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: (
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: (
