Address correspondence and reprint requests to Dr. L. de Meis at Departamento de Bioquímica Médica, Instituto de Ciências Biomédicas, Universidade Federal do Rio de Janeiro, Cidade Universitária, RJ 21941-590, Brazil.
Abstract : In this work, it is shown that the Ca2+-transport ATPase found in the microsomal fraction of the cerebellum can use both glucose 6-phosphate/hexokinase and fructose 1,6-bisphosphate/phosphofructokinase as ATP-regenerating systems. The vesicles derived from the cerebellum were able to accumulate Ca2+ in a medium containing ADP when either glucose 6-phosphate and hexokinase or fructose 1,6-bisphosphate and phosphofructokinase were added to the medium. There was no Ca2+ uptake if one of these components was omitted from the medium. The transport of Ca2+ was associated with the cleavage of sugar phosphate. The maximal amount of Ca2+ accumulated by the vesicles with the fructose 1,6-bisphosphate system was larger than that measured either with glucose 6-phosphate or with a low ATP concentration and phosphoenolpyruvate/pyruvate kinase. The Ca2+ uptake supported by glucose 6-phosphate was inhibited by glucose, but not by fructose 6-phosphate. In contrast, the Ca2+ uptake supported by fructose 1,6-bisphosphate was inhibited by fructose 6-phosphate, but not by glucose. Thapsigargin, a specific SERCA inhibitor, impaired the transport of Ca2+ sustained by either glucose 6-phosphate or fructose 1,6-bisphosphate. It is proposed that the use of glucose 6-phosphate and fructose 1,6-bisphosphate as an ATP-regenerating system by the cerebellum Ca2+-ATPase may represent a salvage route used at early stages of ischemia ; this could be used to energize the Ca2+ transport, avoiding the deleterious effects derived from the cellular acidosis promoted by lactic acid.
The different SERCA isoforms usually possess a high affinity for ATP, the Ka for ATP binding at the catalytic site being in the range of 10-6-10-5M (Inesi et al., 1967 ; Shigekawa et al., 1976 ; Inesi, 1985 ; Michelangeli et al., 1991 ; Engelender and de Meis, 1996). In previous reports, it was shown that the Ca2+-ATPase from skeletal muscle (SERCA 1) can use glucose 6-phosphate and hexokinase as an ATP-regenerating system (de Meis et al., 1992 ; Montero-Lomeli and de Meis, 1992). The affinity of this Ca2+-ATPase isoform for ATP is sufficiently high to permit the formation of the enzyme-substrate complex in the presence of very small concentrations of ATP formed from ADP and glucose 6-phosphate. Thus, in steady-state conditions, the Ca2+-ATPase cleaves the small amount of ATP available, and hexokinase catalyzes the transfer of phosphate from glucose 6-phosphate to ADP :
1 glucose 6−phosphate + ADP⇆glucose + ATP
2 ATP + HOH⇆ADP + Pi
3 glucose 6−phosphate + HOH⇆glucose + Pi
As a result, during steady state, the Ca2+ transport proceeds as if it was supported by the cleavage of glucose 6-phosphate, a compound that has a much lower standard free energy of hydrolysis than ATP.
In this study, it is shown that the Ca2+-ATPase from cerebellum, in addition to glucose 6-phosphate and hexokinase, can also use fructose 1,6-bisphosphate and phosphofructokinase as an ATP-regenerating system.
Ca2+ uptake was measured by the filtration method using 45Ca2+ and Millipore filters (Chiesi and Inesi, 1979). After filtration, the filters were washed five times with 5 ml of 3 mM La(NO3)3, and the radioactivity remaining on the filters was counted by a liquid scintillation counter. The results shown in the figures are representative experiments that were repeated two to four times with different membrane preparations.
ATPase activity and cleavage of glucose 6-phosphate and fructose 1,6-bisphosphate
These were assayed by measuring the release of 32Pi from either [γ-32P]ATP, [32P]glucose 6-phosphate, or fructose 1,6-[1-32P]bisphosphate. The 32Pi produced was extracted from the medium with ammonium molybdate and a mixture of isobutyl alcohol and benzene (de Meis, 1988). Mg2+-dependent activity was measured in the presence of 2 mM EGTA. The Ca2+-stimulated activity was determined by subtracting Mg2+-dependent activity from the activity measured in the presence of both Mg2+ and Ca2+.
[45Ca]CaCl2 was purchased from New England Nuclear. Pyruvate kinase (EC 22.214.171.124), yeast hexokinase (ATP : d-hexose 6-phosphotransferase ; EC 126.96.36.199), and muscle phosphofructokinase (fructose 6-phosphate kinase ; EC 188.8.131.52) were purchased from Sigma.
Ca2+ uptake in the presence of glucose 6-phosphate and fructose 1,6-bisphosphate
Similar to those derived from skeletal muscle, the vesicles derived from cerebellum are also able to accumulate Ca2+ in a medium containing ADP when glucose 6-phosphate and hexokinase are included in the assay medium. We now show that, in addition to the hexokinase system, the cerebellum Ca2+-ATPase can also use fructose 1,6-bisphosphate and phosphofructokinase as an ATP-regenerating system (Fig. 1). The amount of Ca2+ accumulated by the vesicles using either glucose 6-phosphate and hexokinase or fructose 1,6-bisphosphate and phosphofructokinase was 33.6 ± 1.6 and 66.2 ± 4.8 nmol of Ca2+/mg of protein, respectively. These values are the means ± SE of 12 experiments performed using different vesicle preparations and the same experimental conditions as those described in Figs. 1 and 2. The vesicles were not able to use directly the sugar phosphate, because there was no Ca2+ uptake when ADP was not included in the medium (data not shown) or when either hexokinase or phosphofructokinase was omitted from the medium (Fig. 2). The amount of Ca2+ retained by the vesicles with a saturating ATP concentration (2 mM) was larger than that accumulated with fructose 1,6-bisphosphate, which in turn was larger than that accumulated with the use of glucose 6-phosphate (Fig. 1A). The transport of Ca2+ was associated with a Ca2+-dependent cleavage of either ATP or sugar phosphate (Fig. 1B). Similar to the transport of Ca2+, the rate of cleavage of ATP was faster than that of fructose 1,6-bisphosphate and glucose 6-phosphate. The ratio between the velocities of cleavage and of Ca2+ uptake was found to vary between 20 and 40 regardless of whether 2 mM ATP or a sugar phosphate ATP-regenerating system was used.
Under the conditions of Fig. 3, the concentration of ATP available in the medium varied depending on the concentration of sugar phosphate used (Table 1). In Fig. 4, the values of Ca2+ uptake measured in Fig. 3 were plotted as a function of the ATP concentration that was calculated to be available in the medium at each sugar phosphate concentration used in Fig. 3. This was compared with the Ca2+ measured using micromolar ATP concentrations and the ATP-regenerating system pyruvate kinase/phosphoenolpyruvate. The major difference between the use of phosphoenolpyruvate and the use of sugar phosphate in ATP-regenerating systems is that in the first case, the equilibrium of the regenerating reactions is such that practically all the nucleotide will be in the form of ATP (de Meis et al., 1992 ; Montero-Lomeli and de Meis, 1992), whereas in the latter case, the concentration of ADP is much higher than that of ATP (Table 1). The apparent Km for ATP was found to vary between 0.3 and 0.4 μM regardless of whether the glucose 6-phosphate, fructose, 1,6-bisphosphate, or phosphoenolpyruvate was used to regenerate ATP. The experiment of Fig. 4 shows that (a) there was a good correlation between the ATP added to the medium with phosphoenolpyruvate and the calculated ATP concentrations available when sugar phosphates were used as ATP-regenerating systems and (b) the excess ADP found in the medium when the sugar phosphate system was used to regenerate ATP does not represent an impediment to the transport of Ca2+, i.e., ADP does not impair the formation of the enzyme-substrate complex.
Table 1. ATP concentration in media containing 0.1 mM ADP and different concentrations of sugar phosphate and sugar The ATP concentrations shown in the table were calculated as previously described (Montero-Lomeli and de Meis, 1992) using a Keq of 6.27 × 10-4 for the reaction glucose 6-phosphate + ADP ⇆ ATP + glucose (ΔG° = 4.5 kcal/mol) and a Keq of 10-3 for the reaction fructose 1,6-bisphosphate + ADP ⇆ ATP + fructose 6-phosphate (ΔG° = 4.2 kcal/mol). Notice that in all experiments shown in Figs. 1-6 a small concentration (0.05 mM) of either glucose or fructose 6-phosphate was included in the medium. This was done to facilitate the calculations of ATP in the medium and to avoid the situation where the reaction would flow unidirectionally due to the lack of one of the reactants (de Meis et al., 1992 ; Montero-Lomeli and de Meis, 1992). For the ATP concentrations shown in the table and in the presence of 0.1 mM ADP and 20 mM Pi, the ΔG values for the hydrolysis of ATP vary from -6.3 kcal/mol (ATP = 0.63 μM) to -8.4 kcal/mol (ATP = 20.51 μM). For these calculations, the value of the ΔG° for ATP hydrolysis used was 7.0 kcal/mol.
The different SERCA isoforms have two Km values for ATP (Inesi et al., 1967 ; Shigekawa et al., 1976 ; Inesi, 1985 ; Michelangeli et al., 1991 ; Engelender and de Meis, 1996 ; Mintz and Guillain, 1997). The first is a high-affinity Km (~10-6M) that reflects the binding of ATP to the catalytic site of the enzyme. The second Km (0.1-0.2 mM) reflects the binding of ATP with lower affinity to a regulatory site that accelerates the rate of transport, leading to a significant increase in the total amount of Ca2+ retained by the vesicles. In this study, the amount of Ca2+ retained by the vesicles after saturation of both the catalytic and the regulatory site (Fig. 1A, 2 mM ATP) was found to be about two times larger than that measured with ATP concentrations that allow saturation of only the catalytic site of the Ca2+ -ATPase (Fig. 4). The concentration of ATP available in the medium when the glucose 6-phosphate and fructose 1,6-bisphosphate systems were used was not sufficient to saturate the regulatory site of the Ca2+ -ATPase. However, the maximal amount of Ca2+ accumulated by the vesicles with the fructose 1,6-bisphosphate system was always larger than that measured with either glucose 6-phosphate (Figs. 1-4) or a low ATP concentration and phosphoenolpyruvate (Fig. 4). At present, we do not know why more Ca2+ is accumulated with the fructose 1,6-bisphosphate system. The addition of 4 mM fructose 1,6-bisphosphate to a medium containing 2-10 μM ATP and phosphoenolpyruvate did not lead to an increase of Ca2+ uptake (data not shown), thus excluding the possibility that fructose 1,6-bisphosphate may interact with the regulatory site of the ATPase.
Inhibition of Ca2+ uptake
The concentration of ATP generated by hexokinase, glucose 6-phosphate, and ADP decreases when glucose is added to the medium (Table 1). Accordingly, increasing concentrations of glucose inhibit the Ca2+ uptake supported by glucose 6-phosphate and ADP, and half-maximal inhibition was obtained with 0.4 mM glucose (Fig. 5A). Similarly, when fructose 1,6-bisphosphate and phosphofructokinase were used, the transport was inhibited by fructose 6-phosphate (Table 1 and Fig. 5B). Glucose had no effect on the transport of Ca2+ when 2 mM ATP (Fig. 5A) or the phosphofructokinase system (data not shown) was used. Conversely, fructose 6-phosphate had no effect on the transport of Ca2+ supported by either 2 mM ATP (Fig. 5B) or the hexokinase ATP-regenerating system (data not shown).
Activation of glycolysis during cellular ischemia and energy failure is usually associated with production of lactic acid and a decrease of the intracellular pH to values as low as 6.0 (Hope et al., 1987 ; Westerblad et al., 1991). In brain ischemia, lactic acidosis leads to intracellular Ca2+ overload and cellular death (Nedergaard et al., 1991). Acidification of the medium promotes a decrease of both the Ca2+ affinity and the Vmax of Ca2+ transport of the different Ca2+-ATPase isoforms (Almeida and de Meis, 1977 ; de Meis and Tume, 1977 ; Watanabe et al., 1981 ; Pick and Karlish, 1982 ; Forge et al., 1993 ; Wolosker and de Meis, 1994 ; Mintz and Guillain, 1997 ; Wolosker et al., 1997). These changes of the enzyme properties greatly decrease the ability of the endoplasmic reticulum Ca2+-ATPase to drain the intracellular Ca2+, even in the presence of high ATP concentrations. Glucose 6-phosphate can be formed from glycogen and Pi without the need for consuming ATP. The data presented in this report show that the reversal of the reactions catalyzed by either hexokinase or phosphofructokinase might be used for the draining of Ca2+ from the cytosol during energy failure. The advantage of this system is that during O2 deprivation, only the initial steps of glycolysis would be used to avoid the cytosolic Ca2+ overload, and the dephosphorylation of both glucose 6-phosphate and fructose 1,6-bisphosphate would impair the subsequent reactions of glycolysis, thus avoiding the accumulation of lactic acid in the cytosol. The disadvantage is that less ATP would become available to the cell. After the formation of glucose 6-phosphate from glycogen and Pi, three ATP molecules are formed during degradation of glucose 6-phosphate to lactic acid. If either glucose 6-phosphate or fructose 1,6-bisphosphate is used to regenerate ATP, then two molecules of ATP that could be formed in the subsequent steps of glycolysis are not synthesized. Perhaps, at the early stages of anoxia and before the energy failure, the use of glucose 6-phosphate and fructose 1,6-bisphosphate might represent a salvage route whereby the brain cells may energize the Ca2+-ATPase, maintaining a low cytosolic Ca2+ concentration and avoiding the deleterious effect derived from lactic acidosis.
Oxalate (4-10 mM) and Pi (10-20 mM) are usually included in the assay medium in most studies of Ca2+ transport by vesicles derived from the endoplasmic reticulum of muscle, brain, or blood platelet. These anions diffuse through the membrane and form calcium oxalate or calcium phosphate crystals inside the vesicles, thus decreasing the free Ca2+ concentration inside the vesicles and increasing the total amount of Ca2+ retained by the vesicles (Hasselbach, 1964 ; de Meis et al., 1974 ; de Meis, 1981 ; Engelender and de Meis, 1996). The free Ca2+ concentration inside the vesicles is determined by the solubility product of the salt precipitated and is in the range of 2-10 mM when Pi is used and in the range of 10-20 μM when oxalate is used. Because of this large difference in intravesicular Ca2+ concentration, the rate of Ca2+ efflux measured in the presence of Pi is faster than that measured in the presence of oxalate. By using Pi as a Ca2+ precipitating agent, the ratio between the rates of substrate hydrolysis and of Ca2+ uptake was found to vary between 20 and 40 in sarcoplasmic reticulum and blood platelets (Cardoso et al., 1997) and in vesicles derived from the cerebellum (Fig. 1). This value does not represent the true coupling stoichiometry between substrate hydrolysis and Ca2+ accumulation. Due to the high concentration of free Ca2+ inside the vesicles, during the different incubation intervals, part of the Ca2+ previously accumulated leaks from the vesicle and more substrate is cleaved both to maintain the gradient and to allow further Ca2+ accumulation. For the sarcoplasmic reticulum Ca2+ -ATPase, it has been shown that two Ca2+ ions are translocated across the membrane after the cleavage of each ATP molecule. This was determined by using either a high oxalate concentration to decrease both the intravesicular Ca2+ concentration and the Ca2+ efflux to low values (Hasselbach and Makinose, 1961, 1963 ; Hasselbach, 1964) or, with more precision, by measuring in transient kinetic experiments the number of Ca2+ ions translocated across the membrane after a single catalytic cycle of the enzyme (Verjovski-Almeida et al., 1978 ; Fernandez-Belda et al., 1984 ; Inesi, 1985). The amount of Ca2+ accumulated by the cerebellum vesicles in the presence of 4 mM oxalate is small and difficult to measure, because the permeability of the membrane for oxalate is smaller than that of the sarcoplasmic reticulum vesicles (de Meis et al., 1970 ; Trotta and de Meis, 1975) and, as far as we know, the cerebellum Ca2+ -ATPase has not yet been purified to the extent that it can be used for transient kinetic experiments. Thus, at present, we do not know if the true stoichiometry of transport of the cerebellar Ca2+ -ATPase is the same as that of the muscle ATPase.