Plant Physiol. (1996) 11 1 : 195-202
The Role of Magnesium, Pyrophosphate, and Their Complexes as Substrates and Activators of the Vacuolar H+-Pumping lnorganic Pyrophosphatase' Studies Using Ligand Protection from Covalent lnhibitors Ruth Gordon-Weeks*, Susan H. Steele, and Roger A. Leigh
Biochemistry and Physiology Department, IACR-Rothamsted, Harpenden, Hertfordshire AL5 2JQ, United Kingdom although severa1 identical subunits may interact during H+ pumping (Chanson and Pilet, 1989; Maeshima, 1990; Sato et al., 1991; Sarafian et al., 199213). Studies of the structurefunction relations of the H+-PPase are now underway (Kim et al., 1995), and progress will require knowledge of the number of different ligands that bind to the enzyme so that the functional domains involved in catalysis, transport, and regulation can be located. The H+-PPase requires PPi, Mg2+, and K+ for maximum activity (Walker and Leigh, 1981; Wang et al., 1986; White et al., 1990). The available evidence indicates that t h e enzyme uses a complex of Mg2+ and PPi as its substrate, and earlier studies suggested that this was MgPPi (Walker and Leigh, 1981; Johannes and Felle, 1989; White et al., 1990). However, kinetic modeling has indicated that it may be Mg,PPi, with a K , in the range of 2 to 5 PM (Leigh et al., 1992; Rea et al., 1992a; Baykov et al., 1993). In addition, both kinetic and covalent inhibitor studies (Maeshima, 1991; Leigh et al., 1992; Baykov et al., 1993) indicate the presence of a high-affinity Mg2+ binding site with a K , in the range of 25 to 42 FM, but a second site with a K , of 0.25 to 0.46 mM was suggested by the kinetic modeling undertaken by Baykov et al. (1993). In this paper we have determined the binding constants for Mg,PPi and Mg2+ by making use of the observation that the H+-PPase is sensitive to inhibitors that react preferentially with certain amino acids and that inhibition is less when mixtures of Mg2+ and PPi are present in the incubation medium (Britten et al., 1989; Kuo and Pan, 1990; Baykov et al., 1993; Zhen et al., 1994). The results support the conclusion that Mg,PPi is the substrate and that there is a high-affinity Mgz+ binding site (Leigh et al., 1992;Baykov et al., 1993). Both sites bind their ligands with the K , values predicted by the model of Leigh et al. (1992).
lnhibitors preferentially and covalently reactive with cysteine, arginine, histidine, and carboxyl-containing residues were inhibitory to the plant vacuolar H+-transporting inorganic pyrophosphatase (H+-PPase) from Vigna radiafa (mung bean) and Befa vulgaris (red beet), but hydrophobic compounds and those reactive with tyrosine and lysine were less effective. lnhibition by 1-ethyl3-(3-dimethylaminopropyl)carbodiimide,phenylglyoxal, and N-ethylmaleimide was decreased i n the presence of Mgz+ or mixtures of Mg2+ and inorganic pyrophosphate (PPi) but not by PPi alone. None of these ligands affected inhibition by reagents reactive with histidine. The Mg2+ dependence of protection from 1-ethyl-3-(3-dimethylaminopropy1)carbodiimide inhibition followed first-order kinetics and yielded a K, for free Mg2+ of 20 t o 23 p ~ Protection . from inhibition by N-ethylmaleimide and phenylglyoxal varied as a function of Mg,PPi concentration, suggesting that this i s the substrate for the H+-PPase. Protection by Mg,PPi followed MichaelisMenten kinetics w i t h a K, of approximately 2 PM. These results are consistent with the predictions of a kinetic model for the H+-PPase (R.A. Leigh, A.J. Pope, I.R. Jennings, D. Sanders [1992] Plant Physiol 100: 1698-1 750), which identified free Mg2+ as an allosteric activator (K, = 25 PM) and Mg,PPi as the substrate (K, = 2.5-5 PM).
The tonoplast contains two H+ pumps responsible for acidifying the vacuolar sap (Rea and Sanders, 1987).One is a vacuolar-type ATPase (Sze et al., 1992), the other an inorganic pyrophosphatase (H+-PPase; Rea and Poole, 1993; Leigh et al., 1994). Isolation and cloning of cDNAs encoding the major catalytic subunit of the H+-PPase (Sarafian et al., 1992a; Tanaka et al., 1993; Kim et al., 199413) has enabled progress to be made in the molecular characterization of the enzyme. Using yeast, it has been shown that expression of the cDNA encoding the substrate-binding polypeptide of the Arabidopsis H+-PPase is sufficient for a11 of the enzyme's known catalytic functions (Kim et al., 1994a), implying that it contains only a single subunit,
Abbreviations: BTP, bis-Tris-propane; EDAC, l-ethyl-3-(3-dimethylaminopropy1)carbodiimide;H+-PPase; vacuolar H+-transporting inorganic pyrophosphatase; Kapp, first-order rate constant measured in the presence of ligands; Kobs, pseudo-first-order rate constant; NBDCI, 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole; NEM, N-ethylmaleimide; PGO, phenylglyoxal.
' IACR-Rothamsted receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom. * Corresponding author; e-mail
[email protected]; fax 44-1582-760981. 195
Gordon-Weeks et al.
196 MATERIALS AND METHODS Plant Material and lsolation of Tonoplast Vesicles
Red beet (Beta vulgaris L. cv Detroit Crimson Globe) was grown in a glasshouse, and storage roots were harvested immediately before use. Seeds of mung bean ( V i g n a radiata L.) were germinated and grown for 5 d in the dark at 25°C on water-saturated vermiculite, and hypocotyls were harvested on the day of vesicle isolation. Tonoplast vesicles were isolated from red beet storage roots using the method of Rea and Poole (1985). Mung bean tonoplast vesicles were prepared according to the method of Rea et al. (1992b). Vesicles were stored at -80°C in the resuspension medium appropriate to the preparation method used. These media contained components (e.g. DTT, EDTA) that potentially could interfere with the subsequent inhibitor studies. Therefore, on the day of the inhibition experiment, the vesicles were centrifuged at 80,OOOg for 35 min and resuspended in 250 mM glycerol containing 20 mM Hepes-BTP, pH 8.0.
Preincubation with lnhibitors
Vesicles were preincubated with inhibitors before the hydrolytic activity of the H+-PPase was measured. The standard preincubation mixture (total volume 50 pL) contained inhibitor, 20 mM Hepes-BTP, pH 8.0, and 3 mM EGTA (to ensure that there was no free Ca2+ in the medium; Rea et al., 1992a). When mixtures of MgSO, and PPi-BTP were included in the preincubation medium, the total concentrations needed to give the required concentrations of free Mg”, MgPPi, or Mg,PPi were calculated using the computer program SOLCON (written by D.C.S. White, University of York, UK, and Y.E. Goldman, University of Pennsylvania) and the stability constants listed by Leigh et al. (1992). These calculations made appropriate allowance for the binding of Mg2+ to the EGTA present in the media. The preincubation was at 0°C when NEM or N-phenylmaleimide were used and at 25°C for other inhibitors. It was started by the addition of vesicles (25 pL, 6-24 pg of protein) and was terminated either by the addition of 5 pL of 30 mM DTT (Cys-reactive inhibitors) or by dilution directly into the H+-PPase assay mixture (for other inhibitors). The effectiveness of the latter procedure was confirmed by appropriate experiments. Routinely, three 15-pL samples from each preincubation were used in the subsequent assay of H+-PPase hydrolytic activity. When using the labile reagent diethylpyrocarbonate, its concentration was calculated prior to incubation by the method of Pelton and Ganzhorn (1992). Incubation mixtures containing Rose Bengal (4,5,6,7-tetrachloro-2’,4’,5’,7’tetraiodofluorescein) were prepared in the dark, and the inhibitor was activated by a 45-s exposure to a 300-W spotlight, and H+-PPase activity was assayed immediately in the dark. Fluorescein 5’-isothiocyanate and N,N’-dicyclohexylcarbodiimide were dissolved in DMSO before further dilution in water. Control experiments using DMSO alone showed that the solvent did not affect the activity of the H+-PPase.
Plant Physiol. Vol. 1 1 1 , 1996
Measurement of H+-PPase Activity
Hydrolytic activity of the H+-PPase was determined as described by Leigh et al. (1992), except that 10 pg mL-’ L-a-lysophosphatidylcholine and 300 pg mL-l Triton X-100 were included in the assay media to disrupt vesicles and maximize hydrolytic activity. Final concentrations of MgSO, and PPi-BTP were 1.5 and 0.3 mM, respectively. Appropriate compensation was made for the amounts of Mg2+ or PPi carried over from the inhibitor incubation media. Incubation was at 25°C for 1 h and activity was a linear function of time over this period. Typical mung bean tonoplast preparations had a H+-PPase specific activity of about 20 pmol PPi hydrolyzed (mg protein)-’ h-l (range 15-32), whereas the equivalent value for red beet tonoplast was 9 pmol PPi hydrolyzed (mg protein)-l h-l (range 6-12). Protein Assay
Protein was measured by the method of Appleroth and Angsten (1987) with BSA as a standard. RESULTS The Effects of Covalent lnhibitors
A variety of reagents preferentially reactive with particular amino acids were tested for their ability to inhibit the hydrolytic activity of the red beet and mung bean H+PPases. His- and Cys-reactive reagents were found to produce full inhibition of the enzyme at low concentrations (Table I). Inhibition by mersalyl was rapidly reversed (1 min) in the presence of 5 mM DTT, confirming that it reacted with Cys residues. The effects of NBDCl were probably also due to its reactivity with thiol groups rather than with Tyr (Cantley et al., 1978),since full activity couId be recovered by incubation of the NBDC1-inhibited enzyme with mercaptoethanol (I%, w/v). The lack of effect of other Tyr-reactive reagents supports this conclusion (Table I). Lys-reactive reagents were only weak inhibitors and the hydrophilic Arg-reactive inhibitor 2,3-butanedione gave only partia1 inhibition even when present at a concentration of 50 mM. The other Arg-reactive reagent PGO was more effective, although high concentrations (>30 mM) were required for maximal effectiveness. The enzyme displayed different sensitivities to the two carboxyl-reactive inhibitors EDAC and N,N’-dicyclohexylcarbodiimide,with EDAC being more inhibitory. Previous studies of the inhibition of the mung bean and red beet H’-PPase by NEM indicated that the reaction follows pseudo-first-order kinetics, indicating that modification of a single residue is sufficient for inhibition (Britten et al., 1989; Zhen et al., 1994). A similar analysis for the reaction of EDAC with the mung bean enzyme showed that this inhibitor also gives pseudo-first-order kinetics (Fig. 1A). A double-logarithmic plot of Kobs against EDAC concentration yielded a straight line with a slope of 0.86 5 0.009 (mean -+ SE of three experiments), indicating that modification of a single residue is probably responsible for inhibition (Fig. 1B). Inhibition by PGO also followed
Ligand Protection Studies of the Vacuolar H+-Transporting lnorganic Pyrophosphatase psuedo- first-order kinetics (Fig. 2A), and a slope of 2.00 5 0.046 (mean t SE of four experiments) was obtained from the double-logarithmic plot of Kobs against PGO concentration (Fig. 2B), suggesting modification of more than one residue.
Protection by Free Mgz+ and Complexes of Mg and PPi Neither free PPi, free Mg2+, nor mixtures of the two provided any protection against inhibition by the Hisreactive reagents diethylpyrocarbonate and Rose Bengal (Table 11). Pyrophosphate alone provided no protection against any of the inhibitors and actually enhanced the degree of inhibition by NEM, EDAC, and PGO (not shown; see also Britten et al., 1989). Free Mg2+ alone provided some degree of protection against inhibition by NEM, Nphenylmaleimide, PGO, and NBDCl, and was as effective as a mixture of Mg2+ and PPi at protecting from inhibition by EDAC. However, the mixture was more effective than Mg2+ alone in protecting the H+-PPase against the other inhibitors (Table 11),indicating that a complex of Mg2+ and PPi was providing some additional protection. The ability of free Mg2+ to protect from EDAC inhibition was exploited to measure the binding constant for Mg2+. When protection of the mung bean H+-PPase from inhibition by EDAC was measured as a function of free Mg2+ concentration, Michaelis-Menten kinetics were obtained (Fig. 3) yielding a K , for Mg2+ of 23.3 2 2.7 /.LM (mean 'I SE of three experiments). This result was checked by measuring the effects of free Mg2+ concentration on the initial rate of EDAC inhibition. Free Mg2' in the preincubation medium reduced the measured rate constant (Kapp) of the inhibition reaction (Fig. 4A). Assuming the following relationship,
(where Kobs and K,,, are the rate constants measured in the absence and presence of free Mg2', respectively), a plot of
197
1/ Kapp against Mg2" concentration yielded a K, for free Mg2+ of 20 5 1.7 p~ (mean 2 SE of three experiments; Fig. 48). Estimates of the K , for Mg2+ were also made from the ability of free Mg2+ to protect from inhibition by PGO and NEM. The experiments with PGO gave a K , for Mg2+ of 31.0 % 6.6 p~ (mean 2 SE of three experiments), whereas the corresponding value from the NEM experiments was 27.6 -t 3.9 PM (mean t SE of four experiments). The complex of Mg2+ and PPi responsible for additional protection against NEM and PGO inhibition was identified by determining the effects of known concentrations of free Mg2+, MgPPi, and Mg,PPi. Vesicles were preincubated with inhibitor in the presence of various concentrations of free Mg2+ at a series of fixed concentrations of Mg,PPi. AS expected (see Table TI), free Mg2+ alone gave some protection against both inhibitors, but the degree of protection increased with increasing Mg,PPi concentration (Fig. 5). The degree of protection was not related to the concentration of MgPPi in the solutions because the concentration of this complex decreased as the protection increased (see Fig. 5A). These results indicate that Mg,PPi is the complex that protects the enzyme and therefore is likely to be the substrate of the H+-PPase. The effect of Mg,PPi concentration on protection from PGO inhibition was used to obtain a value for the K, for Mg,PPi binding. Protection was plotted as a function of Mg,PPi concentration at fixed Mg2+ concentrations between 0.05 and 0.6 mM. The effect of free Mg2+ alone was subtracted from the total protection to obtain the protection attributable only to Mg,PPi (Fig. 6). The response conformed to Michaelis-Menten kinetics, yielding a K, for Mg,PPi of 1.9 5 0.15 /.LM (mean 5 SE averaged over a11 Mg2+ concentrations used in three experiments). These results were supported by an analysis of the effects of increasing Mg,PPi concentration on the initial rate of PGO inhibition. At a concentration of 0.6 mM free Mg2+, the presence of Mg,PPi in the incubation medium reduced the measured rate constant of the inhibition reaction (Fig. 7A).
Table 1. lnhibitor sensitivities of H+-PPase activity associated with red beet and mung bean tonoplast vesicles Values are means 5 SE of three separate experiments. N-PheNEM, N-Phenylmaleimide; BD, 2,3-butanedione; DEPC, diethylpyrocarbonate; Acl, N-acetylimidazole; FLTE, fluorescein 5' isothiocyanate; TNM, tetranitromethane. lnhibitor
lncubation Conditions
Target Amino Acid/Functional Group
lnhibition Red beet
Mung bean %
NEM NPheNEM Mersalyl PGO BD DEPC Rose Bengal Acl FlTC NBDCI TNM DCCD EDAC a
NT, Not tested.
40 p ~ 10 , min 40 p ~ 10 , min 50 p ~ 10 , min 50 mM, 30 min 50 mM, 30 min 10 mM, 10 min 60 p ~ ,1 min 2 mM, 30 min 2 mM, 30 min 300 p ~ 20 , min 20 mM, 20 min 20 mM, 20 rnin 20 mM, 20 min
93 i 2 9s t 2 89 ? 2 82 i 2 22 t 3 93 2 s 1O 0 O
90 t 8 92 L 3 95 t 1 95 t 3 43 i 2 1O0 NTa O
17 ? 4 1O 0
48 i 3
O
NT 42 t 6
13 i 1 94 ? 2
1O 0
88
?
3
Gordon-Weeks et al.
198
Plant Physiol. Vol. 11 1, 1996
Mgz+ concentrations that could be achieved. In contrast, for NEM, Mg,PPi-dependent protection increased sigmoidally as free Mg2+ increased, suggesting positive cooperative kinetics. A Hill plot confirmed this and yielded a Hill coefficient of 2.3 (Fig. 8, inset). O 10
1
O
I
0.1 1 O
R
DISCUSSION
\
The results in this paper support the conclusion that Mg,PPi is the substrate of the H+-PPase, as suggested by kinetic modeling of the response of the enzyme to changes in Mg2+ and PPi concentrations (Leigh et al., 1992; Rea et al., 1992a; Baykov et al., 1993). This complex protects the H+-PPase from inhibition by both NEM and PGO, whereas high concentrations of MgPPi provide no protection, and protection increases as the concentration of MgPPi decreases (Fig. 5 ) . Further, the K , for Mg,PPi, determined from the kinetics of protection from inhibi-
i
I
I
I
I
I
4
8
12
16
20
Time (min) Log [EDAC] (M) -3.5
-3.0
-2.5
-2.0
I
I
I
-1.5 -0.5
B -1.o 2
9 o)
O
1
-1.5
-2.0 Figure 1. Time-dependent kinetics of inhibition of the mung bean H+-PPase by EDAC. A, Semilogarithmic plot of activity (percent control) versus time. Vesicles were preincubated with O (O),2 (A),5 (W), or 10 (7)mM EDAC in the presence of 20 mM Hepes-BTP, pH 8.0, and 3 mM EGTA in a total volume of 50 pL. Reactions were terminated at the times indicated by the addition of 15 p L of the above mixture directly to the H+-PPase assay mixture. The experiment was performed in triplicate and a typical result is shown. B, Double logarithmic plot of Kobs,the pseudo-first-order rate constant (derived from the slopes of the lines i n A) against EDAC concentration. The fitted line has the equation: log Kobs = 0.86 log [EDAC] + 0.69. This graph includes concentrations of EDAC that, to aid clarity, are omitted from A.
An analysis similar to that described above (see Fig. 4) yielded a K, for Mg,PPi of 2.8 ? 0.45 PM (mean ? SE of three experiments; Fig. 7B). Closer inspection of the way protection by Mg,PPi varied as a function of free Mg2+ concentration (Fig. 5) indicated that there were differences in the responses obtained with NEM and PGO. This is illustrated more clearly in Figure 8, where the Mg2’ dependence of protection by 20 PM Mg,PPi is plotted for both inhibitors. Over the range of free Mg2+ concentrations used, the ability of Mg,PPi to protect from PGO inhibition was similar at low and high Mg” concentrations. These data could be fitted to a Michaelis-Menten function with a K , for free Mg2+ of 18 ~ L M ,although there were no experimental points below 50 ~ L Mfree Mg2+ because of restrictions on the range of free
I
Figure 2. Time-dependent kinetics of inhibition of mung bean HtPPase by PCO. A, Semilogarithmic plot of activity (percent control) versus time. Experimental conditions were as for Figure 1 except that vesicles were preincubated with O (O),15 (A),25 (m), or 35 (V)mM PCO. The experiment was performed in triplicate and a typical result i s shown. B, Double logarithmic plot of Kobr, the first-order rate constant (derived from the slopes of the lines in A) against PGO concentration. The fitted line has the equation: log KObs= 2.0 log [PCO] + 1.89. This graph includes concentrations of PGO that, to aid clarity, are omitted from A.
Ligand Protection Studies of the Vacuolar H+-Transporting lnorganic Pyrophosphatase
199
Table II. Effect of ligands on the inhibition of the red beet and mung bean tonoplast H+-PPase activities b y covalent inhibitors Experimental conditions and abbreviations are as in Table I. Values are means -C- SE of three separate experiments. DEPC, Diethylpyrocarbonate. Protection ~~
lnhibitor
Red beet Mg2+
Mg2+/PPi
Mung bean PPi
Mg2+
Mg2+/PPi
PPi
%
NEM (40 p M ) PCO (30 mM) DEPC (10 p ~ ) Rose Bengal (60 p ~ ) NBDCl (300 p ~ ) EDAC (20 mM) a
80 2 2 71 5 3
O O
43 f 8 1924
59 t 10 53 2 5
O
35 t 5 O O
O O
O
53 ? 2 80 ? 2
76 2 4 84 t 5
O
O NT" NT
O NT NT
O NT NT
O
82 f 1
60 f 1
O
O
88 2 4
O
NT, Not tested.
tion by PGO, is 1.9 to 2.8 PM, a value that agrees well with those (2-5 PM) determined from kinetic modeling (Leigh et al., 1992; Rea et al., 1992a; Baykov et al., 1993). Therefore, it seems likely that the predictions of these models are correct both in the identity of the substrate and in its affinity for the H+-PPase. The kinetic model of Leigh et al. (1992) suggested that free Mg2+ activated the H+-PPase by binding to a site with a K , for free Mg*+ of 25 PM. Maeshima (1991) and Baykov et al. (1993) also found evidence for such a high-affinity Mg2+ binding site, whereas the kinetic modeling by Baykov et al. (1993) suggested a second, low-affinity site with a K , in the range of 0.25 to 0.46 mM. The results in the present paper are consistent with the presence of two Mg2+
.c 7 .c
2
O
l B
= 60
8
2
20
Time (min)
80
.b > ._
15
10
5
I
40
20
O 0.0
-20 0.1
0.2
0.3
0.4
0.5
0.6
Free Mg2' concentration (mM)
Figure 3. The effect of free Mg2+ concentration on the protection of the mung bean H+-PPase from inhibition by EDAC. The preincubation reaction was performed as described in Figure 1 except that the solution contained 1O mM EDAC and the indicated concentration of free Mg2+. The symbols show the experimental results and the line indicates a Michaelis-Menten relationship with a K, for free Mg2+ of 23 p~ and a V,,, (maximal protection) of 82%.
-10
o
10
20
Free Mg" concentration (pM) Figure 4. The effect of free Mg2+ concentration on the rate of EDAC inhibition of the mung bean H+-PPase. A, Vesicles were incubated with 10 mM EDAC under the conditions described i n Figure 1 but in the presence of O (O),5 10 (A),or 20 (V)p~ free Mg2+. The experiment was performed in triplicate and a typical result is shown. 6, Plot of reciproca1 of Kapp(derived from the slopes in A) against free Mg2+ concentrations. The intercept on the x axis yields a K, for free Mg2+ of 20 p ~ .
(m),
Gordon-Weeks et al.
200
1O0
L
IA
2000
f a
v
S
80
1500
!
c S
60
1000
8
.-o
40 500
20
O
a a
r"
O
0.0
0.2
0.4
0.6
Free Mg2+concentration (mM)
B
-0 0.0
0.2
0.4
0.6
Plant Physiol. Vol. 11 1 , 1996
(Forsén and Linse, 1995). It is not possible from the analysis of the protection responses to say which of the sites is the one involved in Mg2' activation of the H+-PPase, and this must now be a target for future studies. The two sites suggested by the experiments reported here could result from the different temperatures at which different inhibitors were used, e.g. the experiments with NEM were conducted at 0°C and those with other inhibitors were conducted at 25°C. However, this explanation is unlikely, since experiments using NEM to measure the K, for free Mg2+ at 0°C gave a value comparable to that measured at 25OC with EDAC (Figs. 3 and 4), PGO (see "Results"), and NEM (Baykov et al., 1993). This suggests that the differences between NEM and PGO in Figure 8 are not due to temperature, and that the second site we have identified is real and distinct from the high-affinity site that seems to have the same K, whether measured at O or 25°C. Nonetheless, it would now be interesting to determine whether the second site showing cooperative kinetics has temperature-dependent properties. This study supports earlier findings that Cys, Arg, and carboxyl residues may be present in the vicinity of domains involved in the binding of substrate and/or free Mg2+ (Britten et al., 1989; Kuo and Pan, 1990; Romero and Celis, 1992; Zhen et al., 1994). The location of the residues sensitive to PGO and EDAC remains to be established, but substrate-protectable NEM inhibition is due to binding to Cys"', which is located on a cytoplasmic loop between membrane-spanning helices X and XI (Sarafian et al.,
40
Free Mg2' concentration (mM) Figure 5. The effects of combinations of free Mg2+ and Mg,PPi on protection of the mung bean H+-PPase from inhibition by 20 FM NEM (A) and 25 mM PGO (B). The preincubation conditions were as described i n "Materials and Methods" except that the solutions contained O (O),1 (U),5 (A),20 (V),or 30 ( + ) p~ Mg,PPi at each of the free Mg'+ concentrations indicated. The amounts of MgSO, and PPi-BTP required to obtain the final concentrations of free MgZ+ and Mg,PPi were calculated using SOLCON. The dashed lines in A show the concentrations of MgPPi for the 5, 20, and 30 FM Mg,PPi treatments. Results are the mean of three to seven experiments.
binding sites of differing affinities. Thus, the kinetics of the ability of free Mg2' to protect from inhibition by EDAC (Figs. 3 and 4), PGO, and NEM (see text), and the Mg2+ dependence of Mg,PPi protection from PGO inhibition (Fig. 8) both provide evidence for a Mg2+ binding site with . the Mg2+ a K , of approximately 25 p ~ Additionally, dependence of Mg,PPi protection from NEM inhibition suggests but does not provide conclusive evidence for a second site having cooperative kinetics, with half-maximal stimulation of protection by Mg,PPi occurring at about 0.2 mM Mg2+ (Fig. 8). The cooperative kinetics for Mg2+ binding to this site could result from the interaction of severa1 H+-PPase subunits during catalysis (Chanson and Pilet, 1989; Sato et al., 1991; Sarafian et al., 1992b), although single-subunit enzymes may also show cooperativity
O
! 0'
5
10
15
20
25
30
35
Mg,PPi concentration (pM)
Figure 6. The effect of Mg,PPi concentration on the protection of the mung bean H*-PPase from inhibition by 25 mM PCO. The Mg,PPidependent protection was calculated from the data in Figure 56 by subtracting the protection obtained in the absence of Mg,PPi from that measured with Mg,PPi at the free MgZt concentrations of 0.05 (O),0.1 (U),0.3 (A),or 0.6 (V)mM. The symbols show the experimental results and the line indicates a Michaelis-Menten relationship with a K, for Mg,PPi of 1.9 FM and a V,,, (maximal protection) of 33%.
Ligand Protection Studies of the Vacuolar H+-Transpqrting lnorganic Pyrophosphatase 1992a; Zhen et al., 1994). This loop also contains a pair of adjacent Arg residues, three Glu residues, and three Asp residues (two of which are adjacent) but no His residues (Sarafian et al., 1992a). Therefore, it would be interesting to determine whether the substrate- and Mg*+-protectable sensitivity of the H+-PPase to NEM, EDAC, and PGO all result from modification of residues in this loop. The lack of His residues may explain why the effects of His-reactive reagents are not modified by the presence of substrate or Mg2+ (Table 11). It is interesting that this is the only cytoplasmic loop that contains this combination of residues, although there are two other cytoplasmic loops that contain Arg, Glu, and Asp residues, but His is also present in these loops, whereas Cys is present in only one of them (Sarafian et al., 1992a; Tanaka et a]., 1993; Kim et al., 199413; Zhen et al., 1994). The loop containing Cys634also contains Tyr and Lys residues, but the H+-PPase was not greatly affected by reagents reactive with these amino acids (Table I). However, the reagents used are generally more reactive
201
40 35
-
E
30
c o
25
c
O
:20
c
c
U a,
5
n
15
a,
10 5
O
t
Log Free [Mg2+ 1(mM)
7 Free Mg '+ concentration (mM)
1 O0 1
A
50 -
Figure 8. The effect of free Mg2+ concentration on the Mg,PPidependent protection of the mung bean H+-PPase from inhibition by NEM (O) or PGO (A).The results are from the 20 p~ Mg,PPi treatments in Figure 5, and the Mg,PPi-dependent protection was calculated by subtracting protection measured in the treatments without Mg,PPi from that measured with 20 p~ Mg,PP,. The dashed line shows a Michaelis-Menten relationship with a K, of 18 p~ and a V,,, (maximal protection) of 36% fitted to the data for PGO. N o relationship was fitted to the NEM data. The inset shows a Hill plot transformation of the NEM data, which yields a Hill coefficient of 2.3.
with these residues when they are in a hydrophobic environment (e.g. Tzeng et al., 1992), and thus they may not have been effective against residues in a hydrophilic cytoplasmic loop. The residues involved in Mg2'-protectable inhibition by EDAC could also be located on other cytoplasmic loops, since a11 but one contain acidic residues (Sarafian et al., 1992a). In particular, the loop between amino acids 216 and 328 contains 5 Glu and 11 Asp residues, giving it a large net negative charge that could be involved in Mg2+ binding. Although this analysis ignores the effects of conformational or steric interactions resulting from the folding of the catalytic subunit, it nonetheless provides a hypothesis that can be tested by site-directed mutagenesis and labeling studies to identify the reactive amino acids. Received October 30, 1995; accepted February 2, 1996. Copyright Clearance Center: 0032-0889/96/ 111/0195/08.
LITERATURE ClTED
Appleroth KJ, Angsten H (1987) An improvement of the protein determination in plant tissues with the dye binding method according to Bradford. Biochem Physiol Pflanzen 182: 85-89
Baykov A, Bakuleva NP, Rea PA (1993) Steady-state kinetics of substrate hydrolysis by vacuolar H+-pyrophosphatase. A simple three state model. Eur J Biochem 217: 755-762
Gordon-Weeks et al.
202
Britten CJ, Turner JC, Rea PA (1989) Identification and purification of substrate-binding subunit of higher plant H+-translocating inorganic pyrophosphatase. FEBS Lett 256 200-206 Cantley LC Jr, Gelles J, Josephson L (1978) Reaction of (Na-K) evidence for ATPase with 7-chloro-4-nitrobenzo-2-oxa-1,3-diazole: an essential tyrosine at the active site. Biochemistry 17:419-425 Chanson A, Pilet PE (1989) Target molecular size and sodium dodecyl sulphate-polyacrylamide gel electrophoresis analysis of the ATP- and pyrophosphate-dependent proton pumps from maize root tonoplast. Plant Physiol 90: 934-938 Forsén S , Linse S (1995) Cooperativity: over the Hill. Trends Biochem Sci 20: 495-497 Johannes E, Felle HH (1989). The role of Mg2+in proton transport by the tonoplast pyrophosphatase in Riccia fluitans vacuoles. Physiol Plant 77: 326-331 Kim E], Zhen R-G, Rea PA (1994a) Heterologous expression of plant vacuolar pyrophosphatase in yeast demonstrates sufficiency of the substrate-binding subunit for proton transport. Proc Natl Acad Sci USA 91: 6128-6132 Kim E], Zhen R-G, Rea PA (1995) Site-directed mutagenesis of vacuolar H+-pyrophosphatase:necessity of for inhibition by maleimides but not catalysis. J Biol Chem 270 2630-2635 Kim Y, Kim E], Rea PA (199413) Isolation and characterization of cDNAs encoding the vacuolar H+-pyrophosphatase of Beta vulgaris. Plant Physiol 106: 37.5382 Kuo SK, Pan RL (1990) An essential arginyl residue in the tonoplast pyrophosphatase from etiolated mung bean seedlings. Plant Physiol93: 1128-1133 Leigh RA, Gordon-Weeks R, Steele SH, Koren'kov VD (1994) The H+-pumping pyrophosphatase of the vacuolar membrane of higher plants. In MR Blatt, RA Leigh, D Sanders, eds, Membrane Transport in Plants and Fungi. Company of Biologists, Cambridge, UK, pp 61-75 Leigh RA, Pope AJ, Jennings IR, Sanders D (1992) Kinetics of the vacuolar H+-pyrophosphatase from higher plants. The role of magnesium, pyrophosphate, and their complexes as substrates, activators, and inhibitors. Plant Physiol 100: 1698-1750 Maeshima M (1990) Oligmeric structure of H+-translocating inorganic pyrophosphatase of plant vacuoles. Biochem Biophys Res Commun 168: 1157-1162 Maeshima M (1991) H+-translocatin inor ganic . . J? yrop hosp hatase of plant vacuoles. Inhibition by Ca ,stabilization by Mg2+ and immunological comparison with other pyrophosphatases. Eur J Biochem 196: 11-17 Pelton PD, Ganzhorn A] (1992) The effect of histidine modification on the activity of myo-inositol monophosphatase from bovine brain. J Biol Chem 267: 5916-5920 Rea PA, Britten CJ, Jennings IR, Calvert CM, Skiera L, Leigh RA, Sanders D (1992a) Regulation of vacuolar H+-pyrophosphatase
g+
Plant Physiol. Vol. 111, 1996
by free calcium. A reaction kinetic analysis. Plant Physiol 100: 1706-1715 Rea PA, Britten CJ, Sarafian, V (1992b) Common identity of substrate-binding subunit of vacuolar H+-translocating inorganic pyrophosphatase of higher plant cells. Plant Physiol 1 0 0 723-732 Rea PA, Poole RJ (1985) Proton-translocating inorganic pyrophosphatase of red beet (Beta vulgaris L.) tonoplast vesicles. Plant Physiol77: 46-52 Rea PA, Poole RJ (1993) Vacuolar Hf-translocating pyrophosphatase. Annu Rev Plant Physiol Plant Mo1 Biol 4 4 157-180 Rea PA, Sanders D (1987)Tonoplast energisation: two pumps, one membrane. Physiol Plant 71: 131-141 Romero I, Celis H (1992) Evidence of an essential carboxyl residue in membrane bound pyrophosphatase of Rhodospirillum rubrum. J Bioenerg Biomembr 2 4 617-624 Sarafian V, Kim Y, Poole RJ, Rea PA (1992a) Molecular cloning and sequence of cDNA encoding the pyrophosphate-energised vacuolar membrane proton pump of Arabidopsis thaliana. Proc Natl Acad Sci USA 89: 1775-1779 Sarafian V, Potier M, Poole RJ (199213) Radiation-inactivation analysis of vacuolar H+-ATPase and H+-pyrophosphatase from Beta vulgaris L. Functional size for substrate hydrolysis and for Hf transport. Biochem J 283: 493-497 Sato MH, Maeshima M, Ohsumi Y, Yoshida M (1991) Dimeric structure of H+-translocating pyrophosphatase from pumpkin vacuolar membranes. FEBS Lett 290: 177-180 Sze H, Ward JM, Lai S (1992) Vacuolar H+-translocatingATPases from plants: structure, function and isoforms. J. Bioenerg Biomembr 24: 371-381 Tanaka Y, Chiba K, Maeda M, Maeshima M (1993) Molecular cloning of cDNA for vacuolar membrane proton translocating inorganic pyrophosphatase in Hordeum vulgare. Biochem Biophys Res Commun 190: 1110-1114 Tzeng CM, Hsu LH, Pan RL (1992) Inhibition of tonoplast ATPase from etiolated mung bean seedlings by fluorescein 5'-isothiocyanate. Biochem J 285: 737-743 Walker RR, Leigh RA (1981) Mg"-dependent, cation-stimulated inorganic pyrophosphatase associated with vacuoles isolated from storage roots of red beet (Beta vulgaris). Planta 153: 150-155 Wang Y, Leigh RA, Kaestner KH, Sze H (1986) Electrogenic Hf pumping pyrophosphatase in tonoplast vesicles of oat roots. Plant Physiol 81: 497-502 White PJ, Marshall J, Smith JAC (1990) Substrate kinetics of the tonoplast H + translocating inorganic pyrophosphatase and its activation by free Mg2+.Plant Physiol 93: 1063-1070 Zhen R-G, Kim EJ, Rea PA (1994) Localization of cytosolically orientated maleimide-reactive domain of vacuolar H'-pyrophosphatase. J Biol Chem 269 23342-23350
