ARCHIVES
OF BIOCHEMISTRY
Vol. 298, No. 2, November
AND
BIOPHYSICS
1, pp. 688-696, 1992
Subunit Interactions of Rubisco Activase: Polyethylene Glycol Promotes Self-Association, Stimulates ATPase and Activation Activities, and Enhances Interactions with Rubisco’ Michael
E. Salvucci
U.S. Department of Agriculture, Agricultural Research Service and Agronomy Department, University of Kentucky, Lexington, Kentucky 40546-0076
Received May 13, 1992, and in revised form July 7, 1992
The effect of polyethylene glycol (PEG) on the enzymatic and physical properties of ribulose- 1,5-bisphosphate carboxylasefoxygenase (Rubisco) activase was examined. In the presence of PEG, Rubisco activase exhibited higher ATPase and Rubisco activating activities, concomitant with increased apparent affinity for ATP and Rubisco. Specific ATPase activity, which was dependent on Rubisco activase concentration, was also higher in the presence of Ficoll, polyvinylpyrrolidone, and bovine serum albumin. The ability of Rubisco activase to facilitate dissociation of the tight-binding inhibitor 2-carboxyarabinitol l-phosphate from carbamylated Rubisco was also enhanced in the presence of PEG. Mixing experiments with Rubisco activase from two different sources showed that tobacco Rubisco activase, which exhibited little activation of spinach Rubisco by itself, was inhibitory when included with spinach Rubisco activase. Polyethylene glycol improved the ability of tobacco and a mixture of tobacco plus spinach Rubisco activase to activate spinach Rubisco. Estimates based on rate zonal sedimentation and gel-filtration chromatography indicated that the apparent molecular mass of Rubisco activase was two- to fourfold higher in the presence of PEG. The increase in apparent molecular mass was consistent with the propensity of solvent-excluding reagents like PEG to promote self-association of proteins. Likewise, the change in enzymatic properties of Rubisco activase in the presence of PEG and the dependence of specific activity on protein concentration resembled changes that often accompany self-association. For Rubisco activase, high concentrations of protein in the chloroplast stroma would provide an environment conducive to self-association and cause expression of properties that would enin vivo. o 1992 hance its ability to function efficiently Academic
Press,
Inc.
The initial reaction in the assimilation of COZ by green plants is the carboxylation of ribulose l,&bisphosphate (RuBP),~ catalyzed by ribulose-l+bisphosphate carboxylase/oxygenase (Rubisco, EC 4.1.1.39). In photosynthesis, irradiance-dependent changes in the activation state of Rubisco play an important regulatory role by adjusting the rate of CO, fixation to the rate of electron transport activity (1). Rubisco activase, a chloroplastic regulatory protein (2), mediates changes in the Rubisco activation state by facilitating carbamylation of Rubisco (3). Carbamylation of a specific lysyl residue completes the binding site for the essential metal ion, making Rubisco competent for catalysis (4, 5). Rubisco activase affects carbamylation by promoting dissociation of tightly bound sugar-phosphates, including RuBP, from the active site of decarbamylated Rubisco (6, 7). When bound to decarbamylated Rubisco, RuBP prevents carbamylation (8) and its removal by Rubisco activase apparently converts Rubisco into a form that has a high affinity for carbamylation (3, 8). Rubisco activase also catalyzes a related reaction, the removal of 2carboxyarabinitol l-phosphate (CAlP) from the active site of Rubisco (9). Carboxyarabinitol l-phosphate is a naturally occurring transition state analogue that binds to carbamylated Rubisco, thereby inhibiting enzyme activity (10). Rubisco activase overcomes the inhibition by facilitating dissociation of CAlP from the Rubisco active site (9). 1 The investigation reported in this paper (No. 92-3-56) is in connection with a project of the Kentucky Agricultural Experiment Station. ’ Abbreviations used: RuBP, ribulose, 1,5-bisphosphate; Rubisco, ribulose-1,5-bisphosphate carboxylase/oxygenase; CAlP, 2-carboxyarabinitol l-phosphate; DTT, dithiothreitol; PEG, polyethylene glycol; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; FPLC, fast protein liquid chromatography.
688 All
Copyright 0 1992 rights of reproduction
0003~9861/92 $5.00 by Academic Press, Inc. in any form reserved.
SELF-ASSOCIATION
OF RUBISCO
The mechanism by which Rubisco activase interacts with Rubisco to enhance sugar-phosphate dissociation is unknown. Rubisco activase catalyzes ATP hydrolysis (ll), a reaction that apparently supplies the energy required for the activation process. However, the ATPase activity of Rubisco activase does not appear to be coupled tightly to Rubisco activation since the rate of ATP hydrolysis is not affected by the presence of Rubisco (11). A major area that has yet to be investigated concerns the nature of protein-protein interactions between Rubisco activase and Rubisco. Of particular interest is how the energy from ATP hydrolysis is transduced into changes in Rubisco conformation that decrease the binding affinity for sugar phosphates and increase the affinity for carbamylation. In the present study, polyethylene glycol 3350 (PEG), a solvent-excluding reagent (12-16), was used to perturb interactions in the Rubisco-Rubisco activase system. The results showed that PEG caused selfassociation of Rubisco activase, stimulating the activation and ATPase activities and enhancing the interaction of Rubisco activase with Rubisco. MATERIALS
AND
METHODS
Chemicals. Sodium [“Clbicarbonate was purchased from Amersham Corp.3 (Arlington Heights, IL). Ribulose 1,5-bisphosphate was synthesized enzymatically from ribose 5-phosphate as described by Horecker et al. (17). Carboxyarabinitol l-phosphate was synthesized from carboxyarabinitol bisphosphate as described in (18). Polyethylene glycol (PEG 3350), coupling enzymes, and other reagents were from Sigma Chemical Co. (St. Louis, MO). Rubisco activase was purified from spinach Enzyme purification. (Spinucia olerucea L.) and tobacco UVicotianu tubucum L. cv. KY-14) leaves as described by Holbrook et al. (19). The enzyme was stored at -80°C with 0.2 mM ATP. Rubisco was purified from tobacco leaves as described previously (20) and was stored at -80°C as a frozen ammonium sulfate suspension. Rubisco, purified from spinach leaves by the procedure of McCurry et al. (21), was a generous gift of Dr. R. Houtz (University of Kentucky). For assays, Rubisco was incubated in 100 mM Tricine-NaOH, pH 8.0, 10 mM MgCl,, 10 mM NaHCO,, and 50 mM dithiothreitol (DTT) for 1 h at room temperature. The enzyme was decarbamylated by removal of Mg2+ and COZ on a Sephadex G-50-80 column (0.8 X 3.5 cm) equilibrated with 20 mM Tricine-NaOH (pH 8.0) and 0.2 mM EDTA. Following desalting, decarbamylated Rubisco was incubated with 0.5 mM RuBP for 30 min at 23°C to convert Rubisco to the Rubisco-RuBP form (i.e., ER form (22)). The Rubisco-RuBP complex was stored at 4°C for up to 6 weeks without loss of activity upon reactivation. Rubisco complexed with CAlP was prepared by incubating carbamylated tobacco Rubisco with excess CAlP. The enzyme was desalted in 100 mM Tricine-NaOH, pH 8.0, 10 mM MgC12,lO mM NaHC03, and 0.5 mM DTT prior to its use in the assay. Enzyme assays. ATPase activity was assayed spectrophotometrically at 25’C by monitoring ADP production via NADH oxidation in 0.5-ml reactions containing 100 mM Tricine-NaOH, pH 8.0, 10 mM MgCl*, 2 mM DTT (buffer A), 20 mM KCl, 2 mM phosphoenolpyruvate, 2 mM DT’I’, 1.7 U of pyruvate kinase, 2.5 U lactate dehydrogenase, ATP at the concentrations indicated in the text, and 7.5 pg purified Rubisco
3 The mention of a trademark, proprietary product, or vendor does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture and does not imply its approval to the exclusion of other products or vendors that may also be suitable.
ACTIVASE:
POLYETHYLENE
GLYCOL
689
activase. One unit (U) of activity represents 1 wmol ATP hydrolyzed per minute. Rubisco activation activity was measured at 25°C in a two-stage assay similar to that in (6). Except where indicated in the text, the first-stage reaction contained buffer A, 10 mM NaHCO,, 1 mM ATP, 4 mM RuBP, 2 mM phosphocreatine, and 2.28 U phosphocreatine kinase. Immediately prior to assay, Rubisco activase was added and the reactions were initiated after precisely 30 s by adding decarbamylated Rubisco complexed with RuBP or carbamylated Rubisco complexed with CAlP. Aliquots (15-25 ~1) were taken 30 s and 5 min after the addition of Rubisco or at other times as indicated in the text and transferred to assays containing buffer A, 10 mM NaH’“C0, (1 Ci/mol), and 0.4 mM RuBP in a total volume of 0.5 ml for measurement of Rubisco activity. Rubisco assays were conducted for 30 s and terminated by the addition of 100 ~1 of 4 N formic acid in 1 N HCl. The assay mixtures were dried in uucuo and “C incorporation into acid-stable products was determined by liquid scintillation spectroscopy. The change in Rubisco activity that occurred after the 30 s incubation with Rubisco activase was used to determine the rate of activation. The rate was based on comparison with the activity of fully carbamylated Rubisco and is expressed as nmol Rubisco active sites activated/min. The activity of Rubisco determined after 5 min incubation with Rubisco activase represents the final extent of Rubisco activation expressed as Rubisco-specific activity in pmol CO2 fixed/min/ mg protein. Unless indicated otherwise in the text, Rubisco and Rubisco activase were from tobacco. Apparent molecular mass was deMolecular muss determinations. termined by rate zonal sedimentation of Rubisco activase (0.57 mg/ml) in lo-40% sucrose gradients (23) containing buffer A, 10 mM NaHCO,, and 5% (w/v) PEG as indicated. The positions of the molecular mass standards were determined by analyzing the composition of the individual fractions by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (23). The position of Rubisco activase was determined by SDS-PAGE and verified by measuring ATPase activity as described above. Gel-filtration chromatography of Rubisco activase (0.68 mg/ml) was performed at 4°C on a 30 X l-cm Superose-12 FPLC column in buffer A and 5% (w/v) PEG as indicated using a flow rate of 0.3 ml/ min. The positions of the molecular mass standards were determined from the A280profile. The position of Rubisco activase was determined from the Aw profile and was verified by measuring ATPase activity as described above. Rubisco activase protein was determined Miscellaneous techniques. by the method of Bradford (24) using serum albumin as a standard. Rubisco protein was determined spectrophotometrically (25). Sodium dodecyl sulfate-PAGE was performed in 10% mini-gels (23). Gels were stained with Coomassie blue R-250 to visualize the proteins.
RESULTS
Effect of PEG on Activation Actiuase
of Rub&o
by Rubisco
The salient features of Rubisco activation in the presence or absence of RuBP and the influence of Rubisco activase on the process are shown in Fig. 1. In the absence of RuBP, tobacco Rubisco activated spontaneously upon incubation with COZ and Mgzf, reflecting the rapid carbamylation of the enzyme (4). When RuBP was present, carbamylation of Rubisco was inhibited and the activation state remained low. Inclusion of tobacco Rubisco activase and ATP increased both the rate and the final level of activity achieved in the presence of RuBP. An identical effect for Rubisco activase and Rubisco from spinach (see (6, 8)) has been reported. At the subsaturating levels of Rubisco activase used here, the rate of activation was
MICHAEL
690
E. SALVUCCI
r
r
1.2
.Z >.= .E 02
0.0 Lc 2
0
6
4
Time
8
1012
(min)
FIG. 1. Time course of Rubisco activation by Rubisco activase in the presence or absence of PEG. Decarbamylated Rubisco (1.5 mg/ml) was incubated with 4 mM RuBP alone (0) or together with either 0.13 mg/ ml Rubisco activase (0) or Rubisco activase plus 10% (w/v) PEG-3350 (m). Decarbamylated Rubisco was incubated in the absence of RuBP and Rubisco activase (A) to measure spontaneous activation.
slower and the final level of activity was lower than those attained by spontaneous activation in the absence of RuBP. However, with higher Rubisco activase/Rubisco ratios, Rubisco was activated in the presence of RuBP to the level attained by spontaneous carbamylation in the absence of RuBP (data not shown, see below). The effect of PEG 3350 in the Rubisco-Rubisco activase system (Fig. 1) was examined. When included with tobacco Rubisco, RuBP, and tobacco Rubisco activase at 10% (w/v), PEG increased both the rate and the extent
1.2
a
I
I
I
I
I
I
I
I
0.0 t
I
0.0
I
0.1
[Rubisco
I
0.2
I
0.3
activase]
I
I
0.4
0.5
(mg/ml )
FIG. 3. The effect of PEG on Rubisco activation by varying concentrations of Rubisco activase. (A) The rate of Rubisco activation was measured in the presence (W) or absence (0) of 5% (w/v) PEG after a 30-s incubation of 1.5 mg/ml Rubisco with the indicated concentrations of Rubisco activase. (B) The extent of Rubisco activity was measured in the presence (m) or absence (0) of 5% (w/v) PEG after a 5-min incubation of 1.5 mg/ml Rubisco with the indicated concentrations of Rubisco activase.
1.0
.=
0.0 0
2
4
6
8
10
12
14
16
[PEG] (%) on Rubisco activation by FIG. 2. The effect of PEG concentration Rubisco activase. Decarbamylated Rubisco (1.5 mg/ml) was incubated with 4 mM RuBP alone (0) or together with 0.18 mg/ml Rubisco activase (0) at the indicated concentrations of PEG for 5 min. Decarbamylated Rubisco was incubated in the absence of RuBP and Rubisco activase (W) to measure spontaneous activation.
of Rubisco activation by approximately twofold (Fig. 1). Similar effects of PEG on Rubisco activation by Rubisco activase were observed with Rubisco and Rubisco activase from spinach (see below). The effect of PEG on Rubisco activation by Rubisco activase was concentration dependent from 0 to 5% (w/v) PEG (Fig. 2). In contrast, these concentrations of PEG had little effect on spontaneous carbamylation of Rubisco (i.e., without Rubisco activase) in either the presence or the absence of RuBP. The Rubisco activase reaction is second-order, dependent on the concentrations of Rubisco and Rubisco activase (6). Therefore, to investigate the effect of PEG on the Rubisco-Rubisco activase interaction, Rubisco activation at several concentrations of Rubisco activase and at a constant Rubisco concentration (Fig. 3) was examined. At 1.5 mg/ml Rubisco, both the rate of activation (Fig. 3A) and the extent, i.e., the level of Rubisco activity that was reached after 5 min reaction with Rubisco ac-
SELF-ASSOCIATION
OF RUBISCO
ACTIVASE:
POLYETHYLENE
GLYCOL
691
Rubisco, the extent of Rubisco activation decreased 2fold in the absence of PEG and nearly 3-fold in the presence of PEG (Fig. 4B). At each Rubisco concentration, Rubisco activity at a constant Rubisco activase concentration was greater in the presence of PEG. However, at 0.2 mg/ml Rubisco, the difference was 2.7-fold compared to only 1.7-fold at 3 mg/ml. Effect of PEG on Dissociation of CAlP from Rub&o the Presence of Rub&o Activase
in
The tight binding inhibitor, CAlP, caused marked inhibition of Rubisco activity (Table I, see also (10)). Rubisco activity increased 2.5fold when the Rubisco-CAlP complex was incubated with either Rubisco activase or RuBP. An even greater increase in Rubisco activity occurred when the complex was incubated with both Rubisco activase and RuBP (see also Ref. (9)). The effectiveness of Rubisco activase in relieving inhibition was enhanced in the presence of PEG. The stimulation of Rubisco activase activity by PEG occurred in the presence or absence of RuBP. However, there was little effect of PEG on Rubisco activity in the absence of Rubisco activase.
0.0 0.0
I
I
I
I
I
I
0.5
1.0
1.5
2.0
2.5
3.0
[Rubiscol
(mglml
Effect of PEG on the ATPase Activity Activase 3.5
)
FIG. 4. The effect of PEG on Rubisco activation at varying concentrations of Rubisco and a constant Rubisco activase concentration. (A) The rate of Rubisco activation was measured in the presence (m) or absence (Cl) of 5% (w/v) PEG after a 30-s incubation of Rubisco, at the indicated concentrations, with 0.065 mg/ml Rubisco activase. (B) The extent of Rubisco activity was measured in the presence (m) or absence (0) of 5% (w/v) PEG after a 5-min incubation of Rub&o, at the indicated concentrations, with 0.065 mg/ml Rubisco activase.
The effect of PEG on the ATPase activity of Rubisco activase in response to ATP concentration (Fig. 5) was examined. The presence of 5% PEG 3350 in the assay caused marked stimulation of ATPase activity, increasing the Max from 0.94 to 1.34 U/mg and decreasing the apparent K,(ATP) from 0.45 to 0.15 mM. An increase in ATPase activity also occurred with a higher molecular
TABLE
tivase (Fig. 3B), increased with increasing Rubisco activase concentrations. The extent of Rubisco activation increased up to a concentration of about 0.2 mg/ml Rubisco activase. When 5% PEG was included in the reactions, the shapes of the response curves were similar, but the rate of activation was faster and the final level of Rubisco activity was higher. With varying concentrations of Rubisco and a constant Rubisco activase concentration of 0.065 mg/ml, the rate of Rubisco activation by Rubisco activase increased with increasing Rubisco concentrations (Fig. 4A). The shape of the response curve was similar in the presence or absence of PEG, but the rate of activation was considerably faster in the presence of PEG. The greatest differences were observed at low Rubisco concentrations where the rata of activation was 4-fold higher in the presence of PEG compared to just 2-fold at higher Rubisco concentrations. In response to increasing concentrations of
of Rub&o
I
Effect of PEG on Inhibition of Rubisco Activity by 2-Carboxyarabinitol l-phosphate (CAlP) in the Presence of Rubisco Activase” Rubisco activity (U/mg protein) Conditions Fully inhibited* (minus RuBP, minus Rubisco activase) Plus RuBP Plus Rubisco activase Complete (plus Rubisco activase, plus RuBP)
-PEG
+5%
PEG
0.09
0.10
0.23 0.24
0.26 0.51
0.96
1.23
’ Rubisco activity was measured after a P-min incubation of 0.73 mg/ ml carbamylated Rubisco-CAlP in the presence or absence of Rubisco activase (0.2 mg/ml) and RuBP (4 mM) as described under Materials and Methods. * The activity of fully activated Rubisco was 1.3 U/mg protein in the absence of CAlP.
692
MICHAEL
E. SALVUCCI
activase at all activase concentrations tested, but the largest increase occurred at the lowest Rubisco activase concentration. For example, the specific activity was 3fold greater in the presence of PEG at 6 pg/ml compared to only 1.2-fold at 150 pg/ml. Effect of PEG on the Apparent Molecular Rub&o Activase
a
0.4
0.2 4
-8
0
4
8
12
118
0.0
~,
0.0
,
,
,
,
0.5
,
,
,
,
1.0
,
,
,
,
,
,
1.5
,
,
,
,
,
,
2.0
[ATPI HW FIG. 5. The effect of PEG on the response of Rubisco activase ATPase activity to ATP concentration. ATPase activity was measured in the presence (m) or absence (0) of 5% (w/v) PEG at the indicated concentrations of ATP. The concentration of Rubisco activase was 15 r&/ml. Inset: Double reciprocal plot of the data.
weight PEG, PEG 20,000, as well as with Ficoll400, polyvinylpyrrolidone-40, and bovine serum albumin (data not shown). Two experiments showed that the stimulatory effect of PEG on the ATPase activity of Rubisco activase required the presence of PEG in the assay. First, when Rubisco activase was preincubated with PEG and then diluted loo-fold into an assay without PEG, there was no effect on ATPase activity (data not shown). Second, when PEG was added during the course of the reaction, the rate of ATPase activity increased immediately from 0.7 to 1.2 U/mg protein. For comparison, the specific ATPase activity of Rubisco activase in assays containing PEG at the start of the reaction was 1.4 U/mg protein. Thus, it appears that PEG affects Rubisco activase both by changing the enzyme to a more active form and by stabilizing the more active form upon dilution into the assay. It should be noted that control experiments with known amounts of ADP showed that PEG had no effect on the ADP-dependent rate of NADH oxidation by the coupling enzymes at concentrations up to 10% (w/v). Effect of ATPase Specific Activity Concentration
on Rub&o
Activase
The specific activity of the ATPase, measured with saturating ATP concentrations, was dependent on the concentration of Rubisco activase (data not shown). Specific activity increased nearly lo-fold when the Rubisco activase concentration was increased from 6 to 150 pg/ ml. PEG increased the specific ATPase activity of Rubisco
Mass of
Polyethylene glycol can promote self-association of proteins (13, 14, 16). A typical feature of proteins that self-associate and exhibit increased activity in the oligomerit state is the response of specific activity to protein concentration (described above and in (26)). To provide physical evidence for self-association, the apparent molecular mass of Rubisco activase was determined by rate zonal sedimentation in sucrose gradients in the presence or absence of PEG (Fig. 6A). In the absence of PEG, the molecular mass of native Rubisco activase was estimated at 58,000, a value similar to that obtained by gel-filtration chromatography during purification on a Sephacryl S300 column (data not shown). In the presence of PEG, the position of Rubisco activase in the sucrose gradients was shifted relative to the positions of the molecular mass standards. Based on the position in the gradient, the apparent molecular mass of Rubisco activase was 217,000 in the presence of PEG. Polyethylene glycol also caused a change in the relative position of Rubisco activase on a Superose-12 FPLC gel-filtration column (Fig. 6B). The shift caused by the presence of 5% (w/v) PEG in the column and buffers corresponded to a doubling of the apparent molecular mass of Rubisco activase from 280,000 to 517,000. Effect of PEG on the Substrate Specificity Activase
of Rubisco
Wang et al. (27) have reported that Rubisco activase exhibits specificity for the Rubisco substrate. For example, Rubisco activase from spinach was unable to activate Rubisco from tobacco and vice versa. In the present study, the effect of PEG on activation of spinach Rubisco by Rubisco activase from spinach and tobacco was examined (Table II). Similar to the results with the tobacco enzymes (Figs. l-4), both the rate and the extent of activation of spinach Rubisco by spinach Rubisco activase were greater in the presence of PEG. Polyethylene glycol also affected activation of spinach Rubisco by tobacco Rubisco activase. In the absence of PEG, there was no detectable activation of spinach Rubisco by Rubisco activase from tobacco. However, in reactions containing PEG, tobacco Rubisco activase increased both the rate and the extent of activation of spinach Rubisco (Table II, Fig. 7). Mixing experiments showed that activation of spinach Rubisco by spinach Rubisco activase was inhibited by tobacco Rubisco activase (Table II). Inclusion of PEG partially overcame the inhibition, improving activation of spinach
SELF-ASSOCIATION
OF RUBISCO
ACTIVASE:
POLYETHYLENE
GLYCOL
693
Rubisco activase, spinach Rubisco activity was two- to threefold higher in the presence of PEG. The rate of activation of spinach Rubisco increased markedly with increasing tobacco Rubisco activase concentrations in the presence of PEG (Fig. 7B). The rate increased linearly about 3-fold from 0.03 to 0.5 mg/ml. In contrast, without PEG, the rate of activation at the highest tobacco Rubisco activase concentration was only 1.5-fold higher than the rate at the lowest Rubisco activase concentration. These rates were marginally higher than the rate without Rubisco activase (Table II). 0
2
4
8
6
Volume
10
(ml )
0.6
0.5
>
0.4
Ym 0.3
0.2
0.1
4
5
Log Molecular
6
Mass
FIG. 6. The effect of PEG on the apparent molecular mass of Rubisco activase estimated by rate zonal sedimentation (A) and gel-filtration chromatography (B). Zonal sedimentation in sucrose gradients and gelfiltration chromatography on a Superose-12 FPLC column were performed in the presence (w) or absence (0) of 5% (w/v) PEG at 0.57 and 0.68 mg/ml Rubisco activase, respectively. The position of Rubisco activase (0, 0) is indicated by an arrow on the regression line relative to the position of standard proteins. The standard proteins were cytochrome c, 12,400; carbonic anhydrase, 29,000; bovine serum albumin, 66,000; alcohol dehydrogenase, 150,000; P-amylase; 200,000; Rubisco, 530,000. K*” = (elution volume - void volume)/(total volume - void volume).
DISCUSSION
The effect of PEG and other inert polymers on protein self-association has been described in thermodynamic terms as a consequence of the excluded volume interaction between PEG and the protein (13-15). Inert polymers like PEG are solvent-excluding reagents that increase the fractional volume occupied by proteins. The change in excluded volume drives protein self-association by increasing the activity coefficient of protein monomers in equilibrium with aggregates (28). For example, PEG has been shown to promote self-association of the pyruvate dehydrogenase complex (29) and heteromerous interactions among citric acid cycle enzymes (30) and glycolytic enzymes (16), and between glycolytic enzymes and F-actin (16). At high concentrations, PEG drives the formation of higher order aggregates and interacts with hydrophobic domains as proteins denature (14, 15). Ultimately, proteins precipitate from solution as a consequence of these effects, hence the familiar use of PEG as a protein precipitating agent. In the present study, the apparent molecular mass of Rubisco activase increased in the presence of 5% (w/v) PEG, indicating that PEG promoted self-association of Rubisco activase. At this concentration of PEG, the effect on protein self-association appeared to be specific since Rubisco activase was the only protein examined whose
TABLE
Rubisco by the mixture of spinach and tobacco Rubisco activases. The stimulatory effect of PEG on activation of spinach Rubisco by tobacco Rubisco activase was analyzed further by measuring activation of spinach Rubisco with varying concentrations of tobacco Rubisco activase (Fig. 7). In the presence of PEG, the extent of activation increased with increasing Rubisco activase concentrations from 0.01 to 0.0625 mg/ml, but was relatively constant above this level (Fig. 7A). In contrast, in the absence of PEG, the specific activity of spinach Rubisco was not affected significantly by changes in tobacco Rubisco activase concentration up to 0.5 mg/ml. Above 0.0625 mg/ml tobacco
Effect
II
of PEG on the Substrate of Rubisco Activase”
Rate of activation (nmol sites/min) Source of Rubisco activase Spinach Tobacco Spinach + tobacco No activase
-PEG 0.04
OF BIOCHEMISTRY
Vol. 298, No. 2, November
AND
BIOPHYSICS
1, pp. 688-696, 1992
Subunit Interactions of Rubisco Activase: Polyethylene Glycol Promotes Self-Association, Stimulates ATPase and Activation Activities, and Enhances Interactions with Rubisco’ Michael
E. Salvucci
U.S. Department of Agriculture, Agricultural Research Service and Agronomy Department, University of Kentucky, Lexington, Kentucky 40546-0076
Received May 13, 1992, and in revised form July 7, 1992
The effect of polyethylene glycol (PEG) on the enzymatic and physical properties of ribulose- 1,5-bisphosphate carboxylasefoxygenase (Rubisco) activase was examined. In the presence of PEG, Rubisco activase exhibited higher ATPase and Rubisco activating activities, concomitant with increased apparent affinity for ATP and Rubisco. Specific ATPase activity, which was dependent on Rubisco activase concentration, was also higher in the presence of Ficoll, polyvinylpyrrolidone, and bovine serum albumin. The ability of Rubisco activase to facilitate dissociation of the tight-binding inhibitor 2-carboxyarabinitol l-phosphate from carbamylated Rubisco was also enhanced in the presence of PEG. Mixing experiments with Rubisco activase from two different sources showed that tobacco Rubisco activase, which exhibited little activation of spinach Rubisco by itself, was inhibitory when included with spinach Rubisco activase. Polyethylene glycol improved the ability of tobacco and a mixture of tobacco plus spinach Rubisco activase to activate spinach Rubisco. Estimates based on rate zonal sedimentation and gel-filtration chromatography indicated that the apparent molecular mass of Rubisco activase was two- to fourfold higher in the presence of PEG. The increase in apparent molecular mass was consistent with the propensity of solvent-excluding reagents like PEG to promote self-association of proteins. Likewise, the change in enzymatic properties of Rubisco activase in the presence of PEG and the dependence of specific activity on protein concentration resembled changes that often accompany self-association. For Rubisco activase, high concentrations of protein in the chloroplast stroma would provide an environment conducive to self-association and cause expression of properties that would enin vivo. o 1992 hance its ability to function efficiently Academic
Press,
Inc.
The initial reaction in the assimilation of COZ by green plants is the carboxylation of ribulose l,&bisphosphate (RuBP),~ catalyzed by ribulose-l+bisphosphate carboxylase/oxygenase (Rubisco, EC 4.1.1.39). In photosynthesis, irradiance-dependent changes in the activation state of Rubisco play an important regulatory role by adjusting the rate of CO, fixation to the rate of electron transport activity (1). Rubisco activase, a chloroplastic regulatory protein (2), mediates changes in the Rubisco activation state by facilitating carbamylation of Rubisco (3). Carbamylation of a specific lysyl residue completes the binding site for the essential metal ion, making Rubisco competent for catalysis (4, 5). Rubisco activase affects carbamylation by promoting dissociation of tightly bound sugar-phosphates, including RuBP, from the active site of decarbamylated Rubisco (6, 7). When bound to decarbamylated Rubisco, RuBP prevents carbamylation (8) and its removal by Rubisco activase apparently converts Rubisco into a form that has a high affinity for carbamylation (3, 8). Rubisco activase also catalyzes a related reaction, the removal of 2carboxyarabinitol l-phosphate (CAlP) from the active site of Rubisco (9). Carboxyarabinitol l-phosphate is a naturally occurring transition state analogue that binds to carbamylated Rubisco, thereby inhibiting enzyme activity (10). Rubisco activase overcomes the inhibition by facilitating dissociation of CAlP from the Rubisco active site (9). 1 The investigation reported in this paper (No. 92-3-56) is in connection with a project of the Kentucky Agricultural Experiment Station. ’ Abbreviations used: RuBP, ribulose, 1,5-bisphosphate; Rubisco, ribulose-1,5-bisphosphate carboxylase/oxygenase; CAlP, 2-carboxyarabinitol l-phosphate; DTT, dithiothreitol; PEG, polyethylene glycol; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; FPLC, fast protein liquid chromatography.
688 All
Copyright 0 1992 rights of reproduction
0003~9861/92 $5.00 by Academic Press, Inc. in any form reserved.
SELF-ASSOCIATION
OF RUBISCO
The mechanism by which Rubisco activase interacts with Rubisco to enhance sugar-phosphate dissociation is unknown. Rubisco activase catalyzes ATP hydrolysis (ll), a reaction that apparently supplies the energy required for the activation process. However, the ATPase activity of Rubisco activase does not appear to be coupled tightly to Rubisco activation since the rate of ATP hydrolysis is not affected by the presence of Rubisco (11). A major area that has yet to be investigated concerns the nature of protein-protein interactions between Rubisco activase and Rubisco. Of particular interest is how the energy from ATP hydrolysis is transduced into changes in Rubisco conformation that decrease the binding affinity for sugar phosphates and increase the affinity for carbamylation. In the present study, polyethylene glycol 3350 (PEG), a solvent-excluding reagent (12-16), was used to perturb interactions in the Rubisco-Rubisco activase system. The results showed that PEG caused selfassociation of Rubisco activase, stimulating the activation and ATPase activities and enhancing the interaction of Rubisco activase with Rubisco. MATERIALS
AND
METHODS
Chemicals. Sodium [“Clbicarbonate was purchased from Amersham Corp.3 (Arlington Heights, IL). Ribulose 1,5-bisphosphate was synthesized enzymatically from ribose 5-phosphate as described by Horecker et al. (17). Carboxyarabinitol l-phosphate was synthesized from carboxyarabinitol bisphosphate as described in (18). Polyethylene glycol (PEG 3350), coupling enzymes, and other reagents were from Sigma Chemical Co. (St. Louis, MO). Rubisco activase was purified from spinach Enzyme purification. (Spinucia olerucea L.) and tobacco UVicotianu tubucum L. cv. KY-14) leaves as described by Holbrook et al. (19). The enzyme was stored at -80°C with 0.2 mM ATP. Rubisco was purified from tobacco leaves as described previously (20) and was stored at -80°C as a frozen ammonium sulfate suspension. Rubisco, purified from spinach leaves by the procedure of McCurry et al. (21), was a generous gift of Dr. R. Houtz (University of Kentucky). For assays, Rubisco was incubated in 100 mM Tricine-NaOH, pH 8.0, 10 mM MgCl,, 10 mM NaHCO,, and 50 mM dithiothreitol (DTT) for 1 h at room temperature. The enzyme was decarbamylated by removal of Mg2+ and COZ on a Sephadex G-50-80 column (0.8 X 3.5 cm) equilibrated with 20 mM Tricine-NaOH (pH 8.0) and 0.2 mM EDTA. Following desalting, decarbamylated Rubisco was incubated with 0.5 mM RuBP for 30 min at 23°C to convert Rubisco to the Rubisco-RuBP form (i.e., ER form (22)). The Rubisco-RuBP complex was stored at 4°C for up to 6 weeks without loss of activity upon reactivation. Rubisco complexed with CAlP was prepared by incubating carbamylated tobacco Rubisco with excess CAlP. The enzyme was desalted in 100 mM Tricine-NaOH, pH 8.0, 10 mM MgC12,lO mM NaHC03, and 0.5 mM DTT prior to its use in the assay. Enzyme assays. ATPase activity was assayed spectrophotometrically at 25’C by monitoring ADP production via NADH oxidation in 0.5-ml reactions containing 100 mM Tricine-NaOH, pH 8.0, 10 mM MgCl*, 2 mM DTT (buffer A), 20 mM KCl, 2 mM phosphoenolpyruvate, 2 mM DT’I’, 1.7 U of pyruvate kinase, 2.5 U lactate dehydrogenase, ATP at the concentrations indicated in the text, and 7.5 pg purified Rubisco
3 The mention of a trademark, proprietary product, or vendor does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture and does not imply its approval to the exclusion of other products or vendors that may also be suitable.
ACTIVASE:
POLYETHYLENE
GLYCOL
689
activase. One unit (U) of activity represents 1 wmol ATP hydrolyzed per minute. Rubisco activation activity was measured at 25°C in a two-stage assay similar to that in (6). Except where indicated in the text, the first-stage reaction contained buffer A, 10 mM NaHCO,, 1 mM ATP, 4 mM RuBP, 2 mM phosphocreatine, and 2.28 U phosphocreatine kinase. Immediately prior to assay, Rubisco activase was added and the reactions were initiated after precisely 30 s by adding decarbamylated Rubisco complexed with RuBP or carbamylated Rubisco complexed with CAlP. Aliquots (15-25 ~1) were taken 30 s and 5 min after the addition of Rubisco or at other times as indicated in the text and transferred to assays containing buffer A, 10 mM NaH’“C0, (1 Ci/mol), and 0.4 mM RuBP in a total volume of 0.5 ml for measurement of Rubisco activity. Rubisco assays were conducted for 30 s and terminated by the addition of 100 ~1 of 4 N formic acid in 1 N HCl. The assay mixtures were dried in uucuo and “C incorporation into acid-stable products was determined by liquid scintillation spectroscopy. The change in Rubisco activity that occurred after the 30 s incubation with Rubisco activase was used to determine the rate of activation. The rate was based on comparison with the activity of fully carbamylated Rubisco and is expressed as nmol Rubisco active sites activated/min. The activity of Rubisco determined after 5 min incubation with Rubisco activase represents the final extent of Rubisco activation expressed as Rubisco-specific activity in pmol CO2 fixed/min/ mg protein. Unless indicated otherwise in the text, Rubisco and Rubisco activase were from tobacco. Apparent molecular mass was deMolecular muss determinations. termined by rate zonal sedimentation of Rubisco activase (0.57 mg/ml) in lo-40% sucrose gradients (23) containing buffer A, 10 mM NaHCO,, and 5% (w/v) PEG as indicated. The positions of the molecular mass standards were determined by analyzing the composition of the individual fractions by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (23). The position of Rubisco activase was determined by SDS-PAGE and verified by measuring ATPase activity as described above. Gel-filtration chromatography of Rubisco activase (0.68 mg/ml) was performed at 4°C on a 30 X l-cm Superose-12 FPLC column in buffer A and 5% (w/v) PEG as indicated using a flow rate of 0.3 ml/ min. The positions of the molecular mass standards were determined from the A280profile. The position of Rubisco activase was determined from the Aw profile and was verified by measuring ATPase activity as described above. Rubisco activase protein was determined Miscellaneous techniques. by the method of Bradford (24) using serum albumin as a standard. Rubisco protein was determined spectrophotometrically (25). Sodium dodecyl sulfate-PAGE was performed in 10% mini-gels (23). Gels were stained with Coomassie blue R-250 to visualize the proteins.
RESULTS
Effect of PEG on Activation Actiuase
of Rub&o
by Rubisco
The salient features of Rubisco activation in the presence or absence of RuBP and the influence of Rubisco activase on the process are shown in Fig. 1. In the absence of RuBP, tobacco Rubisco activated spontaneously upon incubation with COZ and Mgzf, reflecting the rapid carbamylation of the enzyme (4). When RuBP was present, carbamylation of Rubisco was inhibited and the activation state remained low. Inclusion of tobacco Rubisco activase and ATP increased both the rate and the final level of activity achieved in the presence of RuBP. An identical effect for Rubisco activase and Rubisco from spinach (see (6, 8)) has been reported. At the subsaturating levels of Rubisco activase used here, the rate of activation was
MICHAEL
690
E. SALVUCCI
r
r
1.2
.Z >.= .E 02
0.0 Lc 2
0
6
4
Time
8
1012
(min)
FIG. 1. Time course of Rubisco activation by Rubisco activase in the presence or absence of PEG. Decarbamylated Rubisco (1.5 mg/ml) was incubated with 4 mM RuBP alone (0) or together with either 0.13 mg/ ml Rubisco activase (0) or Rubisco activase plus 10% (w/v) PEG-3350 (m). Decarbamylated Rubisco was incubated in the absence of RuBP and Rubisco activase (A) to measure spontaneous activation.
slower and the final level of activity was lower than those attained by spontaneous activation in the absence of RuBP. However, with higher Rubisco activase/Rubisco ratios, Rubisco was activated in the presence of RuBP to the level attained by spontaneous carbamylation in the absence of RuBP (data not shown, see below). The effect of PEG 3350 in the Rubisco-Rubisco activase system (Fig. 1) was examined. When included with tobacco Rubisco, RuBP, and tobacco Rubisco activase at 10% (w/v), PEG increased both the rate and the extent
1.2
a
I
I
I
I
I
I
I
I
0.0 t
I
0.0
I
0.1
[Rubisco
I
0.2
I
0.3
activase]
I
I
0.4
0.5
(mg/ml )
FIG. 3. The effect of PEG on Rubisco activation by varying concentrations of Rubisco activase. (A) The rate of Rubisco activation was measured in the presence (W) or absence (0) of 5% (w/v) PEG after a 30-s incubation of 1.5 mg/ml Rubisco with the indicated concentrations of Rubisco activase. (B) The extent of Rubisco activity was measured in the presence (m) or absence (0) of 5% (w/v) PEG after a 5-min incubation of 1.5 mg/ml Rubisco with the indicated concentrations of Rubisco activase.
1.0
.=
0.0 0
2
4
6
8
10
12
14
16
[PEG] (%) on Rubisco activation by FIG. 2. The effect of PEG concentration Rubisco activase. Decarbamylated Rubisco (1.5 mg/ml) was incubated with 4 mM RuBP alone (0) or together with 0.18 mg/ml Rubisco activase (0) at the indicated concentrations of PEG for 5 min. Decarbamylated Rubisco was incubated in the absence of RuBP and Rubisco activase (W) to measure spontaneous activation.
of Rubisco activation by approximately twofold (Fig. 1). Similar effects of PEG on Rubisco activation by Rubisco activase were observed with Rubisco and Rubisco activase from spinach (see below). The effect of PEG on Rubisco activation by Rubisco activase was concentration dependent from 0 to 5% (w/v) PEG (Fig. 2). In contrast, these concentrations of PEG had little effect on spontaneous carbamylation of Rubisco (i.e., without Rubisco activase) in either the presence or the absence of RuBP. The Rubisco activase reaction is second-order, dependent on the concentrations of Rubisco and Rubisco activase (6). Therefore, to investigate the effect of PEG on the Rubisco-Rubisco activase interaction, Rubisco activation at several concentrations of Rubisco activase and at a constant Rubisco concentration (Fig. 3) was examined. At 1.5 mg/ml Rubisco, both the rate of activation (Fig. 3A) and the extent, i.e., the level of Rubisco activity that was reached after 5 min reaction with Rubisco ac-
SELF-ASSOCIATION
OF RUBISCO
ACTIVASE:
POLYETHYLENE
GLYCOL
691
Rubisco, the extent of Rubisco activation decreased 2fold in the absence of PEG and nearly 3-fold in the presence of PEG (Fig. 4B). At each Rubisco concentration, Rubisco activity at a constant Rubisco activase concentration was greater in the presence of PEG. However, at 0.2 mg/ml Rubisco, the difference was 2.7-fold compared to only 1.7-fold at 3 mg/ml. Effect of PEG on Dissociation of CAlP from Rub&o the Presence of Rub&o Activase
in
The tight binding inhibitor, CAlP, caused marked inhibition of Rubisco activity (Table I, see also (10)). Rubisco activity increased 2.5fold when the Rubisco-CAlP complex was incubated with either Rubisco activase or RuBP. An even greater increase in Rubisco activity occurred when the complex was incubated with both Rubisco activase and RuBP (see also Ref. (9)). The effectiveness of Rubisco activase in relieving inhibition was enhanced in the presence of PEG. The stimulation of Rubisco activase activity by PEG occurred in the presence or absence of RuBP. However, there was little effect of PEG on Rubisco activity in the absence of Rubisco activase.
0.0 0.0
I
I
I
I
I
I
0.5
1.0
1.5
2.0
2.5
3.0
[Rubiscol
(mglml
Effect of PEG on the ATPase Activity Activase 3.5
)
FIG. 4. The effect of PEG on Rubisco activation at varying concentrations of Rubisco and a constant Rubisco activase concentration. (A) The rate of Rubisco activation was measured in the presence (m) or absence (Cl) of 5% (w/v) PEG after a 30-s incubation of Rubisco, at the indicated concentrations, with 0.065 mg/ml Rubisco activase. (B) The extent of Rubisco activity was measured in the presence (m) or absence (0) of 5% (w/v) PEG after a 5-min incubation of Rub&o, at the indicated concentrations, with 0.065 mg/ml Rubisco activase.
The effect of PEG on the ATPase activity of Rubisco activase in response to ATP concentration (Fig. 5) was examined. The presence of 5% PEG 3350 in the assay caused marked stimulation of ATPase activity, increasing the Max from 0.94 to 1.34 U/mg and decreasing the apparent K,(ATP) from 0.45 to 0.15 mM. An increase in ATPase activity also occurred with a higher molecular
TABLE
tivase (Fig. 3B), increased with increasing Rubisco activase concentrations. The extent of Rubisco activation increased up to a concentration of about 0.2 mg/ml Rubisco activase. When 5% PEG was included in the reactions, the shapes of the response curves were similar, but the rate of activation was faster and the final level of Rubisco activity was higher. With varying concentrations of Rubisco and a constant Rubisco activase concentration of 0.065 mg/ml, the rate of Rubisco activation by Rubisco activase increased with increasing Rubisco concentrations (Fig. 4A). The shape of the response curve was similar in the presence or absence of PEG, but the rate of activation was considerably faster in the presence of PEG. The greatest differences were observed at low Rubisco concentrations where the rata of activation was 4-fold higher in the presence of PEG compared to just 2-fold at higher Rubisco concentrations. In response to increasing concentrations of
of Rub&o
I
Effect of PEG on Inhibition of Rubisco Activity by 2-Carboxyarabinitol l-phosphate (CAlP) in the Presence of Rubisco Activase” Rubisco activity (U/mg protein) Conditions Fully inhibited* (minus RuBP, minus Rubisco activase) Plus RuBP Plus Rubisco activase Complete (plus Rubisco activase, plus RuBP)
-PEG
+5%
PEG
0.09
0.10
0.23 0.24
0.26 0.51
0.96
1.23
’ Rubisco activity was measured after a P-min incubation of 0.73 mg/ ml carbamylated Rubisco-CAlP in the presence or absence of Rubisco activase (0.2 mg/ml) and RuBP (4 mM) as described under Materials and Methods. * The activity of fully activated Rubisco was 1.3 U/mg protein in the absence of CAlP.
692
MICHAEL
E. SALVUCCI
activase at all activase concentrations tested, but the largest increase occurred at the lowest Rubisco activase concentration. For example, the specific activity was 3fold greater in the presence of PEG at 6 pg/ml compared to only 1.2-fold at 150 pg/ml. Effect of PEG on the Apparent Molecular Rub&o Activase
a
0.4
0.2 4
-8
0
4
8
12
118
0.0
~,
0.0
,
,
,
,
0.5
,
,
,
,
1.0
,
,
,
,
,
,
1.5
,
,
,
,
,
,
2.0
[ATPI HW FIG. 5. The effect of PEG on the response of Rubisco activase ATPase activity to ATP concentration. ATPase activity was measured in the presence (m) or absence (0) of 5% (w/v) PEG at the indicated concentrations of ATP. The concentration of Rubisco activase was 15 r&/ml. Inset: Double reciprocal plot of the data.
weight PEG, PEG 20,000, as well as with Ficoll400, polyvinylpyrrolidone-40, and bovine serum albumin (data not shown). Two experiments showed that the stimulatory effect of PEG on the ATPase activity of Rubisco activase required the presence of PEG in the assay. First, when Rubisco activase was preincubated with PEG and then diluted loo-fold into an assay without PEG, there was no effect on ATPase activity (data not shown). Second, when PEG was added during the course of the reaction, the rate of ATPase activity increased immediately from 0.7 to 1.2 U/mg protein. For comparison, the specific ATPase activity of Rubisco activase in assays containing PEG at the start of the reaction was 1.4 U/mg protein. Thus, it appears that PEG affects Rubisco activase both by changing the enzyme to a more active form and by stabilizing the more active form upon dilution into the assay. It should be noted that control experiments with known amounts of ADP showed that PEG had no effect on the ADP-dependent rate of NADH oxidation by the coupling enzymes at concentrations up to 10% (w/v). Effect of ATPase Specific Activity Concentration
on Rub&o
Activase
The specific activity of the ATPase, measured with saturating ATP concentrations, was dependent on the concentration of Rubisco activase (data not shown). Specific activity increased nearly lo-fold when the Rubisco activase concentration was increased from 6 to 150 pg/ ml. PEG increased the specific ATPase activity of Rubisco
Mass of
Polyethylene glycol can promote self-association of proteins (13, 14, 16). A typical feature of proteins that self-associate and exhibit increased activity in the oligomerit state is the response of specific activity to protein concentration (described above and in (26)). To provide physical evidence for self-association, the apparent molecular mass of Rubisco activase was determined by rate zonal sedimentation in sucrose gradients in the presence or absence of PEG (Fig. 6A). In the absence of PEG, the molecular mass of native Rubisco activase was estimated at 58,000, a value similar to that obtained by gel-filtration chromatography during purification on a Sephacryl S300 column (data not shown). In the presence of PEG, the position of Rubisco activase in the sucrose gradients was shifted relative to the positions of the molecular mass standards. Based on the position in the gradient, the apparent molecular mass of Rubisco activase was 217,000 in the presence of PEG. Polyethylene glycol also caused a change in the relative position of Rubisco activase on a Superose-12 FPLC gel-filtration column (Fig. 6B). The shift caused by the presence of 5% (w/v) PEG in the column and buffers corresponded to a doubling of the apparent molecular mass of Rubisco activase from 280,000 to 517,000. Effect of PEG on the Substrate Specificity Activase
of Rubisco
Wang et al. (27) have reported that Rubisco activase exhibits specificity for the Rubisco substrate. For example, Rubisco activase from spinach was unable to activate Rubisco from tobacco and vice versa. In the present study, the effect of PEG on activation of spinach Rubisco by Rubisco activase from spinach and tobacco was examined (Table II). Similar to the results with the tobacco enzymes (Figs. l-4), both the rate and the extent of activation of spinach Rubisco by spinach Rubisco activase were greater in the presence of PEG. Polyethylene glycol also affected activation of spinach Rubisco by tobacco Rubisco activase. In the absence of PEG, there was no detectable activation of spinach Rubisco by Rubisco activase from tobacco. However, in reactions containing PEG, tobacco Rubisco activase increased both the rate and the extent of activation of spinach Rubisco (Table II, Fig. 7). Mixing experiments showed that activation of spinach Rubisco by spinach Rubisco activase was inhibited by tobacco Rubisco activase (Table II). Inclusion of PEG partially overcame the inhibition, improving activation of spinach
SELF-ASSOCIATION
OF RUBISCO
ACTIVASE:
POLYETHYLENE
GLYCOL
693
Rubisco activase, spinach Rubisco activity was two- to threefold higher in the presence of PEG. The rate of activation of spinach Rubisco increased markedly with increasing tobacco Rubisco activase concentrations in the presence of PEG (Fig. 7B). The rate increased linearly about 3-fold from 0.03 to 0.5 mg/ml. In contrast, without PEG, the rate of activation at the highest tobacco Rubisco activase concentration was only 1.5-fold higher than the rate at the lowest Rubisco activase concentration. These rates were marginally higher than the rate without Rubisco activase (Table II). 0
2
4
8
6
Volume
10
(ml )
0.6
0.5
>
0.4
Ym 0.3
0.2
0.1
4
5
Log Molecular
6
Mass
FIG. 6. The effect of PEG on the apparent molecular mass of Rubisco activase estimated by rate zonal sedimentation (A) and gel-filtration chromatography (B). Zonal sedimentation in sucrose gradients and gelfiltration chromatography on a Superose-12 FPLC column were performed in the presence (w) or absence (0) of 5% (w/v) PEG at 0.57 and 0.68 mg/ml Rubisco activase, respectively. The position of Rubisco activase (0, 0) is indicated by an arrow on the regression line relative to the position of standard proteins. The standard proteins were cytochrome c, 12,400; carbonic anhydrase, 29,000; bovine serum albumin, 66,000; alcohol dehydrogenase, 150,000; P-amylase; 200,000; Rubisco, 530,000. K*” = (elution volume - void volume)/(total volume - void volume).
DISCUSSION
The effect of PEG and other inert polymers on protein self-association has been described in thermodynamic terms as a consequence of the excluded volume interaction between PEG and the protein (13-15). Inert polymers like PEG are solvent-excluding reagents that increase the fractional volume occupied by proteins. The change in excluded volume drives protein self-association by increasing the activity coefficient of protein monomers in equilibrium with aggregates (28). For example, PEG has been shown to promote self-association of the pyruvate dehydrogenase complex (29) and heteromerous interactions among citric acid cycle enzymes (30) and glycolytic enzymes (16), and between glycolytic enzymes and F-actin (16). At high concentrations, PEG drives the formation of higher order aggregates and interacts with hydrophobic domains as proteins denature (14, 15). Ultimately, proteins precipitate from solution as a consequence of these effects, hence the familiar use of PEG as a protein precipitating agent. In the present study, the apparent molecular mass of Rubisco activase increased in the presence of 5% (w/v) PEG, indicating that PEG promoted self-association of Rubisco activase. At this concentration of PEG, the effect on protein self-association appeared to be specific since Rubisco activase was the only protein examined whose
TABLE
Rubisco by the mixture of spinach and tobacco Rubisco activases. The stimulatory effect of PEG on activation of spinach Rubisco by tobacco Rubisco activase was analyzed further by measuring activation of spinach Rubisco with varying concentrations of tobacco Rubisco activase (Fig. 7). In the presence of PEG, the extent of activation increased with increasing Rubisco activase concentrations from 0.01 to 0.0625 mg/ml, but was relatively constant above this level (Fig. 7A). In contrast, in the absence of PEG, the specific activity of spinach Rubisco was not affected significantly by changes in tobacco Rubisco activase concentration up to 0.5 mg/ml. Above 0.0625 mg/ml tobacco
Effect
II
of PEG on the Substrate of Rubisco Activase”
Rate of activation (nmol sites/min) Source of Rubisco activase Spinach Tobacco Spinach + tobacco No activase
-PEG 0.04
