to oxaloacetate 1 This research was supported in part by Grant DMB-8’704237 from the National Science Foundation and is published as Paper No. 8732, Journal Series, Nebraska Agricultural Research Division. 2 To whom correspondence should be addressed. 3 Abbreviations used: PEPCase, phosphoenolpyruvate carboxylase; PEP, phosphoenolpyruvate; CAM, Crassulacean acid metabolism; PVP, polyvinylpolypyrrolidone; ATP-y-S, adenosine-5’-0-(3-thiotriphosphatc); PEG, polyethylene glycol; Mops, 4-morpholinepropanesulfonic acid; DTT, dithiothreitol; FPLC, fast-protein liquid chromatography; PEPC-PK, phosphoenolpyruvate carboxylase protein kinase; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid; 0003-9861/89 $3.00 Copyright0 1989by AcademicPress,Inc. All rights of reproductionin any form reserved.

SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis. 526

527

&-LEAF PHOSPHOenolPYRUVATE CARBOXYLASE the structures of the catalytic- and regulatory-site domains as well as the subunit arrangement of the homotetrameric holoenzyme (2, 4-9). It has also been established that the tetrameric quaternary structure is closely related to the catalytic activity of the &-enzyme (10-12). Another vigorously investigated area of research related to PEPCase is the effect of dark .- light transitions on the catalytic activity and regulatory properties of the C,- (13-20) and CAM-leaf (21-23) enzymes. Evidence has accumulated from several research groups (13-20) indicating that PEPCase activity in illuminated Cq- leaves is about two- to threefold greater than that from darkened plants when assayed at suboptimal, but physiological levels of PEP and pH. In parallel with these lightinduced changes in catalytic activity, the light-form enzyme is less sensitive to feedback inhibition by malate. Concomitant with these changes in enzymatic properties, both Cd- (1’7, 18, 24) and CAM- (21) PEPCases undergo changes in covalent phosphorylation status upon light/dark transitions of the parent leaf tissue. We have recently reported that the differing catalytic and regulatory properties of maize PEPCase purified from light- and dark-adapted green leaf tissue are related, at least in part, to the degree of covalent seryl-phosphorylation of the protein in vivo (18). studies Related in vitro phosphorylation in our laboratory with maize leaf extracts indicated the presence of an ATP-dependent soluble protein kinase(s) which modified crude PEPCase exclusively on serine residues (25). In the present study, we have further developed a reconstituted system composed of electrophoretically pure PEPCase and a partially purified soluble protein kinase preparation from dark- and light-adapted maize leaves, respectively, to perform detailed analysis of the effects of in vitro phosphorylation on the catalytic and regulatory properties of C,-leaf PEPCase. By exploiting this reconstituted system, we provide the first unequivocal evidence for a direct correlation between changes in PEPCase activity/malate sensitivity and the degree of seryl-phosphorylation of the &-leaf protein.

MATERIALS

AND

METHODS

Materials. PEP (monocyclohexylammonium salt), NADH, ATP, glycerol, L-malate, blue dextran-agarose (No. B-3760), insoluble PVP, and pig heart lactate dehydrogenase and malate dehydrogenase were obtained from Sigma, and ATP-r-S from BoehringerMannheim. [y-32P]ATP was purchased from Amersham, PEG 8000 from J. T. Baker, and hydroxylapatite (Bio-Gel HTP) from Bio-Rad. All other reagents were of analytical grade. The generous gift of purified spinach leaf thioredoxin h was kindly provided by Dr. Bob B. Buchanan of the University of California at Berkeley. Maize (Zea rnags L., cv. Golden Cross Bantam) plants were grown from seed in an environmentally controlled chamber (12-h photoperiod, 25°C day/ 16°C night) at about 500 PE. m-*. s-i (400-700 nm) at leaf height. Laminar tissue from 4- to B-week-old plants was used for sample preparation. PurQication of PEPCase. Both light- and darkform PEPCases were purified to apparent electrophoretie homogeneity (e.g., see (18) and Fig. 6A, lane 2) from green maize leaves as previously described (18) with some modifications. The enzyme was extracted by blending diced leaves in a Waring Blender in 50 mM Mops-KOH, pH ‘7.3, 5 mM malate, 1 mM EDTA, 7.5 mM 2-mercaptoethanol, 10 mM MgC&, and 2% (w/ v) insoluble PVP. The crude homogenate was filtered through an 80-pm nylon net and then centrifuged at 30,OOOgfor 20 min at 4°C. Solid PEG 8000 was added to the supernatant fluid to a final concentration of 85 mg/ml. Following incubation with gentle stirring at 0°C for 15 min, the suspension was centrifuged for 15 min at 30,OOOg.The supernatant liquid was recovered, brought to 14% (w/v) PEG 8000, and then incubated as before. The precipitated protein from the 8.5-14% PEG fraction was recovered by a lo-min centrifugation at 10,OOOg.The pellet was dissolved in 20 mM Bistris propane/HCl, pH 7.0, containing 20% (v/v) glycerol, 1 mM DTT, 0.1 mM EDTA, and 5 mM malate. The sample was then fractionated on an hydroxylapatite column (18). The fractions containing PEPCase activity were pooled and directly chromatographed twice on a Pharmacia Mono Q anion-exchange column (0.5 X 5 cm) connected to an automated LCC-500 Pharmacia FPLC system (18). A linear elution gradient of O-O.35 M NaCl in 20 ml 50 mM Tris-HCl, pH 8.0,l mM DTT, 5 mM malate, and 5% (v/v) glycerol was employed. Purified PEPCase was stored directly in 50% (v/v) glycerol and 5 mM malate at -20°C (18). Partial purijcatim of PEPGPK. All the following steps were performed at 4°C. (i) Homogenization and ammonium sulfate fractionation: About 40 g of preilluminated (2 h) green leaf tissue was used for each preparation. The tissue was diced and homogenized with 5 vol of Buffer A (100 mM Tris-HCl, pH 8.0,lO mM MgClz, 14 mM 2-mercaptoethanol, 1 mM EDTA, and 5% (v/v) glycerol) containing 2% (w/v) insoluble PVP in a Waring Blender.

528

JIAO

AND

The homogenate was filtered through an 80-pm nylon net and then centrifuged at 30,000~ for 20 min. The crude supernatant fluid was brought to 60% saturation (0°C) with solid ultrapure ammonium sulfate and the precipitated protein collected by eentrifugation (lO,OOOg, 10 min) after about 1 h of gentle stirring. (ii) Blue dextran-agarose affinity chromatography: The O-60% saturated (NH&SO., protein precipitate was resuspended in a minimal volume of Buffer B (50 mM Tris-HCI, pH 7.5,l mhl DTT, and 5% (v/v) glycerol) and desalted through a Sephadex G-25 column (2.5 X 15 cm) equilibrated with the same buffer. The gel-filtered protein sample was clarified by centrifugation (3O,OOOg,10 min) and loaded onto a 25-ml column (2.5 X 5.1 cm) of blue dextran-agarose (flow rate, 0.2 ml/min) equilibrated with Buffer B. The affinity column was thoroughly washed with Buffer B to remove unbound material and then batch-eluted with 0.5 M NaCl in Buffer B. Azm peak fractions from the salt elution were pooled and the protein was concentrated by ammonium sulfate (60% saturation) precipitation at 4°C. This suspension was allowed to stand overnight at 0-4°C and then centrifuged for 10 min at 10,000~. The pellet was dissolved in a minimal volume of Buffer C (50 mrvf Tris-HCl, pH 7.8, 1 mM DTT, 20% (v/v) glycerol) and desalted on a Pharmacia PD-10 column equilibrated with the same buffer. The desalted protein preparation was used as the source of PEPC-PK. For long-term storage, the desalted sample was divided into 0.2-ml aliquots and stored at -20°C. Under these conditions, the preparation was stable for at least 4 weeks. Assays. PEPCase activity was assayed spectrophotometrically at 340 nm by coupling to exogenous NADH-malate dehydrogenase/lactate dehydrogenase (26) at 25°C. The standard reaction mixture (1.0 ml) contained 50 mM Hepes-KOH, pH 7.3, 2.5 mM PEP, 5 mM MgC&, 1 mM NaHCOa 0.2 mM NADH, 10 units of malate dehydrogenase, and 5 units of lactate dehydrogenase, f0.5 mM L-malate. The reaction was initiated by the addition of PEPCase and was linear for up to 2 min; the velocity during the first minute was computed by an enzyme kinetics program. Soluble protein content was determined by a sensitive dye-binding method (27) using the Bio-Rad dye reagent and crystalline bovine serum albumin as the standard. In vitro activatiodphosphorylation of PEPCase. Purified PEPCase was thoroughly dialyzed against Buffer C (enzyme to buffer volume ratio of 1:lOOO) for about 4 h at 4°C. The standard activation/phosphorylation reaction was effected by preincubating the dialyzed PEPCase sample with partially purified PEPCPK and 0.5 mM ATP in Buffer C containing 5 mM MgClz for up to 1 h at 30°C. Under these conditions, the -ATP control PEPCase was stable with respect to its activity and malate sensitivity (e.g., see Figs. 1 and 2, respectively). Activation/phosphorylation was

CHOLLET initiated by the addition of ATP, and 20-111aliquots were withdrawn at the specified times and assayed spectrophotometrically for PEPCase activity or for phosphorylation of the lOO-kDa PEPCase subunit by SDS-PAGE (see below), in which case [T-~~P]ATP was used. The difference between PEPCase activity at zero time and after ATP addition is a measure of ATP-dependent activation (expressed as a relative percentage of the zero-time sample (100%)). For analysis of “P-incorporation, aliquots were removed from the preincubation mixture at various times and mixed with an equal volume of SDS sample buffer (125 mM Tris-HCl, pH 6.8,4% (w/v) SDS, 20% (v/v) glycerol, 10% (v/v) 2-mercaptoethanol, 0.02% (w/v) bromophenol blue). The samples were then heated in boiling water for 90 s and fractionated by SDS-PAGE (18,28). The PEPCase subunit band was excised from the destained gel, and the level of 32P-incorporation was determined by Cerenkov counting. Phosphorylated protein samples for autoradiography were separated by SDS-PAGE (l&28) and then autoradiographs were made on Kodak X-Omat AR film utilizing two Lightning-Plus intensifying screens (DuPont) at -80°C. Y2P-labeled phosphoamino acid analysis. The 32P-labeled PEPCase subunit band was excised from the destained denaturing gel and extracted by electroelution using an electrophoretic concentrator (Model 1750, Isco, Inc.). The eluted protein sample was thoroughly dialyzed against distilled water for 3 h at 4°C and then partially hydrolyzed in 6 N HCl at 105°C for 3 h. 32P-labeled phosphoamino acids were analyzed by two-dimensional thin-layer electrophoresis on sheets of 13255 cellulose (Kodak) according to (25, 29), followed by autoradiography. RESULTS

AND

DISCUSSION

Partial puri&zatim of PEPC-PK. Previous studies in our laboratory (25) indicated that the PEPC-PK is a soluble, ATP-dependent protein kinase, as demonstrated by the in vitro seryl-phosphorylation of crude PEPCase in an (NH&SO,-fractionated maize leaf extract (i.e., desalted O60% saturation fraction). Based on these initial findings, we further developed this in vitro phosphorylation system by employing affinity chromatography on blue dextran-agarose to partially purify the protein kinase. Following desalting against Buffer C, the activity of the resulting PEPC-PK preparation was stable for at least 2 days at 4°C or 4 weeks at -20°C. While this partially purified preparation still contained several polypeptides, it was devoid of detectable PEPCase protein (e.g., see Fig. 6A, lane 4) and activity.

Cd-LEAF PHOSPHOenolPYRUVATE CARBOXYLASE Preliminary experiments indicated that there were no dramatic changes in activity and malate sensitivity of PEPCase when crude maize leaf extracts (i.e., the desalted O-60% saturated (NH&SO4 fraction), prepared from either dark- or light-adapted tissue, were preincubated separately or together in the presence of ATP and MgCl, at 30°C (data not shown and (25)). However, when 5 mM ATP-r-S, an ATP analog, was substituted for ATP as a P-donor, changes in activity of PEPCase (about 50% activation when assayed in the presence of 0.5 mM L-malate) in extracts from darkadapted leaves were reproducibly measured following a 90-min preincubation at 30°C. Given that thiophosphorylated proteins are extremely resistant to the action of protein phosphatases (30, 31), it is reasonable to suggest that our inability to detect an ATP-dependent activation of PEPCase in crude leaf extracts, even in the presence of an ATP-regenerating system and an adenylate kinase inhibitor (25), was due to endogenous protein phosphatase(s) in the crude O-60% (NH&SO, fraction. This was likely the case since (i) our previous studies (25) indicated that the in vitro phosphorylation reaction mixture allowed for maximum [y-32P]ATP-dependent phosphorylation of crude PEPCase within 15 to 20 min, after which the level of seryl-phosphorylation declined to a lower steadystate amount, presumably due to hydrolysis, and (ii) PEPCase activation can be readily observed following fractionation of the crude leaf extract on blue dextran-agarose (see below). Because of the above-described limitations with the crude extract system, it was not possible to estimate the recovery and fold purification of the PEPC-PK preparation during purification. ATP- and protein kinase-dependent activation of purijed PEPCase. Both the light and dark forms of green maize leaf PEPCase are phosphorylated in vivo (17, 18), with the degree of seryl-phosphorylation being -50% greater in the former (18). Evidence that this difference in phosphorylation status was correlated with the light-induced increase in PEPCase activity and decrease in malate sensitivity was obtained by treatment of the purified light and dark enzyme forms with exogenous

529

alkaline phosphatase (18). Phosphatase treatment converted the more active, less malate-sensitive light-form PEPCase into a less active, more malate-sensitive enzyme form which closely resembled darkform PEPCase (18). In the presence of ATP and MgC12, preincubation of purified light- or dark-form PEPCase with the PEPC-PK partially purified from light-adapted maize leaves resulted in an increase in activity of both enzyme forms. However, the extent of activation of the dark-form PEPCase was about 120% when assayed in the absence of malate, compared to only -40% with the light enzyme (data not shown). These comparative results suggest that the lightform enzyme was not maximally phosphorylated under the growth or illumination (25% of full sunlight) conditions employed and/or partially dephosphorylated during purification, and that dark-form PEPCase is the preferred substrate for the PEPC-PK preparation. Since the protein kinase-mediated activation results were more pronounced using dark-form PEPCase as substrate, all subsequent experiments with this reconstituted system were performed with the electrophoretically pure (18), partially phosphorylated (17,18) dark-form enzyme. The time course of the protein kinasemediated activation of dark-form PEPCase at 30°C in the presence of ATP and MgC12, is shown in Fig. 1. PEPCase activity, measured in either the absence or presence of L-malate, increased dramatically over the first 30 min of preincubation with ATP and PEPC-PK, thereafter remaining constant for at least 20 min. Substitution of buffer for ATP (e.g., see Figs. 1 and 4), MgC12 (e.g., see Fig. 3), or the protein kinase produced little or no activation of PEPCase, indicating that activation was dependent on the simultaneous presence of these three factors. Moreover, heat treatment (100°C for 5 min) of the PEPC-PK preparation led to a complete loss of its effect on PEPCase activity. Activation was also temperature dependent, as indicated by the low degree of activation when preincubation was performed at 0 or 10°C. Typically, the ratios of PEPCase activity following preincubation with the protein

530

JIAO

AND

CHOLLET

OO

0

IO

I

I

I

I

20

30

40

50

Preincubotion

tATP(mm

60

1

FIG. 1. ATP-dependent activation of PEPCase. Purified dark-form PEPCase (33 pg) and the partially purified protein kinase (96 pg) were preincubated at 30°C in 0.5 ml 50 mM Tris-HCl, pH 7.8, containing 20% (v/v) glycerol, 1 mM DTT, 5 mM MgClz, f0.5 mM ATP. At the indicated times, 20-4 aliquots were removed for assay of PEPCase activity in the presence (0) or absence (m) of 0.5 mM L-malate at pH 7.3 and 2.5 mM PEP. The zero-time (100%) specific activities were 6.70 and 1.95 pmol NADH oxidized mini mg protein-’ when measured in the absence and presence of malate, respectively. The activity of the -ATP control PEPCase (A) was assayed in the absence of malate. Little or no activation was observed with the complete preincubation mixture lacking the protein kinase.

kinase, MgClz, and ATP to those obtained minus ATP were about 2 when PEPCase was assayed under standard conditions (pH 7.3, 2.5 mM PEP) without malate, increasing to 4-5 when assayed in the presence of 0.5 MM L-malate (Fig. 1). In related studies, the effect of preincubation with PEPC-PK and ATP on the malate sensitivity of the dark-form enzyme was investigated (Fig. 2). It is evident that the changes in malate sensitivity of PEPCase are kinetically coupled to the concomitant increase in catalytic activity (Figs. 1 and 2). For example, accompanied by about a 120% increase in PEPCase activity, malate inhibition decreased from about 70% at zero time to about 40% after a 30-min preincubation with PEPC-PK and ATP at 30°C. Thus, like the light-induced effects on Cl-PEPCase activity in vivo (13-20), treatment of the dark-form enzyme in vitro with ATP, PEPC-PK, and MgClz re-

FIG. 2. Changes in malate sensitivity of dark-form PEPCase following preincubation with the protein kinase in the absence (A) or presence (0) of 0.5 mM ATP. Experimental conditions were the same as described for Fig. 1. PEPCase activity was assayed under standard conditions (pH 7.3, 2.5 mM PEP), rt0.5 mM L-malate.

sulted in marked changes in its catalytic activity and feedback inhibition by malate. Studies on the effect of exogenous MgClz concentration on PEPCase activation indicated that M$’ is required for the activity of the PEPC-PK, saturating at a total concentration of -2 mM in the presence of 0.5 mM ATP (Fig. 3). Activation was not observed in the absence of exogenous MgClz. An equimolar total concentration of Mn2+

MgC12 (mM1

FIG. 3. ATP- and protein kinase-mediated activation of PEPCase as a function of exogenous MgClz. Six micrograms of dark-form PEPCase and 17 pg of PEPC-PK were preincubated in 0.1 ml Buffer C containing 0.5 mM ATP and varying total concentrations of MgClz. After 30 min at 3O”C, PEPCase activity was assayed in the absence of L-malate.

C&-LEAF

Premcubotvx

PHOSPHOenolPYRUVATE

with ATP(mM)

FIG. 4. Protein kinase-mediated activation of PEPCase as a function of ATP. Dark-form PEPCase (5.3 fig) and the PEPC-PK (17 fig) were preincubated in 0.1 ml Buffer C containing 5 IIIM MgClz and varying total concentrations of ATP. After 30 min at 3O”C, PEPCase activity was assayed in the absence of L-malate.

was able to partially substitute for M$+ in the preincubation mixture, resulting in a -50% reduction in the extent of PEPCase activation when assayed in the presence of 0.25 mM malate. NaCl was severely inhibitory. In the presence of 5 mM MgClz, the total concentration of ATP required for maximal PEPC-PK activation of PEPCase was about 1 mM, with higher concentrations being inhibitory (Fig. 4). The reason for this inhibition is not known at present. Evidence for the protein kinase-mediated seryl-phosphorylation of PEPCase. 32P-incorporation experiments were performed under the same conditions as described for PEPCase activation (e.g., Fig. 1) except that [T-~‘P]ATP was used instead of ATP. Figure 5 shows the time course of the PEPC-PK-mediated incorporation of 32P into the partially phosphorylated (17, 18) dark-form enzyme. As in the case of the changes in catalytic activity and malate sensitivity (Figs. 1 and 2, respectively), phosphorylation of the lOO-kDa PEPCase subunit displayed similar kinetics, plateauing after about 30 min at 30°C. At this time, the molar 32P-incorporation value was about 0.25 per lOO-kDa subunit. This stoichiometry is essentially equivalent to the difference in the covalent phosphorylation status of the light and dark forms of PEPCase observed in vivo (18).

531

CARBOXYLASE

Although the PEPC-PK preparation was only partially purified (Fig. 6A, lane 4), the PEPCase monomer was the major polypeptide phosphorylated in this reconstituted in vitro system (Fig. 6B, lane 6). As in the case of the activation-related experiments (Figs. 1 and 3), phosphorylation of PEPCase was not observed in the absence of either the protein kinase (Fig. 6B, lane 5) or exogenous Me (data not shown). Related phosphoamino acid analysis of the 32P-labeled PEPCase subunit by two-dimensional thin-layer electrophoresis revealed the exclusive presence of phosphoserine (Fig. ‘7), in excellent agreement with previous in vivo (17,18,24) and in vitro (25) phosphorylation studies of the Cdleaf enzyme. Most notably, when the time courses for 32P-incorporation (Fig. 5) and activation (Fig. 1) are replotted on a common basis (Fig. S), there is a direct correlation between the protein kinase- and ATPdependent changes in PEPCase activity/ malate sensitivity and the degree of serylphosphorylation of the dark-form enzyme.

c ': a %

OO

IO Incutdion

20

x)

with Cy-32PlATP

40

50

60

(mm)

FIG. 5. Time course of 32P-incorporation from [y“‘P]ATP into dark-form PEPCase. Experimental conditions were the same as described in the legend to Fig. 1 except that 0.5 mM [“P]ATP (3.89 X lo7 dpm/ pmol) was used instead of ATP. At the indicated times, aliquots were removed from the 0.5-ml incubation mixture and added to an equal volume of SDS sample buffer. After heating and separation by SDSPAGE (18, 28), the lOO-kDa PEPCase-subunit band was excised and “‘P-dpm determined by Cerenkov counting. All incorporation values were corrected for background radiation, which was determined by counting a region of the gel not containing protein.

532

JIAO 1

66.2

2

3

AND 4

CHOLLET 5

6

7

-

-

21.5 14.4

-

FIG. 6. Phosphorylation of PEPCase in the in vitro reconstituted system. Dark-form PEPCase was phosphorylated for 1 h at 30°C as described in Fig. 5. (A) Denaturing gel stained with Coomassie blue R-250; (B) Autoradiograph made from A. Lane 1, molecular weight markers (values in kDa); lanes 2 and 5, purified dark-form PEPCase plus r’P]ATP; lanes 3 and 6, purified dark-form PEPCase plus PEPC-PK and [aaP]ATP; lanes 4 and 7, PEPC-PK plus [aaP]ATP.

Possible redox regulation of PEPCase and PEPC-PK by thioredoxin h. Buchanan and co-workers (32) have recently identified a specific type of thioredoxin, called thioredoxin h, in the cytosolic fraction of green leaves of spinach and pea. Although the function of this NADP/thioredoxin system is unknown, it could represent a regulatory link coupling the redox status of the chloroplast with certain cytoplasmic target enzymes involved in photosynthetic carbon assimilation. Since PEPCase and presumably the protein kinase are localized in the mesophyll cytoplasm, we explored the possible role of this redox system in regulating these two enzymes by preincubating the reconstituted phosphorylation system (containing PEPCase, PEPC-PK, ATP, and MgC12) or dark-form PEPCase alone with purified spinach leaf thioredoxin h which had been chemically reduced by pretreatment with 50 mM DTT for 20 min at 25°C. Subsequent assay of PEPCase activity (+I,-malate) indicated

that neither the protein kinase nor darkform PEPCase was affected by a l-h preincubation with reduced thioredoxin h at pH 7.1 or 7.8 and 30°C. This negative finding, however, does not unequivocally exclude the possibility that thioredoxin h may play a role in the redox regulation of PEPCase or PEPC-PK in &-photosynthesis, because the thioredoxin preparation used in this experiment was purified from &-leaf tissue. CONCLUDING

REMARKS

A number of soluble enzymes involved in Cs- or &-photosynthesis are known to be reversibly light-activated. Of these, only C4 mesophyll-chloroplast pyruvate,orthophosphate dikinase (33, 34) and C4 mesophyll-cytoplasm PEPCase (17,18,24) have been shown to undergo reversible changes in phosphorylation status in response to dark/light transitions of the parent leaf tissue. Generally, regulatory protein phos-

&-LEAF

PHOSPHOenolPYRUVATE

CARBOXYLASE

pH

1.9

+-

FIG. 7. Two-dimensional thin-layer electrophoretic separation of phosphoamino acids. Dark-form PEPCase was phosphorylated in vitro and separated by SDS-PAGE. The 32P-1abeled 100-kDa PEPCase-subunit band was excised, electroeluted, thoroughly dialyzed against distilled water, acid-hydrolyzed at 105°C for 3 h, and electrophoresed according to Refs. (25, 29). Circled areas represent the location of the phosphoamino acid standards as detected by ninhydrin. An autoradiograph was prepared from the thin-layer chromatogram. In addition to phosphoserine, an unidentified peptide is also 32P-1abe1ed. x, origin.

phorylation can lead to either activation or inactivation of the target enzyme (35). However, the extent of these changes in activity varies greatly among different target enzymes. For some enzymes, like the dikinase (33), the difference in catalytic activity between the phosphorylated and dephosphorylated forms is very large, to the extent that one form is active and the other considered kinetically inactive. On the other hand, changes in phosphorylation status of other enzymes result only in several-fold changes in activity, so that one of the two forms is less active compared to the other. In addition, protein phosphorylation can also alter the regulatory properties of the target enzyme (35). In the case of &-leaf PEPCase, both the light and dark enzyme forms have been shown to contain

phosphoserine and possess catalytic activity (17, 18, 24). However, further serylphosphorylation of the partially phosphorylated dark-form PEPCase, both in viva (17,18,24) and in vitro (Figs. 1,2,7, a), results in an increase in its catalytic activity and a decrease in its sensitivity to feedback inhibition by malate. The results presented in this report clearly demonstrate the presence of a soluble protein kinase(s) in green maize leaves that activates/phosphorylates dark-form PEPCase in vitro. Although the partially purified preparation of PEPC-PK contains several other proteins, some of which undergo an ATP-dependent phosphorylation (lanes 4 and ‘7 in Figs. 6A and 6B, respectively), the lOO-kDa PEPCase subunit is the major polypeptide phosphorylated in

534

JIAO

Preincubation

AND

with ATPTP(min)

FIG. 8. Direct correlation between the ATP- and protein kinase-mediated changes in activity and seryl-phosphorylation of PEPCase i?z vitro. Changes in activity and 32P-incorporation were equated to 100% after a 30-min preincubation at 30°C with 0.5 mM ATP plus PEPC-PK and 0% at 0 min. PEPCase activity was assayed at pH 7.3 and 2.5 mM PEP, with (m) or without (0) 0.5 mML-malate. (A) 32P-incorporation into the lOO-kDa PEPCase-subunit band.

this reconstituted system (Fig. 6B, lane 6). The direct correlation between the protein kinase-mediated changes in the catalytic and regulatory properties and seryl-modification of the dark-form PEPCase (Fig. 8) provides the first unequivocal evidence in support of the view (17,18,24) that the covalent seryl-phosphorylation of the target enzyme by an ATP-dependent (25) soluble protein kinase is the mechanism to which the light-induced changes in catalytic activity and malate sensitivity of G-leaf PEPCase observed in viva (13-20) are attributed. This hypothesis is further supported by the excellent correspondence between the magnitude of the protein kinaseand ATP-dependent changes in activity, malate sensitivity, and phosphorylation status of the dark-form enzyme observed in vitro (Figs. 1, 2, 5) to those occurring during dark + light transitions of the parent leaf tissue in viva (18). It is noteworthy that several of the mesophyll cell-specific steps of the &-cycle are light-modulated in G-plants, including the transport of pyruvate across the chloroplast envelope-membrane (36) and the ensuing reaction sequence of pyruvate + PEP -+ oxaloacetate + malate cat-

CHOLLET

alyzed by pyruvate,Pi dikinase (33, 34), PEPCase (13-20), and NADPH-malate dehydrogenase (33), respectively. Exactly why such a major portion of the &-cycle is light-modulated is not known, but perhaps it serves to coordinately regulate the activities of the mesophyll-specific carbon-assimilating reactions with those of the light-activated bundle sheath-specific enzymes of the Calvin cycle (e.g., ribulosebisphosphate carboxylase, fructosebisphosphatase, sedoheptulosebisphosphatase, phosphoribulokinase) during Cd-photosynthesis. While the specific nature of the dark + light signal controlling the lightmodulation of pyruvate transport and NADPH-malate dehydrogenase activity in the mesophyll chloroplast is thought to be the photosynthetic electron transportinduced changes in pH across the envelope-membrane (36) and in the redox status of stromal thioredoxin and NADP (33, 37, 38), respectively, little is known with certainty with respect to how light modulates the phosphorylation status (activation state) of the dikinase and PEPCase. The light activation of these sequential, mesophyll-cell enzymes is known to be dependent, either directly or indirectly, on photosynthetic electron transport and/or photophosphorylation (13, 14, 33, 38). In the case of chloroplastic pyruvate,Pi dikinase, a role for stromal metabolite effectors (e.g., pyruvate, adenine nucleotides) in the reversible light-activation/dephosphorylation process has been postulated (33, 38-40), but recently questioned (41, 42). With respect to cytoplasmic PEPCase, while the sulfhydryl status of neither PEPC-PK nor the dark form of its target enzyme appears to be involved in the overall light activation process (this study and Ref. (20)), photosynthetically derived metabolite effecters and calcium/calmodulin have been recently implicated (14,19, 43). It is obvious that further purification of the PEPC-PK preparation described in this report would enable us to perform more detailed studies on its regulatory, enzymatic, and structural properties. Along these lines, we are currently developing FPLC-based procedures for purification of this soluble protein kinase following the affinity chromatography step.

&-LEAF

PHOSPHOewZPYRUVATE

ACKNOWLEDGMENTS We thank Ms. Shirley Condon for her excellent technical assistance and Professor Bob B. Buchanan for providing us with a sample of purified spinach leaf thioredoxin h.

REFERENCES 1. EDWARDS, G. E., AND HUBER, S. C. (1981) Biothem. Plants 8,237-281. 2. ANDREO, C. S., GONZALEZ, D. H., AND IGLESIAS, A. A. (1987) FEBSLett. 213,1-8. 3. YANAGISAWA, S., IZUI, K., YAMAGUCHI, Y., SHIGESADA, K., AND KATSUKI, H. (1988) FEBS Lett. 229,107-110. 4. ANDREO, C. S., IGLESIAS, A. A., PODEST~, F. E., AND WAGNER, R. (1986) B&him. Biophys. Acta 870,292-301. 5. WAGNER, R., PODEST~, F. E., GONZ~EZ, D. H., AND ANDREO, C. S. (1988) Eur. J Biochem. 173, 561-568. 6. D~Az, E., O’LAUGHLIN, J. T., AND O’LEARY, M. H. (1988) Biochemistry 27,1336-1341. 7. SIKKEMA, K. D., AND O’LEARY, M. H. (1988) Biochemistry 27,1342-1347. 8. JIAO, J.-A., AND SHI, J.-N. (1987) Sci. Sin. (Ser. B) 30,1060-1069. 9. GONZ.&LEZ, D. H., AND ANDREO, C. S. (1989) Trends B&hem. Sci. 14, in press. 10. SHI, J.-N., Wu, M.-X., AND ZHA, J.-J. (1981) Acta PhytophysioL Sin. 7,317-326. 11. WALKER, G. H., Ku, M. S. B., AND EDWARDS, G. E. (1986) Plant PhysioL 80,848-855. 12. WAGNER, R., GONZ~EZ, D. H., PODEST~, F. E., AND ANDREO, C. S. (1987) Eur. J B&hem. 164, 661-666. 13. KARABOURNIOTIS, G., MANETAS, Y., AND GAVALAS, N. A. (1983) Plant PhysioL 73,735-739. 14. SAMARAS, Y., MANETAS, Y., AND GAVALAS, N. A. (1988) Photosynth Res. 16,233-242. 15. HUBER, S. C., AND SUGIYAMA, T. (1986) Plant PhysioL 81,674-677. 16. JIAO, J.-A., AND SHI, J-N. (1987) Actu PhytophysioL Sin. 13,190-196. 17. NIMMO, G. A., MCNAUGHTON, G. A. L., FEWSON, C. A., WILKINS, M. B., AND NIMMO, H. G. (1987) FEBS Lett. 213,18-22. 18. JIAO, J.-A., AND CHOLLET, R. (1988) Arch. Bicthem. Biophys. 261,409-417. 19. DONCASTER, H. D., AND LEEGOOD, R. C. (1987) Plant PhysioL 84,82-87.

CARBOXYLASE

535

20. RODR~QUEZ-SOTRES, R., AND MU%OZ-CLARES, R. A. (1987) J. Plant PhysioL 128,361-369. 21. NIMMO, G. A., NIMMO, H. G., HAMILTON, I. D., FEWSON, C. A., AND WILKINS, M. B. (1986) Bicthem. J. 239,213-220. 22. WV, M.-X., AND WEDDING, R. T. (1985) Arch. Bicthem. Biophys. 240,655-662. 23. KRUGER, I., AND KLUGE, M. (1987) Bot. Acta 101, 24-27. 24. GUIDICI-ORTICONI, M.-T., VIDAL, J., LE MAR%CHAL, P., THOMAS, M., GADAL, P., AND RI?MY, R. (1988) Biochimie 70,769-772. 25. BUDDE, R. J. A., AND CHOLLET, R. (1986) Plant PhysioL 82,1107-1114. 26. MEYER, C. R., RUSTIN, P., AND WEDDING, R. T. (1988) Plant PhysioL 86,325-328. 27. BRADFORD, M. M. (1976) AnaL Biochem. 72,248254. 28. LAEMMLI, U. K. (1970) Nature (Lcdon) 227,680685. 29. HUGANIR, R. L., MILES, K., AND GREENGARD, P. (1984) Proc. NutL Acad Sci. USA 81,6968-6972. 30. GRATECOS, D., AND FISCHER, E. H. (1974) Biothem. Biophys. Res. Commun 58,960-967. 31. CASSEL, D., AND GLASER, L. (1982) Proc. NatL Acud. Sci. USA 79,2231-2235. 32. FLORENCIO, F. J., YEE, B. C., JOHNSON, T. C., AND BUCHANAN, B. B. (1988) Arch. Biochem Biophys. 266,496-507. 33. EDWARDS, G. E., NAKAMOTO, H., BURNELL, J. N., AND HATCH, M. D. (1985) Annu. Rev. Plant Physiol. 36,255-286. 34. BUDDE, R. J. A., HOLBROOK, G. P., AND CHOLLET, R. (1985) Arch. Biochem. Biophys. 242,283-290. 35. BUDDE, R. J. A., AND CHOLLET, R. (1988) PhysioL Plant. 72,435-439. 36. OHNISHI, J.-I., AND KANAI, R. (1987) Plant Cell PhysioL 28,243-251. 37. REBEILLE, F., AND HATCH, M. D. (1986) Arch Biothem. Biophys. 249,164-170. 38. NAKAMOTO, H., AND EDWARDS, G. E. (1986) Plant PhysioL 82,312-315. 39. BUDDE, R. J. A., ERNST, S. M., AND CHOLLET, R. (1986) B&hem. J. 236,579-584. 40. BURNELL, J. Ei., JENKINS, C. L. D., AND HATCH, M. D. (1986) Aust. .I Plant Physiol. 13,203-210. 41. USUDA, H. (1988) Plant Physiol. 88.1461-1468. 42. ROESKE, C. A., AND CHOLLET, R. (1989). Plant PhysioL 89, in press. 4 43. ECHEVARR~A, C., VIDAL, J., LE MAR&HAL, P., BRULFERT. J., RANJEVA, R., AND GADAL, P. (1988) Biochem. Biophys. Res. Commun. 155, 835-840.