- Email: [email protected]
Plant cytosolic pyruvate kinase: a kinetic study
Biochimica et Biophysica Acta, 1160 (1992) 213-220 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/$05.00
213
BBAPRO 34324
Plant cytosolic pyruvate kinase: a kinetic study Florencio E. Podest~i and William C. Plaxton Departments of Biology and Biochemistry, Queen's University, lO'ngston, Ontario (Canada)
(Received 21 October 1991) (Revised manuscript received 28 May 1992)
Key words: Oil seed; Pyruvate kinase; Enzyme regulation; Enzyme kinetics; Glycolysis;Phosphoenolpyruvate The kinetic properties of cytosolic pyruvate kinase (PK c) from germinating castor oil seeds (COS) have been investigated. From experiments in which the free Mg 2+ concentration was varied at constant levels of either the complexed or free forms of the substrates it was determined that the true substrates are the free forms of both phosphoenolpyruvate (PEP) and ADP. This conclusion is corroborated by the quenching of intrinsic PK c tryptophan fluorescence by free PEP and ADP. Mg 2+ is bound as the free bivalent cation but is likely released as MgATP. The fluorescence data, substrate interaction kinetics, and pattern of inhibition by products and substrate analogues (adenosine 5'-O-(2-thiodiphosphate) for ADP and phenyl phosphate for PEP) are compatible with a sequential, compulsory-ordered, Tri-Bi type kinetic reaction mechanism. PEP is the leading substrate, and pyruvate the last product to abandon the enzyme. The dissociation constant and limiting K m for free PEP (8.2 to 22 and 38 ~M, respectively) and the limiting K m for free ADP (2.9/zM) are considerably lower than those reported for the non-plant enzyme. The results indicate that COS PK c exists naturally in an activated state, similar to the fructose 1,6-bisphosphate-activated yeast enzyme. This deduction is consistent with a previous study (F.E. Podest~i and W.C. Plaxton (1991) Biochem. J. 279, 495-501) that failed to identify any allosteric activators for the COS PK¢, but which proposed a regulatory mechanism based upon ATP levels and pH-dependent alterations in the enzyme's response to various metabolite inhibitors. As plant phosphofructokinases display potent inhibition by PEP, the overall rate of glyeolytic flux from hexose 6-phosphate to pyruvate in the plant cytosol will ultimately depend upon variations in PEP levels brought about by the regulation of PKg.
Introduction Pyruvate kinase (PK, EC 2.7.1.40) catalyses the substrate level phosphorylation of A D P at the expense of PEP and is considered to be a key regulatory enzyme of glycolysis in all phyla [1-5]. Both allosteric mechanisms and reversible phosphorylation can account for the regulation of non-plant PKs [1,4-7]. Relative to the extensive work done with PK from non-plant sources, comparatively little is known about the detailed kinetic characteristics of the cytosolic PK (PK c) or plastidic PK isoenzymes of plants. Plant PK is, however, of considerable interest, because a substantial amount evidence indicates that this enzyme is the primary control site of glycolytic flux to pyruvate in the plant cytosol [8-11]. A reduction in P E P levels, brought about by an enhancement of PK activity, will stimulate
Correspondence to: W.C. Plaxton, Dept. of Biology, Queen's University, Kingston, Ontario, Canada K7L 3N6. Abbreviations: PK, pyruvate kinase; PKc, cytosolic pyruvate kinase; PFK, phosphofructokinase (EC 2.7.1.11); COS, castor oil seed; PEP, phosphoenolpyruvate; ADP-/3-S, adenosine 5'-O-(2-thiodiphosphate).
plant phosphofructokinase (PFK), because PEP is a potent inhibitor of PFKs from plant sources [11,12]. By contrast, P F K is believed to be the most important control element in non-plant glycolysis, with secondary regulation vested at the level of PK [13]. Early in the study of plant glycolysis, PK c was demonstrated to play a fundamental role in regulating a metabolic reorganization from gluconeogenesis to glycolysis that occurs when germinated COS undergoes anaerobiosis [8]. We have recently reported that PK c purified from this tissue exhibits substrate saturation kinetics characteristic of a fully active enzyme and does not display sensitivity to the major allosteric effectors of other PKs [14]. As well, no evidence was obtained that the enzyme was a phosphoprotein in either anoxic or aerobic germinating COS [14]. It was concluded that the PKc activity of germinating COS is enhanced following anaerobiosis through concerted decreases in A T P levels, cytosolic p H and concentrations of several key inhibitors [14]. In this paper, we examine the kinetic reaction mechanism of the homogeneous PK~ from germinating COS. Our results reinforce the concept that COS PK~ exists in an activated state similar to the fructose 1,6-bisphos-
214 phate-activated form of the non-plant enzyme and further exemplify the peculiarities of plant versus nonplant glycolysis.
substrate not complexed with Mg2+: PEP 3-, HPEP 2and KPEP 2-. There is only one possible Mg-complexed form of PEP, designated MgPEP. Free Mg 2+ is referred to as Mg 2+.
Materials and Methods
Chemicals and plant material Mes, Bis-tris-propane, Hepes, PEP and dithiothreitol were from Research Organics (Cleveland, OH, USA). NADH was from Boehringer-Mannheim (Montreal, Quebec, Canada). Other biochemicals and coupling enzymes were purchased from Sigma (St. Louis, MO, USA). All other chemicals were of analytical grade obtained from BDH (Toronto, Ontario, Canada). All solutions were prepared with Milli-Qprocessed water. Seeds of the castor plant (Ricinus communis L., cv. Hale) were germinated for 5 days as described in Ref. 8, dissected free from hypocotyls and the endosperms were stored at -80°C until used. PK~ purification was performed as in Ref. 15, except that the heat step was omitted [14]. Typical values of specific activity for the purified enzyme were around 180 /zmol of pyruvate produced/min per mg protein at pH 7.2 and 30°C. Enzyme assay The PK activity was monitored at pH 7.2 and 30°C by following the reduction of pyruvate catalysed by lactic dehydrogenase in a medium containing 25 mM Mes/Bis-tris-propane/HC1, 20 mM KCI, 2 mM dithiothreitol, 0.15 mM NADH and 2 units of lactic dehydrogenase. For examination of pyruvate inhibition, NADH and lactic dehydrogenase were replaced by 0.5 mM NAD +, 2 mM glucose, 2 units of yeast hexokinase and 2 units of Leuconostoc mesenteroides glucose 6-phosphate dehydrogenase. Substrates and MgC1z are as indicated in the figure legends and text. Absorbance changes at 340 nm were measured on a Varian DMS200 spectrophotometer. Coupling enzymes were desalted before use. Assays were initiated by the addition of PK c. No difference in the specific activity of PK¢ was evident between the two coupling systems used. Assays were linear for at least 10 min. Stock solutions of substrates, analogues and products were adjusted to pH 7. The concentration of nucleotides was checked spectrophotometrically using published extinction coefficients [16]. Terminology The notations used to differentiate among the different substrate species existing in solution are as follows: ADPf represents the sum of all species of ADP not complexed with Mg2+: ADP 3-, HADP 2- and KADP 2-, while MgADP denotes the sum of the two Mg-complexed forms of ADP: M g A D P - and MgHADP. Similarly, PEPf stands for the forms of this
Data analysis Analysis of the initial velocity data was done with a non-linear least-squares regression computer kinetics program (S. Brooks, unpublished work) or alternatively by fitting of the data to the general equation V
V=
1+
A
-K- m
a
Km B
+ --
b
+
KsAKmS
(1)
ab
using the built-in curve fitting program from Sigmaplot version 4.1 (Jandel Scientific), where Km A and Kam are the Michaelis constants for substrates A and B, respectively; K~ is the dissociation constant for the interaction between the enzyme and the first substrate to bind and a and b are the concentrations of substrates A and B. Patterns of inhibition were evaluated by Lineweaver-Burk plots. Concentrations of the different substrate forms in the presence of MgCI 2 and KC1 were calculated using computer programs compiled in Turbo Pascal based on the algorithms given by McFarlane and Ainsworth (see Appendix to Ref. 17). Briefly, the programs allow the calculation of the concentration of any of the substrate forms at a given pH, preset free Mg 2÷ and free K ÷ concentration and at fixed concentration of one selected species of ADP and PEP. Additionally, terms were added to the polynomial equation to perform the same calculations while correcting for the chelating effects of ATP, pyruvate, ADP-fl-S and phenyl phosphate. The total amount of MgC12, KCI and substrates or effectors to be added is given by the programs. The dissociation constants for the Mg-complexes with substrates and products were obtained from Ref. 17 and for phenyl phosphate from Ref. 18. No data were available concerning ADP-/3-S-chelating properties and it, thus, was assumed to be equivalent to ADP in this respect.
True substrates determination The form of ADP and PEP bound to PK¢ was assessed by measuring enzyme activity at different levels of Mg 2÷ while maintaining constant concentrations of the chosen species of ADP and PEP. The rationale for the test can be found elsewhere [17]. Briefly, the inverted form of the general equation may consist of terms including the reciprocal concentration of Mg 2÷ and combinations of those of ADPf, PEPf, MgADP or MgPEP, according to the possible complexes with the enzyme. Taking into consideration the Mg-binding constants at fixed pH and free K ÷, the reciprocal velocity
215 at constant concentrations of the chosen substrate forms may be expressed as a polynomial in Mg2+: V- 1 = ~xfx(Mg2+
)i
(2)
whose power (i) is determined by the substrate form selected [17]. Fluorescence measurement The intrinsic fluorescence of COS PKc was determined on a Perkin-Elmer LS 50 spectrofluorometer using a stirred quartz cuvette thermostated at 30°C. Data were analyzed using the Perkin-Elmer FL Data Manager version 3.0. Freshly purified COS PK¢ was desalted in a Centricon 30 microconcentrator (Amicon) using a buffer containing 25 mM Hepes-KOH and 20% (v/v) glycerol (pH 7.0). Aliquots (30/zl) of the desalted enzyme were diluted to a final concentration of 5 / z g / m l in a volume of 700 /xl of the desalting buffer prior to each measurement. Excitation or emission spectra were recorded for the native enzyme using the desalting buffer as reference. PK~ steady-state intrinsic fluorescence was determined by exciting at 290 or 295 nm (slit width 5 nm) and detecting the fluorescence emission at 340 nm (slit width 7.5 or 10 nm). Under these conditions most of the fluorescence can be attributed to tryptophan residues. Additions of substrates a n d / o r Mg 2+ (maximum volume 10 /xl) were made from concentrated stock solutions brought to pH 7.0. At the concentrations of protein or other additions used, no corrections were necessary due to inner filter effect. The dissociation constants (K s) for PEP or ADP were obtained by measuring the degree of quenching caused by increasing concentrations of the substrates in the presence or absence of 10 mM Mg 2÷ and several concentrations of PEP in the case of ADP. The data were analyzed by non-linear least-squares fitting to Eqn. 3:
AF/Fo
(AFm~,/Fo)"S h Ks + Sh
(3)
where AF is the degree of quenching in the presence of a determined amount of substrate S, F o is the fluorescence in the absence of S, K s is the dissociation constant and h is the Hill number. The total fluorescence quenching was 2-fold higher for PEP than for ADP. Due to the better signal-to-noise ratio, the standard error affecting the results was lower for PEP (approx. 10%) than for ADP (approx. 30%). Protein determination Protein concentration was determined by the dye-binding method of Bradford [19] using bovine-T-globulin as standard.
Results
True substrates determination Fig. 1 shows the activity dependence on Mg 2+ concentration at fixed concentrations of combinations of the free or complexed forms of the substrates. Linear relationships (i.e., i values of 0 and - 1 ) should be observed for any combination of PEPf or MgPEP with ADPf or MgADP representing the true substrates. Inhibition was observed at high Mg 2÷ concentrations for all combinations, except the free forms of PEP and ADP. It is evident from the plot that only when the free forms of the substrates are kept constant is a linear relation obtained, suggesting that these are the catalytically viable forms. The inhibition observed at high Mg 2÷ concentration may arise as a consequence of the sequestration of the free forms of PEP and ADP from the medium. The lack of inhibition at high Mg 2+ when PEPf and ADPf are held constant generates two other important conclusions: (i), that Mg 2÷ does not elicit product inhibition, raising the possibility that it is eliminated as MgATP and not as Mg 2+ and (ii), MgADP and MgPEP are not dead-end inhibitors of PK c, since the concentrations of these rise with Mg 2÷ as PEPf and ADPf are held constant. Quenching o f intrinsic fluorescence by substrates COS PK c showed a broad emission spectrum centred at 340 nm when excited at 290 nm (inset Fig. 2) or 295 nm (not shown). Fig. 2 presents the effect of
0.15
•
,
.
,
.
,
.
I
,
0.10
00 .5= ~ 0.00
,
0.000
I
0.025
,
I
0.050
,
0.075
I
0.I00
1/[Mg ~*] (raM) -l Fig. 1. Dependence of PK c reaction rate on the concentration of Mg 2+. Enzyme activity was determined at constant levels of: (o), MgPEP (0.170 mM) and MgADP (0.187 mM); (e), total PEP (0.30 mM) and total ADP (0.20 mM); ( v ) , PEPf (0.094 mM) and MgADP (0.187 mM); ( v ) , MgPEP (0.170 raM) and ADPf (0.007 mM); (t3), PEPf (0.094 mM) and ADPf (0.007 mM). Activity units are expressed a s / z m o l / m i n per mg protein.
216 substrates and Mg 2+ on PKc tryptophanyl fluorescence. The quenching of fluorescence that occurs following the addition of 2.5 mM PEP demonstrates that PEP can bind to PK¢ in the presence or absence of MgCl 2 (Fig. 2, trace A and B, respectively). At the same time, it is evident that Mg 2+ does not have an effect over the steady-state PKc fluorescence, whether it is added alone (Fig. 2, trace A) or in combination with PEP (Fig. 2, trace A and B) or PEP plus A D P (Fig. 2, trace C). ADPf (0.12 raM) could not quench PKc intrinsic fluorescence unless it was preceded by 2.5 mM PEP (Fig. 2, traces C and D), in which case a rapid decline of the enzyme's fluorescence was observed. Taken together these results corroborate the experiments reported in Fig. 1, which suggested that PEPf and ADPf are the catalytically effective forms. Binding constants for both substrates were determined in the presence and absence of 10 mM MgC12. The K s for total PEP increased from 8.2 ~ M in the
] 300
320
340
360
380
Wavelength (nm) 5mM MgCI2
2JmM PEP
Bv~P~P B v ~
P~P
L_~mM
1
g -
0.1mM ADP
51 _oI
I
"2
h
IIIL mM MgCI2 5
'
.
ADP
O,03"2mM
O.064mM
~"
12
:t 9 m
6
I
0.0
i
I
I
I
0.5
1.0
1.5
[PEP]
2.0
(raM)
Fig. 3. Dependence of the dissociation constant for ADP on PEP concentration. Binding of ADP to PK c was analyzed by fluorescence quenching as described in Materials and Methods. The excitation wavelength was 295 nm (slit width 5 nm) and emission was detected at 340 nm (slit width 10 nm). PK c intrinsic fluorescence was excited
at 295 nm in order to avoid interference by the adenine ring of ADP.
absence of MgC12 to 1 8 / z M in its presence. Correcting the latter value for the effect of Mg-complexing renders a K s of 6.5 /zM for PEPf in the presence of 10 mM MgC12. The addition of 0.1 or 0.2 mM A D P increased the K s for PEP to 13 and 14 /xM, respectively. The K s for total A D P was 70 /zM in the presence of 10 mM MgCI 2 and 2 m M PEP. The corrected value for ADPf is 3.4 /zM, which is essentially equivalent to the K s of 3.1 /xM observed in the absence of MgCI 2. The binding of A D P to PK c was dependent on the amount of PEP present, with the K s for A D P decreasing with increasing PEP concentration (Fig. 3). These data suggest that the Mg2÷-induced apparent decrease in the affinity of PK c for PEP or A D P is due to sequestration of the free forms of the substrates by complexing with the cofactor.
32mM
95mMPEP
LL
15
O.12~mM
)
Time
Fig. 2. Effect of substrates and Mg2+ on PK c intrinsic fluorescence. The intrinsic fluorescence of PK¢ (expressed as arbitrary units, AU) was recorded as a function of time as described in Materials and Methods. Arrows indicate the addition of substrates or MgCI2. The spikes are the result of noise introduced by the syringe needle at the moment of addition. Inset: Intrinsic fluorescence of PKc. The emission spectrum was recorded using an excitation wavelength of 290 nm (slit width 5 nm) and detecting emission between 300 and 380 nm (slit width 7.5 nm) as described in Materials and Methods.
Substrate interaction kinetics Fig. 4A and B show the fit to Eqn. 1 for ADPf or PEPf binding at several concentrations of the co-substrate. The plots demonstrate a linear response of the reciprocal velocity with the concentration of each of the true substrates. The type of plot (intersecting lines) obtained in both cases suggests a sequential mechanism of reaction. The lines in the plots were fitted assuming that PEPf is the first substrate to bind and the considerations described below. From the intersection point in Fig. 4B, the K s for PEPf was calculated to be 22 ~M. The same value was obtained by plotting the slopes in Fig. 4B against 1 / P E P (not shown). This value is higher than that obtained by fluorometry and probably reflects the adverse effect of A D P on PEP binding. Plotting the intercepts in Fig. 4B against the reciprocal concentration of PEPf rendered a limiting K m for PEPf of 38 /zM (not shown). The limiting K m for ADPf was determined to be 2.9 ~ M from a plot of the intercepts on the 1/activity axis in Fig. 4A against the reciprocal of ADPf concentration (not shown).
217 0.016
0.06
•
r
.
~
•
I
A
A 0.04 0
0.008
..~ 0.0~ 0.000
,
i
-15 0.00
-800
0
I
I
800
400
i
600
I
i
0
15
I/[PEP]t
(mM) -t
I
30
0.016
I/[ADP]t (raM)-t
B
0.030
B O
0.008
J
,=4
0.015 0
0.000 -150
0.000 j p . - ~
,
0
I
150
J
300
1 / [ A D e ] f (raM) -~
-0.015 -50
-25
i
i
25
50
t/[PEP]t ( m M ) - I Fig. 4. Substrate interaction kinetics of PK c. Enzyme activity was determined at varying concentrations of ADPf (A) or PEP r (B). In A, fixed levels of PEPf were: (<3), 0.053; (e), 0.107; ( v ) , 0.215 and ( v ) , 0.438 raM. In B, fixed levels of ADPf were: (©), 7.1; (o), 14.6; (v), 30.2 and ( v ) , 65.3 p~M. The free Mg 2. concentration was 10 raM. Activity is expressed as p.mol/min per mg protein.
Product inhibition studies Fig. 5A reveals the pattern of product inhibition by MgATP with PEPf as the variable substrate. The inhibition was mixed whether ADPf was saturating (Fig. 5A) or subsaturating (Table I). Likewise, inhibition by MgATP when ADPf was the variable substrate was of the mixed type at saturating (Fig. 5B) and subsaturating PEPf (Table I). Pyruvate inhibition was competitive with respect to PEP at saturating ADPf (Fig. 6) and subsaturating ADPf (Table I). Pyruvate acted as non-competitive inhibitor with respect to ADPf at subsaturating PEPf (Fig. 7), but showed no inhibition when PEP was saturating (Table I). The different inhibition patterns offered by ATP and pyruvate are not compatible with a Theorell-Chance mechanism. The competition between PEPf and pyruvate (and the lack of response to pyruvate when ADPf was varied) could be explained by a compulsory ordered mechanism where PEPy is the first product to bind and pyruvate the last product to leave the catalytic site. A random-order mechanism could also account for the inhibition pattern observed.
Fig. 5. Inhibition of PK c by MgATP. (A), PEPf as variable substrate at saturating (0.1 mM) ADPf; (B), ADPf as variable substrate at saturating (0.5 mM) PEPf. The MgATP concentrations were: (©), 0 raM; (e), 2.5 mM; ( v ) , 5.0 mM; ( v ) , 7.5 raM; ([]), 10.0 raM. The free Mg 2+ concentration was 10 raM. Activity is expressed as ~,mol/min per mg protein.
Substrate analogue studies Further characterization of the reaction mechanism was achieved by the use of the substrate analogues ADP-/3-S and phenyl phosphate. The pattern of inhibition by these two compounds is shown in Figs. 8 and 9 (see also summary in Table I). The ADP analogue, ADP-/3-S, was not a substrate for PK¢, as no pyruvate production was observed following a 15 min incubation of 1 mM ADP-fl-S and 2 mM PEP with 0.015 units of PK c (i.e., about 4-fold the amount of PK c used routinely for an assay). Mixed TABLE I Inhibition patterns displayed by reaction products and substrate analogues of PK c
Abbreviations: Sat, saturating; Ssat, subsaturating; M, mixed inhibition; C, competitive inhibition; NC, non-competitive inhibition. Effector
Variable substrate PEPf
MgATP Pyruvate ADP-/3 -S Phenyl phosphate
ADP r
Sat ADP
Ssat ADP
Sat PEP
Ssat PEP
M C None
M C M
M None C
M NC M
C
C
None
NC
218
y
0.075
0.070
0.050
:E
0
•,~ 0.035 <
0.025
,-4
0.000 0.000 ' -15
~
i
0 1/[PEP][
I
i
I
i
15
30
i
15
30
1/[PEP]I
( m M ) -1
0
150
(raM) -I
Fig. 6. Inhibition of PK c by pyruvate with PEPf as variable substrate at saturating (0.I mM) ADPf. The concentrations of pyruvate were: (0), 0 mM; (•), 25 raM; ( v ) , 50 raM; ( v ) , I00 mM. The free Mg 2+ concentration was 10 raM. Activity is expressed as p.mol/min per mg protein.
inhibition was obtained against PEPf at subsaturating ADPf (Fig. 8A), while no changes in activity were evident at saturating ADPf (0.1 mM). Inhibition with respect to ADPf was competitive at saturating PEPf (Fig. 8B) and mixed at subsaturating PEPf (Table I). Clearly ADP-fl-S does not affect PEP binding to the free enzyme, but is capable of binding to the enzymePEP complex. The PEP analogue, phenyl phosphate [14], was assayed for its effect on PK~, since the high concentrations of pyruvate required to observe inhibition (Fig. 6) may have distorted the enzyme's response to this product by altering the ionic strength of the medium. Inhibition by phenyl phosphate with respect to PEPf was competitive at saturating ADPf (Fig. 9A) or subsaturating ADPf (Table I), while with respect to ADPf it inhibited non-competitively at subsaturating PEPf (Fig. 9B) and did not inhibit (up to 20 raM) at saturating PEPf. Reciprocal plots against ADPf were linear, indicating that ADP cannot bind to the enzyme-phenyl phosphate complex. The results indicate an interaction only with the PEP binding site.
0.024
¢9
<
-
B
0,012 •
0.000 ~ -150
'
I
I/[ADPIr (raM)
'./()() -1
Fig. 8. Inhibition of PK c by ADP-fl-S. (A), PEPf as variable substrate with subsaturating (7 IzM) ADP[; (B), ADPf as variable substrate at saturating (0.5 mM) PEPf. The concentrations of ADP-fl-S were: (©), 0 mM; (o), 2.5 mM; ( v ) , 5.0 raM; ( t ) , 7.5 mM. The free Mg 2+ concentration was 10 mM. Activity is expressed as/zmol/min per mg protein.
0.024
p,
A"
•
i
•
0.016
C~
<
0.008
0.000 -0.025
0.000
1/[PEPlf
0.15
0.025
0.050
( m M ) -1
0.040
••~4 0.10
.4 ~
•
0.020 <
O
<
0.05
0.000
0.00 -300
-o.oto 0
300
600
I/[ADP]f (mM) -t Fig. 7. Inhibition of PK c by pyruvate with ADPf as variable substrate at subsaturating (0.025 mM) PEPf. The concentrations of pyruvate were: (©), 0 raM; (o), 25 raM; ( v ) , 50 raM; ( • ) , 100 raM. The free Mg z÷ concentration was 10 mM. Activity is expressed as ~moi/rnin per mg protein.
,
o.ooo
o.olo
i
0.020
1/[ADP]f (mM) -t Fig. 9. Inhibition of PK c by phenyl phosphate. (A), PEP~ as variable substrate at saturating (0.1 raM) ADPf; (B), ADPf as variable substrate at subsaturating (25 IzM) PEPf. The concentrations of phenyl phosphate were (o), 0 mM; (o), 2.5 mM; ( v ), 5.0 mM; ( • ), 7.5 mM; (C3), 10.0 mM. The free Mg 2+ concentration was 10 mM. Activity is expressed as/xmol/min per mg protein.
219 pEp3-
ADp3-
Mg 2÷
MgATP
Pyruvate
pK c ............................................................................................................ ~PKc
Scheme I
Discussion
A recent study characterizing germinating COS PK c [14] demonstrated this enzyme as unique compared to other plant [2,20], green algal [21], animal [1,3,4] and yeast [3,5] PKs with respect to its kinetic properties. Namely, the COS enzyme presented hyperbolic kinetics for PEP binding, distinguishing it from other PKs recognized as regulatory enzymes [1,3-5,7,17,22]. No activators were identified for germinating COS PK~ [14], in contrast to the spinach leaf [20] and green algal [21] PK. In fact, germinating COS PK~ behaves as a fully active enzyme under a variety of conditions, being sensitive to changes in pH within the physiological range and to a variety of metabolite inhibitors [14]. Kinetic reaction mechanism
Isolated efforts to characterize the kinetic reaction mechanism of a plant PK have had no clear outcome [23,24], perhaps as a consequence of a failure to identify the true substrates. A report on the kinetic properties of the partially purified PK~ and plastidic PK from developing COS suggested a compulsory-ordered, sequential type reaction mechanism for both isoenzymes [24]. The results presented here indicate that germinating COS PK c has a compulsory-ordered Tri-Bi reaction mechanism. The following evidence support this hypothesis. First, K m values for both substrates are not independent of the co-substrate concentration (Fig. 4A and B). Second, PEPf appears to bind independently to PKc, whereas ADPf can bind only if PEPf is already present (Figs. 2 and 3). None of the above would be applicable in a random-order mechanism. Third, although the product inhibition data are not unambiguous, the results can certainly be explained in light of the known properties of PKg. The inhibition by MgATP and ADP-/3-S deviate from the expected behaviour in a compulsory-ordered mechanism when PEPf is varied. At saturating ADP, MgATP should be uncompetitive (if the reaction was reversible to some degree) or not inhibit at all (for an irreversible reaction, which is probably the case with PK~). It has been shown, however, that a high ADP concentration (> 1.5 mM) produces mixed inhibition of PK~ [14]. It is, therefore, plausible for ATP (and more so for ADP-/3-S) to interact with the ADP inhibitory site producing mixed inhibition. ADP-/3-S would also be expected to produce uncompetitive inhibition due to sequestering of the PKc-PEP complex. That it does not could be the result of a masking effect caused by its interaction with
the ADP inhibitory site, as discussed for MgATP, with the addition that it may be unable to bind or have a very low affinity for the catalytic site of the PKc-PEP complex. The reaction mechanism proposed here for the COS PKc is similar to that described for the fructose 1,6-bisphosphate-activated yeast PK [17], but different to the Ping-Pong or rapid equilibrium random-type mechanisms postulated respectively for the enzyme from pig liver and rabbit or Carcinus maenas muscles [25-27]. The substrates are bound in their free form, with PEP being the leading substrate, as shown in Scheme I. As discussed previously for the yeast enzyme [17], the separate binding of Mg 2÷ and its elimination as MgATP (positioned between the y- and /3-phosphate) suggests that the role of the bivalent cation in catalysis is in bridging the /3-phosphate group of ADP and the PEP phosphate group. Lack of an effect of Mg 2÷ on the intrinsic fluorescence of PK c suggests that an independent interaction of this co-factor with the enzyme may be very limited and supports the assertion that Mg 2÷ is not liberated as the free cation. Binding constants are comparatively lower than the values reported for the yeast enzyme for PEPf (K s COS PKc: 8.2 and 22 /zM (fluorometric and kinetic determination, respectively); K s yeast PK: 69 to 87/zM [17,28]) and more notably for ADPf (limiting K m COS PKc: 2.9/xM; g m yeast PK: 840 ~M [17]). Metabolic consequences o f COS P K c kinetic properties
The significance of the current and a related study [14] for germinating COS metabolism is that PK c from this tissue exists in an activated form at all times and that its activity is enhanced in response to anoxia not by activation but rather through concerted decreases in ATP levels, cytosolic pH and concentrations of several inhibitors. The high affinities of germinating COS PKc for its substrates may reflect the particular metabolic needs that differentiate this plant enzyme from other PKs. The fact that only the free forms of ADP and PEP serve as substrates makes it tempting to speculate that fluctuations in cytosolic Mg 2+ levels may regulate the activity of COS PKg. Large changes in metabolite concentrations take place in germinating COS upon anaerobiosis [8]. Notably, estimated cytosolic ATP levels plunge from about 6 to 2 mM [8,14]. The cytosolic pH is also expected to quickly drop [29], not only releasing PK¢ from inhibition by various metabolites, but also allowing different interactions between PEP and ADP and Mg 2+. The resulting redistribution of
220 substrate species could have an influential role in regulating PK c activity. It is important to note that the overall glycolytic rate will be controlled by PK c activity, since PFK activity is largely dependent on the cytosolic concentration of its allosteric inhibitor, PEP.
Acknowledgement This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). References 1 Kayne, F.J. (1973) in The Enzymes (Boyer, P.D., ed.), 3rd Edn., Vol. 8, pp. 353-382, Academic Press, New York. 2 Duggleby, R.G. and Dennis, D.T. (1973) Plant Physiol. 52, 312317. 3 Munday, K.A., Giles, I.G. and Poat, P.C. (1980) Comp. Biochem. Physiol. 67B, 403-411. 4 Plaxton, W.C. and Storey, K.B. (1984) Eur. J. Biochem. 143, 257-265. 5 De Arriaga, D., Busto, F., Del Valle, P. and Soler, J. (1989) Biochim. Biophys. Acta 998, 221-230. 6 Blair, J.B. and Walker, R.G. (1984) Arch. Biochem. Biophys. 232, 202-213. 7 Callens, M., Kuntz, D.A. and Opperdoes, F.R. (1991) Mol. Biochem. Parasitol. 47, 19-30. 8 Kobr, M.J. and Beevers, H. (1971) Plant Physiol. 47, 48-52. 9 Turpin, D.H., Botha, F.C., Smith, R.G., Feil, R., Horsey, A.K. and Vanlerberghe, G.C. (1990) Plant Physiol. 65, 1188-1193
10 Hatzfeld, W. and Stitt, M. (1991) Physiol. Plant. 81, 103-110 11 Copeland, L. and Turner, J.F. (1987) in The Biochemistry of Plants (Stumpf, P.K. and Conn, E.E., eds.), Vol. 11, pp. 107-128, Academic Press, New York. 12 Dennis, D.T. and Greyson, M.F. (1987) Physiol. Plant. 69, 395404. 13 Stryer, L. (1988) in Biochemistry, 3rd Edn., pp. 349-372, Freeman, New York. 14 Podest~, F.E. and Plaxton, W.C. (1991) Biochem. J., 279, 495-501. 15 Plaxton, W.C. (1988) Plant Physiol. 86, 1064-1069. 16 Bock, R.M., Ling, N-S., Morrel, S.A. and Lipfon, S.H. (1956) Arch. Biochem. Biophys. 62, 253-264. 17 McFarlane, N. and Ainsworth, S. (1972) Biochem. J. 129, 10351047. 18 Rodrlguez-Sotres, R. and Mufioz-Clares, R.A. (1990) Arch. Biochem. Biophys. 276, 180-190. 19 Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. 20 Baysdorfer, C. and Bassham, J.A. (1984) Plant Physiol. 74, 374379. 21 Lin, M., Turpin, D.H. and Plaxton, W.C. (1989) Arch. Biochem. Biophys. 269, 228-238. 22 Kahn, A., Marie, J., Garreau, H. and Sprengers, E.D. (1978) Biochim. Biophys. Acta 523, 59-74. 23 Ireland, R.J., DeLuca, V. and Dennis, D.T. (1980) Plant Physiol. 65, 1188-1193. 24 Gupta, V.K. and Singh, R. (1989) Plant Physiol. Biochem. 27, 703-711. 25 McFarlane, N. and Ainsworth, S. (1974) Biochem. J. 139, 499-508. 26 Ainsworth, S. and McFarlane, N. (1973) Biochem. J. 131,223-236. 27 Newton, C.J., Poat, P.C. and Munday, K.A. (1976) Biochem. Soc. Trans. 4, 479-481. 28 Mildvan, A.S. and Cohn, M. (1966) J. Biol. Chem. 241, 1178-1193. 29 Saint-Ges, V., Roby, C., Bligny, R., Pradet, A. and Douce, R. (1991) Eur. J. Biochem. 200, 477-482.
213
BBAPRO 34324
Plant cytosolic pyruvate kinase: a kinetic study Florencio E. Podest~i and William C. Plaxton Departments of Biology and Biochemistry, Queen's University, lO'ngston, Ontario (Canada)
(Received 21 October 1991) (Revised manuscript received 28 May 1992)
Key words: Oil seed; Pyruvate kinase; Enzyme regulation; Enzyme kinetics; Glycolysis;Phosphoenolpyruvate The kinetic properties of cytosolic pyruvate kinase (PK c) from germinating castor oil seeds (COS) have been investigated. From experiments in which the free Mg 2+ concentration was varied at constant levels of either the complexed or free forms of the substrates it was determined that the true substrates are the free forms of both phosphoenolpyruvate (PEP) and ADP. This conclusion is corroborated by the quenching of intrinsic PK c tryptophan fluorescence by free PEP and ADP. Mg 2+ is bound as the free bivalent cation but is likely released as MgATP. The fluorescence data, substrate interaction kinetics, and pattern of inhibition by products and substrate analogues (adenosine 5'-O-(2-thiodiphosphate) for ADP and phenyl phosphate for PEP) are compatible with a sequential, compulsory-ordered, Tri-Bi type kinetic reaction mechanism. PEP is the leading substrate, and pyruvate the last product to abandon the enzyme. The dissociation constant and limiting K m for free PEP (8.2 to 22 and 38 ~M, respectively) and the limiting K m for free ADP (2.9/zM) are considerably lower than those reported for the non-plant enzyme. The results indicate that COS PK c exists naturally in an activated state, similar to the fructose 1,6-bisphosphate-activated yeast enzyme. This deduction is consistent with a previous study (F.E. Podest~i and W.C. Plaxton (1991) Biochem. J. 279, 495-501) that failed to identify any allosteric activators for the COS PK¢, but which proposed a regulatory mechanism based upon ATP levels and pH-dependent alterations in the enzyme's response to various metabolite inhibitors. As plant phosphofructokinases display potent inhibition by PEP, the overall rate of glyeolytic flux from hexose 6-phosphate to pyruvate in the plant cytosol will ultimately depend upon variations in PEP levels brought about by the regulation of PKg.
Introduction Pyruvate kinase (PK, EC 2.7.1.40) catalyses the substrate level phosphorylation of A D P at the expense of PEP and is considered to be a key regulatory enzyme of glycolysis in all phyla [1-5]. Both allosteric mechanisms and reversible phosphorylation can account for the regulation of non-plant PKs [1,4-7]. Relative to the extensive work done with PK from non-plant sources, comparatively little is known about the detailed kinetic characteristics of the cytosolic PK (PK c) or plastidic PK isoenzymes of plants. Plant PK is, however, of considerable interest, because a substantial amount evidence indicates that this enzyme is the primary control site of glycolytic flux to pyruvate in the plant cytosol [8-11]. A reduction in P E P levels, brought about by an enhancement of PK activity, will stimulate
Correspondence to: W.C. Plaxton, Dept. of Biology, Queen's University, Kingston, Ontario, Canada K7L 3N6. Abbreviations: PK, pyruvate kinase; PKc, cytosolic pyruvate kinase; PFK, phosphofructokinase (EC 2.7.1.11); COS, castor oil seed; PEP, phosphoenolpyruvate; ADP-/3-S, adenosine 5'-O-(2-thiodiphosphate).
plant phosphofructokinase (PFK), because PEP is a potent inhibitor of PFKs from plant sources [11,12]. By contrast, P F K is believed to be the most important control element in non-plant glycolysis, with secondary regulation vested at the level of PK [13]. Early in the study of plant glycolysis, PK c was demonstrated to play a fundamental role in regulating a metabolic reorganization from gluconeogenesis to glycolysis that occurs when germinated COS undergoes anaerobiosis [8]. We have recently reported that PK c purified from this tissue exhibits substrate saturation kinetics characteristic of a fully active enzyme and does not display sensitivity to the major allosteric effectors of other PKs [14]. As well, no evidence was obtained that the enzyme was a phosphoprotein in either anoxic or aerobic germinating COS [14]. It was concluded that the PKc activity of germinating COS is enhanced following anaerobiosis through concerted decreases in A T P levels, cytosolic p H and concentrations of several key inhibitors [14]. In this paper, we examine the kinetic reaction mechanism of the homogeneous PK~ from germinating COS. Our results reinforce the concept that COS PK~ exists in an activated state similar to the fructose 1,6-bisphos-
214 phate-activated form of the non-plant enzyme and further exemplify the peculiarities of plant versus nonplant glycolysis.
substrate not complexed with Mg2+: PEP 3-, HPEP 2and KPEP 2-. There is only one possible Mg-complexed form of PEP, designated MgPEP. Free Mg 2+ is referred to as Mg 2+.
Materials and Methods
Chemicals and plant material Mes, Bis-tris-propane, Hepes, PEP and dithiothreitol were from Research Organics (Cleveland, OH, USA). NADH was from Boehringer-Mannheim (Montreal, Quebec, Canada). Other biochemicals and coupling enzymes were purchased from Sigma (St. Louis, MO, USA). All other chemicals were of analytical grade obtained from BDH (Toronto, Ontario, Canada). All solutions were prepared with Milli-Qprocessed water. Seeds of the castor plant (Ricinus communis L., cv. Hale) were germinated for 5 days as described in Ref. 8, dissected free from hypocotyls and the endosperms were stored at -80°C until used. PK~ purification was performed as in Ref. 15, except that the heat step was omitted [14]. Typical values of specific activity for the purified enzyme were around 180 /zmol of pyruvate produced/min per mg protein at pH 7.2 and 30°C. Enzyme assay The PK activity was monitored at pH 7.2 and 30°C by following the reduction of pyruvate catalysed by lactic dehydrogenase in a medium containing 25 mM Mes/Bis-tris-propane/HC1, 20 mM KCI, 2 mM dithiothreitol, 0.15 mM NADH and 2 units of lactic dehydrogenase. For examination of pyruvate inhibition, NADH and lactic dehydrogenase were replaced by 0.5 mM NAD +, 2 mM glucose, 2 units of yeast hexokinase and 2 units of Leuconostoc mesenteroides glucose 6-phosphate dehydrogenase. Substrates and MgC1z are as indicated in the figure legends and text. Absorbance changes at 340 nm were measured on a Varian DMS200 spectrophotometer. Coupling enzymes were desalted before use. Assays were initiated by the addition of PK c. No difference in the specific activity of PK¢ was evident between the two coupling systems used. Assays were linear for at least 10 min. Stock solutions of substrates, analogues and products were adjusted to pH 7. The concentration of nucleotides was checked spectrophotometrically using published extinction coefficients [16]. Terminology The notations used to differentiate among the different substrate species existing in solution are as follows: ADPf represents the sum of all species of ADP not complexed with Mg2+: ADP 3-, HADP 2- and KADP 2-, while MgADP denotes the sum of the two Mg-complexed forms of ADP: M g A D P - and MgHADP. Similarly, PEPf stands for the forms of this
Data analysis Analysis of the initial velocity data was done with a non-linear least-squares regression computer kinetics program (S. Brooks, unpublished work) or alternatively by fitting of the data to the general equation V
V=
1+
A
-K- m
a
Km B
+ --
b
+
KsAKmS
(1)
ab
using the built-in curve fitting program from Sigmaplot version 4.1 (Jandel Scientific), where Km A and Kam are the Michaelis constants for substrates A and B, respectively; K~ is the dissociation constant for the interaction between the enzyme and the first substrate to bind and a and b are the concentrations of substrates A and B. Patterns of inhibition were evaluated by Lineweaver-Burk plots. Concentrations of the different substrate forms in the presence of MgCI 2 and KC1 were calculated using computer programs compiled in Turbo Pascal based on the algorithms given by McFarlane and Ainsworth (see Appendix to Ref. 17). Briefly, the programs allow the calculation of the concentration of any of the substrate forms at a given pH, preset free Mg 2÷ and free K ÷ concentration and at fixed concentration of one selected species of ADP and PEP. Additionally, terms were added to the polynomial equation to perform the same calculations while correcting for the chelating effects of ATP, pyruvate, ADP-fl-S and phenyl phosphate. The total amount of MgC12, KCI and substrates or effectors to be added is given by the programs. The dissociation constants for the Mg-complexes with substrates and products were obtained from Ref. 17 and for phenyl phosphate from Ref. 18. No data were available concerning ADP-/3-S-chelating properties and it, thus, was assumed to be equivalent to ADP in this respect.
True substrates determination The form of ADP and PEP bound to PK¢ was assessed by measuring enzyme activity at different levels of Mg 2÷ while maintaining constant concentrations of the chosen species of ADP and PEP. The rationale for the test can be found elsewhere [17]. Briefly, the inverted form of the general equation may consist of terms including the reciprocal concentration of Mg 2÷ and combinations of those of ADPf, PEPf, MgADP or MgPEP, according to the possible complexes with the enzyme. Taking into consideration the Mg-binding constants at fixed pH and free K ÷, the reciprocal velocity
215 at constant concentrations of the chosen substrate forms may be expressed as a polynomial in Mg2+: V- 1 = ~xfx(Mg2+
)i
(2)
whose power (i) is determined by the substrate form selected [17]. Fluorescence measurement The intrinsic fluorescence of COS PKc was determined on a Perkin-Elmer LS 50 spectrofluorometer using a stirred quartz cuvette thermostated at 30°C. Data were analyzed using the Perkin-Elmer FL Data Manager version 3.0. Freshly purified COS PK¢ was desalted in a Centricon 30 microconcentrator (Amicon) using a buffer containing 25 mM Hepes-KOH and 20% (v/v) glycerol (pH 7.0). Aliquots (30/zl) of the desalted enzyme were diluted to a final concentration of 5 / z g / m l in a volume of 700 /xl of the desalting buffer prior to each measurement. Excitation or emission spectra were recorded for the native enzyme using the desalting buffer as reference. PK~ steady-state intrinsic fluorescence was determined by exciting at 290 or 295 nm (slit width 5 nm) and detecting the fluorescence emission at 340 nm (slit width 7.5 or 10 nm). Under these conditions most of the fluorescence can be attributed to tryptophan residues. Additions of substrates a n d / o r Mg 2+ (maximum volume 10 /xl) were made from concentrated stock solutions brought to pH 7.0. At the concentrations of protein or other additions used, no corrections were necessary due to inner filter effect. The dissociation constants (K s) for PEP or ADP were obtained by measuring the degree of quenching caused by increasing concentrations of the substrates in the presence or absence of 10 mM Mg 2÷ and several concentrations of PEP in the case of ADP. The data were analyzed by non-linear least-squares fitting to Eqn. 3:
AF/Fo
(AFm~,/Fo)"S h Ks + Sh
(3)
where AF is the degree of quenching in the presence of a determined amount of substrate S, F o is the fluorescence in the absence of S, K s is the dissociation constant and h is the Hill number. The total fluorescence quenching was 2-fold higher for PEP than for ADP. Due to the better signal-to-noise ratio, the standard error affecting the results was lower for PEP (approx. 10%) than for ADP (approx. 30%). Protein determination Protein concentration was determined by the dye-binding method of Bradford [19] using bovine-T-globulin as standard.
Results
True substrates determination Fig. 1 shows the activity dependence on Mg 2+ concentration at fixed concentrations of combinations of the free or complexed forms of the substrates. Linear relationships (i.e., i values of 0 and - 1 ) should be observed for any combination of PEPf or MgPEP with ADPf or MgADP representing the true substrates. Inhibition was observed at high Mg 2÷ concentrations for all combinations, except the free forms of PEP and ADP. It is evident from the plot that only when the free forms of the substrates are kept constant is a linear relation obtained, suggesting that these are the catalytically viable forms. The inhibition observed at high Mg 2÷ concentration may arise as a consequence of the sequestration of the free forms of PEP and ADP from the medium. The lack of inhibition at high Mg 2+ when PEPf and ADPf are held constant generates two other important conclusions: (i), that Mg 2÷ does not elicit product inhibition, raising the possibility that it is eliminated as MgATP and not as Mg 2+ and (ii), MgADP and MgPEP are not dead-end inhibitors of PK c, since the concentrations of these rise with Mg 2÷ as PEPf and ADPf are held constant. Quenching o f intrinsic fluorescence by substrates COS PK c showed a broad emission spectrum centred at 340 nm when excited at 290 nm (inset Fig. 2) or 295 nm (not shown). Fig. 2 presents the effect of
0.15
•
,
.
,
.
,
.
I
,
0.10
00 .5= ~ 0.00
,
0.000
I
0.025
,
I
0.050
,
0.075
I
0.I00
1/[Mg ~*] (raM) -l Fig. 1. Dependence of PK c reaction rate on the concentration of Mg 2+. Enzyme activity was determined at constant levels of: (o), MgPEP (0.170 mM) and MgADP (0.187 mM); (e), total PEP (0.30 mM) and total ADP (0.20 mM); ( v ) , PEPf (0.094 mM) and MgADP (0.187 mM); ( v ) , MgPEP (0.170 raM) and ADPf (0.007 mM); (t3), PEPf (0.094 mM) and ADPf (0.007 mM). Activity units are expressed a s / z m o l / m i n per mg protein.
216 substrates and Mg 2+ on PKc tryptophanyl fluorescence. The quenching of fluorescence that occurs following the addition of 2.5 mM PEP demonstrates that PEP can bind to PK¢ in the presence or absence of MgCl 2 (Fig. 2, trace A and B, respectively). At the same time, it is evident that Mg 2+ does not have an effect over the steady-state PKc fluorescence, whether it is added alone (Fig. 2, trace A) or in combination with PEP (Fig. 2, trace A and B) or PEP plus A D P (Fig. 2, trace C). ADPf (0.12 raM) could not quench PKc intrinsic fluorescence unless it was preceded by 2.5 mM PEP (Fig. 2, traces C and D), in which case a rapid decline of the enzyme's fluorescence was observed. Taken together these results corroborate the experiments reported in Fig. 1, which suggested that PEPf and ADPf are the catalytically effective forms. Binding constants for both substrates were determined in the presence and absence of 10 mM MgC12. The K s for total PEP increased from 8.2 ~ M in the
] 300
320
340
360
380
Wavelength (nm) 5mM MgCI2
2JmM PEP
Bv~P~P B v ~
P~P
L_~mM
1
g -
0.1mM ADP
51 _oI
I
"2
h
IIIL mM MgCI2 5
'
.
ADP
O,03"2mM
O.064mM
~"
12
:t 9 m
6
I
0.0
i
I
I
I
0.5
1.0
1.5
[PEP]
2.0
(raM)
Fig. 3. Dependence of the dissociation constant for ADP on PEP concentration. Binding of ADP to PK c was analyzed by fluorescence quenching as described in Materials and Methods. The excitation wavelength was 295 nm (slit width 5 nm) and emission was detected at 340 nm (slit width 10 nm). PK c intrinsic fluorescence was excited
at 295 nm in order to avoid interference by the adenine ring of ADP.
absence of MgC12 to 1 8 / z M in its presence. Correcting the latter value for the effect of Mg-complexing renders a K s of 6.5 /zM for PEPf in the presence of 10 mM MgC12. The addition of 0.1 or 0.2 mM A D P increased the K s for PEP to 13 and 14 /xM, respectively. The K s for total A D P was 70 /zM in the presence of 10 mM MgCI 2 and 2 m M PEP. The corrected value for ADPf is 3.4 /zM, which is essentially equivalent to the K s of 3.1 /xM observed in the absence of MgCI 2. The binding of A D P to PK c was dependent on the amount of PEP present, with the K s for A D P decreasing with increasing PEP concentration (Fig. 3). These data suggest that the Mg2÷-induced apparent decrease in the affinity of PK c for PEP or A D P is due to sequestration of the free forms of the substrates by complexing with the cofactor.
32mM
95mMPEP
LL
15
O.12~mM
)
Time
Fig. 2. Effect of substrates and Mg2+ on PK c intrinsic fluorescence. The intrinsic fluorescence of PK¢ (expressed as arbitrary units, AU) was recorded as a function of time as described in Materials and Methods. Arrows indicate the addition of substrates or MgCI2. The spikes are the result of noise introduced by the syringe needle at the moment of addition. Inset: Intrinsic fluorescence of PKc. The emission spectrum was recorded using an excitation wavelength of 290 nm (slit width 5 nm) and detecting emission between 300 and 380 nm (slit width 7.5 nm) as described in Materials and Methods.
Substrate interaction kinetics Fig. 4A and B show the fit to Eqn. 1 for ADPf or PEPf binding at several concentrations of the co-substrate. The plots demonstrate a linear response of the reciprocal velocity with the concentration of each of the true substrates. The type of plot (intersecting lines) obtained in both cases suggests a sequential mechanism of reaction. The lines in the plots were fitted assuming that PEPf is the first substrate to bind and the considerations described below. From the intersection point in Fig. 4B, the K s for PEPf was calculated to be 22 ~M. The same value was obtained by plotting the slopes in Fig. 4B against 1 / P E P (not shown). This value is higher than that obtained by fluorometry and probably reflects the adverse effect of A D P on PEP binding. Plotting the intercepts in Fig. 4B against the reciprocal concentration of PEPf rendered a limiting K m for PEPf of 38 /zM (not shown). The limiting K m for ADPf was determined to be 2.9 ~ M from a plot of the intercepts on the 1/activity axis in Fig. 4A against the reciprocal of ADPf concentration (not shown).
217 0.016
0.06
•
r
.
~
•
I
A
A 0.04 0
0.008
..~ 0.0~ 0.000
,
i
-15 0.00
-800
0
I
I
800
400
i
600
I
i
0
15
I/[PEP]t
(mM) -t
I
30
0.016
I/[ADP]t (raM)-t
B
0.030
B O
0.008
J
,=4
0.015 0
0.000 -150
0.000 j p . - ~
,
0
I
150
J
300
1 / [ A D e ] f (raM) -~
-0.015 -50
-25
i
i
25
50
t/[PEP]t ( m M ) - I Fig. 4. Substrate interaction kinetics of PK c. Enzyme activity was determined at varying concentrations of ADPf (A) or PEP r (B). In A, fixed levels of PEPf were: (<3), 0.053; (e), 0.107; ( v ) , 0.215 and ( v ) , 0.438 raM. In B, fixed levels of ADPf were: (©), 7.1; (o), 14.6; (v), 30.2 and ( v ) , 65.3 p~M. The free Mg 2. concentration was 10 raM. Activity is expressed as p.mol/min per mg protein.
Product inhibition studies Fig. 5A reveals the pattern of product inhibition by MgATP with PEPf as the variable substrate. The inhibition was mixed whether ADPf was saturating (Fig. 5A) or subsaturating (Table I). Likewise, inhibition by MgATP when ADPf was the variable substrate was of the mixed type at saturating (Fig. 5B) and subsaturating PEPf (Table I). Pyruvate inhibition was competitive with respect to PEP at saturating ADPf (Fig. 6) and subsaturating ADPf (Table I). Pyruvate acted as non-competitive inhibitor with respect to ADPf at subsaturating PEPf (Fig. 7), but showed no inhibition when PEP was saturating (Table I). The different inhibition patterns offered by ATP and pyruvate are not compatible with a Theorell-Chance mechanism. The competition between PEPf and pyruvate (and the lack of response to pyruvate when ADPf was varied) could be explained by a compulsory ordered mechanism where PEPy is the first product to bind and pyruvate the last product to leave the catalytic site. A random-order mechanism could also account for the inhibition pattern observed.
Fig. 5. Inhibition of PK c by MgATP. (A), PEPf as variable substrate at saturating (0.1 mM) ADPf; (B), ADPf as variable substrate at saturating (0.5 mM) PEPf. The MgATP concentrations were: (©), 0 raM; (e), 2.5 mM; ( v ) , 5.0 mM; ( v ) , 7.5 raM; ([]), 10.0 raM. The free Mg 2+ concentration was 10 raM. Activity is expressed as ~,mol/min per mg protein.
Substrate analogue studies Further characterization of the reaction mechanism was achieved by the use of the substrate analogues ADP-/3-S and phenyl phosphate. The pattern of inhibition by these two compounds is shown in Figs. 8 and 9 (see also summary in Table I). The ADP analogue, ADP-/3-S, was not a substrate for PK¢, as no pyruvate production was observed following a 15 min incubation of 1 mM ADP-fl-S and 2 mM PEP with 0.015 units of PK c (i.e., about 4-fold the amount of PK c used routinely for an assay). Mixed TABLE I Inhibition patterns displayed by reaction products and substrate analogues of PK c
Abbreviations: Sat, saturating; Ssat, subsaturating; M, mixed inhibition; C, competitive inhibition; NC, non-competitive inhibition. Effector
Variable substrate PEPf
MgATP Pyruvate ADP-/3 -S Phenyl phosphate
ADP r
Sat ADP
Ssat ADP
Sat PEP
Ssat PEP
M C None
M C M
M None C
M NC M
C
C
None
NC
218
y
0.075
0.070
0.050
:E
0
•,~ 0.035 <
0.025
,-4
0.000 0.000 ' -15
~
i
0 1/[PEP][
I
i
I
i
15
30
i
15
30
1/[PEP]I
( m M ) -1
0
150
(raM) -I
Fig. 6. Inhibition of PK c by pyruvate with PEPf as variable substrate at saturating (0.I mM) ADPf. The concentrations of pyruvate were: (0), 0 mM; (•), 25 raM; ( v ) , 50 raM; ( v ) , I00 mM. The free Mg 2+ concentration was 10 raM. Activity is expressed as p.mol/min per mg protein.
inhibition was obtained against PEPf at subsaturating ADPf (Fig. 8A), while no changes in activity were evident at saturating ADPf (0.1 mM). Inhibition with respect to ADPf was competitive at saturating PEPf (Fig. 8B) and mixed at subsaturating PEPf (Table I). Clearly ADP-fl-S does not affect PEP binding to the free enzyme, but is capable of binding to the enzymePEP complex. The PEP analogue, phenyl phosphate [14], was assayed for its effect on PK~, since the high concentrations of pyruvate required to observe inhibition (Fig. 6) may have distorted the enzyme's response to this product by altering the ionic strength of the medium. Inhibition by phenyl phosphate with respect to PEPf was competitive at saturating ADPf (Fig. 9A) or subsaturating ADPf (Table I), while with respect to ADPf it inhibited non-competitively at subsaturating PEPf (Fig. 9B) and did not inhibit (up to 20 raM) at saturating PEPf. Reciprocal plots against ADPf were linear, indicating that ADP cannot bind to the enzyme-phenyl phosphate complex. The results indicate an interaction only with the PEP binding site.
0.024
¢9
<
-
B
0,012 •
0.000 ~ -150
'
I
I/[ADPIr (raM)
'./()() -1
Fig. 8. Inhibition of PK c by ADP-fl-S. (A), PEPf as variable substrate with subsaturating (7 IzM) ADP[; (B), ADPf as variable substrate at saturating (0.5 mM) PEPf. The concentrations of ADP-fl-S were: (©), 0 mM; (o), 2.5 mM; ( v ) , 5.0 raM; ( t ) , 7.5 mM. The free Mg 2+ concentration was 10 mM. Activity is expressed as/zmol/min per mg protein.
0.024
p,
A"
•
i
•
0.016
C~
<
0.008
0.000 -0.025
0.000
1/[PEPlf
0.15
0.025
0.050
( m M ) -1
0.040
••~4 0.10
.4 ~
•
0.020 <
O
<
0.05
0.000
0.00 -300
-o.oto 0
300
600
I/[ADP]f (mM) -t Fig. 7. Inhibition of PK c by pyruvate with ADPf as variable substrate at subsaturating (0.025 mM) PEPf. The concentrations of pyruvate were: (©), 0 raM; (o), 25 raM; ( v ) , 50 raM; ( • ) , 100 raM. The free Mg z÷ concentration was 10 mM. Activity is expressed as ~moi/rnin per mg protein.
,
o.ooo
o.olo
i
0.020
1/[ADP]f (mM) -t Fig. 9. Inhibition of PK c by phenyl phosphate. (A), PEP~ as variable substrate at saturating (0.1 raM) ADPf; (B), ADPf as variable substrate at subsaturating (25 IzM) PEPf. The concentrations of phenyl phosphate were (o), 0 mM; (o), 2.5 mM; ( v ), 5.0 mM; ( • ), 7.5 mM; (C3), 10.0 mM. The free Mg 2+ concentration was 10 mM. Activity is expressed as/xmol/min per mg protein.
219 pEp3-
ADp3-
Mg 2÷
MgATP
Pyruvate
pK c ............................................................................................................ ~PKc
Scheme I
Discussion
A recent study characterizing germinating COS PK c [14] demonstrated this enzyme as unique compared to other plant [2,20], green algal [21], animal [1,3,4] and yeast [3,5] PKs with respect to its kinetic properties. Namely, the COS enzyme presented hyperbolic kinetics for PEP binding, distinguishing it from other PKs recognized as regulatory enzymes [1,3-5,7,17,22]. No activators were identified for germinating COS PK~ [14], in contrast to the spinach leaf [20] and green algal [21] PK. In fact, germinating COS PK~ behaves as a fully active enzyme under a variety of conditions, being sensitive to changes in pH within the physiological range and to a variety of metabolite inhibitors [14]. Kinetic reaction mechanism
Isolated efforts to characterize the kinetic reaction mechanism of a plant PK have had no clear outcome [23,24], perhaps as a consequence of a failure to identify the true substrates. A report on the kinetic properties of the partially purified PK~ and plastidic PK from developing COS suggested a compulsory-ordered, sequential type reaction mechanism for both isoenzymes [24]. The results presented here indicate that germinating COS PK c has a compulsory-ordered Tri-Bi reaction mechanism. The following evidence support this hypothesis. First, K m values for both substrates are not independent of the co-substrate concentration (Fig. 4A and B). Second, PEPf appears to bind independently to PKc, whereas ADPf can bind only if PEPf is already present (Figs. 2 and 3). None of the above would be applicable in a random-order mechanism. Third, although the product inhibition data are not unambiguous, the results can certainly be explained in light of the known properties of PKg. The inhibition by MgATP and ADP-/3-S deviate from the expected behaviour in a compulsory-ordered mechanism when PEPf is varied. At saturating ADP, MgATP should be uncompetitive (if the reaction was reversible to some degree) or not inhibit at all (for an irreversible reaction, which is probably the case with PK~). It has been shown, however, that a high ADP concentration (> 1.5 mM) produces mixed inhibition of PK~ [14]. It is, therefore, plausible for ATP (and more so for ADP-/3-S) to interact with the ADP inhibitory site producing mixed inhibition. ADP-/3-S would also be expected to produce uncompetitive inhibition due to sequestering of the PKc-PEP complex. That it does not could be the result of a masking effect caused by its interaction with
the ADP inhibitory site, as discussed for MgATP, with the addition that it may be unable to bind or have a very low affinity for the catalytic site of the PKc-PEP complex. The reaction mechanism proposed here for the COS PKc is similar to that described for the fructose 1,6-bisphosphate-activated yeast PK [17], but different to the Ping-Pong or rapid equilibrium random-type mechanisms postulated respectively for the enzyme from pig liver and rabbit or Carcinus maenas muscles [25-27]. The substrates are bound in their free form, with PEP being the leading substrate, as shown in Scheme I. As discussed previously for the yeast enzyme [17], the separate binding of Mg 2÷ and its elimination as MgATP (positioned between the y- and /3-phosphate) suggests that the role of the bivalent cation in catalysis is in bridging the /3-phosphate group of ADP and the PEP phosphate group. Lack of an effect of Mg 2÷ on the intrinsic fluorescence of PK c suggests that an independent interaction of this co-factor with the enzyme may be very limited and supports the assertion that Mg 2÷ is not liberated as the free cation. Binding constants are comparatively lower than the values reported for the yeast enzyme for PEPf (K s COS PKc: 8.2 and 22 /zM (fluorometric and kinetic determination, respectively); K s yeast PK: 69 to 87/zM [17,28]) and more notably for ADPf (limiting K m COS PKc: 2.9/xM; g m yeast PK: 840 ~M [17]). Metabolic consequences o f COS P K c kinetic properties
The significance of the current and a related study [14] for germinating COS metabolism is that PK c from this tissue exists in an activated form at all times and that its activity is enhanced in response to anoxia not by activation but rather through concerted decreases in ATP levels, cytosolic pH and concentrations of several inhibitors. The high affinities of germinating COS PKc for its substrates may reflect the particular metabolic needs that differentiate this plant enzyme from other PKs. The fact that only the free forms of ADP and PEP serve as substrates makes it tempting to speculate that fluctuations in cytosolic Mg 2+ levels may regulate the activity of COS PKg. Large changes in metabolite concentrations take place in germinating COS upon anaerobiosis [8]. Notably, estimated cytosolic ATP levels plunge from about 6 to 2 mM [8,14]. The cytosolic pH is also expected to quickly drop [29], not only releasing PK¢ from inhibition by various metabolites, but also allowing different interactions between PEP and ADP and Mg 2+. The resulting redistribution of
220 substrate species could have an influential role in regulating PK c activity. It is important to note that the overall glycolytic rate will be controlled by PK c activity, since PFK activity is largely dependent on the cytosolic concentration of its allosteric inhibitor, PEP.
Acknowledgement This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). References 1 Kayne, F.J. (1973) in The Enzymes (Boyer, P.D., ed.), 3rd Edn., Vol. 8, pp. 353-382, Academic Press, New York. 2 Duggleby, R.G. and Dennis, D.T. (1973) Plant Physiol. 52, 312317. 3 Munday, K.A., Giles, I.G. and Poat, P.C. (1980) Comp. Biochem. Physiol. 67B, 403-411. 4 Plaxton, W.C. and Storey, K.B. (1984) Eur. J. Biochem. 143, 257-265. 5 De Arriaga, D., Busto, F., Del Valle, P. and Soler, J. (1989) Biochim. Biophys. Acta 998, 221-230. 6 Blair, J.B. and Walker, R.G. (1984) Arch. Biochem. Biophys. 232, 202-213. 7 Callens, M., Kuntz, D.A. and Opperdoes, F.R. (1991) Mol. Biochem. Parasitol. 47, 19-30. 8 Kobr, M.J. and Beevers, H. (1971) Plant Physiol. 47, 48-52. 9 Turpin, D.H., Botha, F.C., Smith, R.G., Feil, R., Horsey, A.K. and Vanlerberghe, G.C. (1990) Plant Physiol. 65, 1188-1193
10 Hatzfeld, W. and Stitt, M. (1991) Physiol. Plant. 81, 103-110 11 Copeland, L. and Turner, J.F. (1987) in The Biochemistry of Plants (Stumpf, P.K. and Conn, E.E., eds.), Vol. 11, pp. 107-128, Academic Press, New York. 12 Dennis, D.T. and Greyson, M.F. (1987) Physiol. Plant. 69, 395404. 13 Stryer, L. (1988) in Biochemistry, 3rd Edn., pp. 349-372, Freeman, New York. 14 Podest~, F.E. and Plaxton, W.C. (1991) Biochem. J., 279, 495-501. 15 Plaxton, W.C. (1988) Plant Physiol. 86, 1064-1069. 16 Bock, R.M., Ling, N-S., Morrel, S.A. and Lipfon, S.H. (1956) Arch. Biochem. Biophys. 62, 253-264. 17 McFarlane, N. and Ainsworth, S. (1972) Biochem. J. 129, 10351047. 18 Rodrlguez-Sotres, R. and Mufioz-Clares, R.A. (1990) Arch. Biochem. Biophys. 276, 180-190. 19 Bradford, M.M. (1976) Anal. Biochem. 72, 248-254. 20 Baysdorfer, C. and Bassham, J.A. (1984) Plant Physiol. 74, 374379. 21 Lin, M., Turpin, D.H. and Plaxton, W.C. (1989) Arch. Biochem. Biophys. 269, 228-238. 22 Kahn, A., Marie, J., Garreau, H. and Sprengers, E.D. (1978) Biochim. Biophys. Acta 523, 59-74. 23 Ireland, R.J., DeLuca, V. and Dennis, D.T. (1980) Plant Physiol. 65, 1188-1193. 24 Gupta, V.K. and Singh, R. (1989) Plant Physiol. Biochem. 27, 703-711. 25 McFarlane, N. and Ainsworth, S. (1974) Biochem. J. 139, 499-508. 26 Ainsworth, S. and McFarlane, N. (1973) Biochem. J. 131,223-236. 27 Newton, C.J., Poat, P.C. and Munday, K.A. (1976) Biochem. Soc. Trans. 4, 479-481. 28 Mildvan, A.S. and Cohn, M. (1966) J. Biol. Chem. 241, 1178-1193. 29 Saint-Ges, V., Roby, C., Bligny, R., Pradet, A. and Douce, R. (1991) Eur. J. Biochem. 200, 477-482.