Search

The urea cycle occurs widely in vertebrates, serving different functions in different lineages. In elasmobranchs, where it contributes to osmolar balance, the product urea itself is functionally useful. In other vertebrates the synthesis of urea serves functions other than provision of urea. In air-breathing fish, amphibians, and mammals ureagenesis generates protons that contribute to pH homeostasis by titration of metabolically produced bicarbonate, and in ruminants this function is supplemented by large-scale transport of bicarbonate in the form of urea from tissues to the rumen. Disposal of ammonium in ureotelic species generally and delivery of ammonium to the rumen in ruminants appear to be consequences of pH-modulated ureagenesis rather than primary functions. Because only regulatory changes are required for its development from the arginine synthetic pathway and because it consumes no metabolites and relatively little ATP, the urea cycle is a uniquely suitable source of the protons that are needed...

The activity of rat liver glutaminase from sedimented fractions of freeze-thawed mitochondria is strongly affected by variation in the Mg 2+ concentration within the approximate physiological range of activators. A rise in the Mg 2+ concentration stimulates glutaminase by increasing the apparent affinity of the enzyme for its positive modifier phosphate. With the addition of 4 mM Mg 2+ the M 0.5 for phosphate activation decreased from 18 to 9.5 mM at pH 7.1, 10 to 5.8 mM at pH 7.4 and 6.4 to 4.0 mM at pH 7.7. The result is an increase in the apparent affinity of the enzyme for glutamine. With the addition of 4 mM Mg 2+ the S 0.5 of glutaminase for glutamine decreased from 24 to 13 mM at pH 7.1, 14 to 9.6 mM at pH 7.4, and remained unchanged at 8.2 mM at pH 7.7. Since Mg 2+ stimulates glutaminase, as does a rise in pH (Szweda, L.I. and Atkinson, D.E. (1989) J. Biol. Chem. 264, 15357–15360), by increasing the apparent affinity of the enzyme for phosphate, it reduces the inhibitory effect of a decrease in pH and/or phosphate concentration over a physiologically relevant range.

In Responding to the invitation to comment on some aspects of the symposium, I will attempt no evaluation of the internal logic or mathematical merits of the various mathematical treatments that have been discussed. Those and other characteristics of these treatments have been discussed by other participants. Rather, I will focus my comments mainly on two questions that Dr. Cornish-Bowden asked me to consider, and which he posed in his introductory presentation (see Chapter 1): (a) what characteristics should a theory have if it is to be useful to experimentalists and (b) are the main treatments that have been discussed here — those of Kacser and Burns (see especially Chapters 3, 17, 20 and 25), Heinrich and Rapoport (Chapter 28), and Savageau (Chapters 4 and 5) — valuable to experimental metabolic biochemists?

In his preparations for this symposium, Athel Comish-Bowden mentioned a desire to present control theory to experimentalists in such a way as to persuade them that their subject could advance more rapidly with more attention to theoretical ideas. He is also inviting some experimentalists to indicate what they think a useful theory should offer, and has asked me to attempt to assess in the final chapter the extent to which the others have offered experimentalists a workable approach to metabolic control.

Rat heart mitochondria oxidizing pyruvate (in the presence of 20% as much malate) took up nearly the amount of oxygen required for complete oxidation to CO2. Thus pyruvate, a physiological substrate of the citrate cycle, is oxidized through the entire cycle in these mitochondria, and they seem suitable for study of regulation of integrated mitochondrial energy transduction. By addition of graded amounts of hexokinase or pyruvate kinase to the suspending medium (in the presence of excess glucose or phosphoenolpyruvate), a wide range of steady-state values of the ATP ADP concentration ratio was obtained. At a constant concentration of phosphate, the steady-state rate of oxygen uptake by rat heart mitochondria oxidizing pyruvate was a function of the adenylate energy charge or of the ATP ADP ratio, and relatively independent of the absolute concentrations of these nucleotides. The oxygen uptake rates typically spanned a range of about 20-fold. At very high values of the ATP ADP ratio, the rate of oxygen uptake is much lower than the “state 4” rate seen after added ADP has been phosphorylated. This result suggests that “state 4” respiration, at least in these freshly prepared mitochondria, measures the rate at which ADP is made available by ATPase activity, rather than indicating uncoupling of electron transport from phosphorylation. The concentration of orthophosphate affected the rate of oxygen uptake and the pattern of response to the ATP ADP ratio or the energy charge, but the effects did not seem interpretable in terms of the mass-action expression for hydrolysis of ATP, (ATP ADP) (Pi.

Catabolism of protein liberates HCO 3 − , which cannot be eliminated in the necessary amounts through the lungs, kidneys, or intestines, and thus poses a threat of alkalosis to air-breathing animals. That threat is met by biosynthetic sequences that consume the weak acid NH 4 + and liberate its proton for reaction with HCO 3 − . Different kinds of air-breathing animals use different syntheses for this purpose; mammals utilize ureagenesis. Modulation of the rate of urea synthesis in response to pH is an important part of the interacting regulatory systems by which pH homeostasis is maintained in mammals.

In the physiological range of the adenylate energy charge in liver (0.7 to 0.9) the rate of deamination of AMP catalyzed by rat liver adenylate deaminase (EC 3.5.4.6) increases sharply with decreasing energy charge. It is suggested that this response serves to protect against sharp transient decreases in energy charge: when the charge decreases the resulting removal of AMP by deamination will oppose the decrease in charge (the mole fraction of ATP plus half the mole fraction of ADP). The activity of the enzyme decreases sharply as the size of the adenine nucleotide pool decreases in and below the physiological range. This effect may be a self-limiting response to prevent excessive depletion of the pool. These suggestions, based on the properties of the enzyme observed in vitro, are consistent with the results of experiments on liver in vivo reported by other workers.

We have used the technique of continuous culture to study the expression of β-galactosidase in Escherichia coli . In these experiments the cultures were grown on carbon-limited media in which half of the available carbon was supplied as glycerol, glucose, or glucose 6-phosphate, and the other half as lactose. Lactose itself provided the sole source of inducer for the lac operon. The steady-state specific activity of the enzyme passed through a maximal value as a function of dilution rate. Moreover, the rate at which activity was maximal (0.40 h −1 ) and the observed specific activity of the enzyme at a given growth rate were found to be identical in each of the three media tested. This result was unexpected, since the steady-state specific activity can be shown to be equal to the differential rate of enzyme synthesis, and since it is known that glycerol, glucose, and glucose-6- P -cause different degrees of catabolite repression in batch culture. The differential rate of β-galactosidase synthesis was an apparently linear function of the rate of lactose utilization per milligram protein regardless of the composition of the input medium. That is, it is independent of the rate of metabolism of substrates other than lactose which are concurrently being utilized and the enzyme level appears to be matched to the metabolic requirement for it. If this relationship is taken to indicate the existence of a fundamental control mechanism, it may represent a form of attenuation of the rate of β-galactosidase synthesis which is independent of cyclic AMP levels.

The activity of rat liver glutaminase from sedimented fractions of freeze-thawed mitochondria is strongly affected by variation in pH over a physiologically relevant range at approximate physiological concentrations of activators. As pH increases from 7.1 to 7.7 at 0.7 mM ammonium and 10 mM phosphate, the S0.5 for glutamine decreases 3.5-fold, from 38 to 11 mM. This results in an 8-fold increase in reaction velocity at 10 mM glutamine. In addition, the M0.5 for phosphate activation decreases from 21 to 8.9 mM as pH increases from 7.1 to 7.7. This apparent effect of pH on the affinity of glutaminase for phosphate is similar to previous reports of the pH effect on activation by ammonium (Verhoeven, A. J., Van Iwaarden, J. F., Joseph, S. K., and Meijer, A. J. (1983) Eur. J. Biochem. 133, 241-244; McGivan, J. D., and Bradford, N. M. (1983) Biochim. Biophys. Acta 159, 296-302). Glutaminase does not respond to variation in pH between 7.1 and 7.7 when phosphate and ammonium are saturating. The effects of the two modifiers are additive. Each is still effective, as is pH, when the other is saturating. Therefore, it appears that the effects of pH on the apparent affinity of the enzyme for ammonium and phosphate account for the enzyme's response to pH. These results may help explain previous reports of minimal effects of pH on glutaminase at saturating concentrations of related substances (McGivan, J. D., Lacey, J. H., and Joseph, K. (1980) Biochim. J. 192, 537-542; Horowitz, M. L., and Knox, W. E. (1968) Enzymol. Biol. Clin. 9, 241-255; McGivan, J. D., and Bradford, N. M. (1983) Biochim. Biophys. Acta 759, 296-302). Glutaminase binds glutamine cooperatively with Hill coefficients ranging from 1.7 to 2.2, which suggests at least two and probably three or more interacting binding sites for glutamine. The strong response of liver glutaminase to pH and the fact that the reaction can supply metabolites for urea synthesis suggest a possible regulatory role of glutaminase in ureagenesis.

Simple computational and graphic models were used to illustrate necessary design constraints in the evolution of two aspects of metabolic control: regulation at the catalytic site by ratios of metabolic coupling factors such as the ATP/ADP/AMP system and the pyridine nucleotides, and endproduct feedback inhibition. The observed constancy of the ratios of the nucleotide coupling agents, especially of the adenylates, indicates that control in vivo must be exerted primarily by ratios of the components of the pools, and must be insensitive to pool size. Such response of appropriate enzymes has been observed in vitro. The model illustrates by graphs the effects of variation in enzyme affinity fo the nucleotide coupling agents, and of variation in the ratios of these affinities. The graphs demonstrate the previously recognized, but not as formally demonstrated, need for high affinity (low Michaelis constants relative to physiological concentrations) and for higher affinity for the product nucleotide than for the reactant. Similar calculations for enzymes subject to negative feedback control demonstrated that, even if the effect of a modifier on the affinity of binding of substrate is very large, a sensitive response to modifier concentration cannot be attained if the reaction catalyzed is first order with respect to substrate and modifier. The cooperativity that is observed in such cases is thus a necessary design feature without which effective negative feedback could not have evolved.

The reaction catalyzed by calf liver uridine diphosphate glucose synthase (pyrophosphorylase) (EC 2.7.7.9; UTP + glucose 1-phosphate = UDP-glucose + PPi) is an example of an enzymic reaction in which a nucleoside triphosphate other than ATP is the immediate source of metabolic energy. Kinetic properties of the enzyme, acting in the direction of UCP-glucose formation were investigated in vitro. The reaction was inhibited by UDP-glucose (0.072), Pi (11), UDP (1.6), UDP-xylose (0.87), UDP-glucuronate (1.3), and UDP-galacturonate (0.95). The numbers in parentheses indicate the concentration (mM) required for half-maximal inhibition under the conditions used. Other compounds tested, including ATP, ADP, and AMP, had no effect. Over a range of concentrations of UTP (0.04-0.8 MM) and UDP-glucose (0.05-0.03 mM), the reaction rate was more dependent on the concentration ratio [UDP-glucose]/[UTP] than on the absolute concentration of either compound. Comparison of the kinetic properties in vitro with estimates of metabolite levels in vivo suggests that (1) the enzyme operates in a range far from its maximal rate, and (2) the concentrations of glucose 1-phosphate and Pi and the ratio [UDP-glucose]/[UTP] may be the most important determinants of UDP-glucose synthase activity.

The isoelectric points, ultraviolet absorption spectra, and electrophoretic mobilities between pH 4.15 and 8.0 of eight strains of tobacco mosaic virus are presented. The results confirm an earlier classification of these strains into four groups. Some of the differences in ultraviolet spectra and in slope of the mobility-vs-pH curves are correlated with differences in tyrosine, tryptophan, and histidine content. Strains within a group do not differ in any of the properties determined. Infrared absorption spectra of all strains are essentially identical.

Kemp, John D. (University of California, Los Angeles), and Daniel E. Atkinson . Nitrite reductase of Escherichia coli specific for reduced nicotinamide adenine dinucleotide. J. Bacteriol. 92: 628–634. 1966.—A nitrite reductase specific for reduced nicotinamide adenine dinucleotide (NADH 2 ) appears to be responsible for in vivo nitrite reduction by Escherichia coli strain Bn. In extracts, the reduction product is ammonium, and the ratio of NADH 2 oxidized to nitrite reduced or to ammonium produced is 3. The Michaelis constant for nitrite is 10 μ m . The enzyme is induced by nitrite, and the ability of intact cells to reduce nitrite parallels the level of NADH 2 -specific nitrite reductase activity demonstrable in cell-free preparations. Crude extracts of strain Bn will also reduce hydroxylamine, but not nitrate or sulfite, at the expense of NADH 2 . Kinetic observations indicate that hydroxylamine and nitrite may both be reduced at the same active site. The high apparent Michaelis constant for hydroxylamine (1.5 m m ), however, seems to exclude hydroxylamine as an intermediate in nitrite reduction. In vitro activity is enhanced by preincubation with nitrite, and decreased by preincubation with NADH 2 .

Phosphoenolpyruvate synthetase (PEP synthetase) and the pyruvate dehydrogenase complex, both partially purified from lactate-grown Escherichia coli B, were assayed simultaneously in the same reaction mixture to evaluate some of the factors that must affect the partition of pyruvate between degradation and biosynthesis in vivo. The enzymes responded independently to control parameters; thus, under assay conditions in vitro there were no protein-protein interactions that affected the kinetic properties of the enzymes. When equal activities of the enzymes were present in the absence of modifiers, the two reactions proceeded at equal velocities (corresponding to equipartition of pyruvate between degradation and biosynthesis) at an energy charge value of about 0.85. At higher values of charge, conversion to PEP (which would lead in vivo to biosynthesis) predominates, and at lower values of charge the partition favors conversion to acetyl coenzyme A (which would lead in vivo to degradation of substrate and regeneration of ATP). Inhibitors of PEP synthetase (PEP, oxalacetate, and α-ketoglutarate) alter the partition ratios, increasing the fraction converted to acetyl coenzyme A at all values of energy charge. Acetyl coenzyme A, which inhibits the first enzyme of the pyruvate dehydrogenase complex, has the opposite effect on the partition ratio. In cultures growing on glucose, PEP synthetase has no apparent function, and the ratio of pyruvate dehydrogenase activity to that of PEP synthetase was between 10 and 20 in such cultures. In lactate-grown cultures this ratio was as high as 4 at early stages, but fell to 0.5 during growth. These variations in relative enzyme levels, together with the responses to adenylate energy charge and feedback inhibition reported here, seem well adapted for controlling the partitioning of pyruvate between degradation and biosynthesis as appropriate to the momentary metabolic needs of the cell.

A possible mechanism for the frequently proposed involvement of phosphofructokinase in regulation of the rate of carbohydrate metabolism was suggested by the finding of Lardy and Parks (1) that this enzyme is inhibited by excess adenosine triphosphate. Mansour and Mansour (2) showed phosphofructokinase of the liver fluke to be stimulated by cyclic adenosine 3’,5’-monophosphate. Inhibition of the corresponding enzyme from guinea pig heart by ATP is overcome by 3’, 5’-AMP, 5’-AMP, or ADP (3). Similar properties of phosphofructokinase from skeletal muscle were suggested as a possible explanation of the Pasteur effect (4). Since 3’,5’-AMP seems to be generally associated with hormonal control mechanisms, we investigated the properties of phosphofructokinase from yeast, the classical organism for study of the Pasteur effect and one presumably lacking hormones. The effect of 5’-AMP on the kinetics of yeast DPN-specific isocitrate dehydrogenase (5) had led us to suspect that this nucleotide may play an important role in the control of energy metabolism in yeast. The results presented here suggest that the ratio of the concentrations of AMP and ATP may regulate the activity of phosphofructokinase in yeast.

1. 1.Citrate synthase (citrate oxaloacetate-lyase (CoA-acetylating), EC 4.1.3.7) has been partially purified from beef liver, beef heart and Escherichia coli. The beef heart and beef liver enzymes were quite similar in their pH optima, substrate affinities and sensitivity to ATP inhibition. The E. coli enzyme had a lower substrate affinity and had a lower Ki for ATP than the mammalian enzymes. The E. coli citrate synthase differed markedly from the mammalian enzymes in its response to pH changes. 2. 2.ATP was competitive with respect to CoASAc with the beef liver and beef heart enzymes. In E. coli, ATP was not competitive with CoASAc but changed both V and Km for oxaloacetate. Mg+ also inhibits the activity of the E. coli citrate synthase and tends to relieve the ATP inhibition. 3. 3.The ATP inhibition of citrate synthase may act in concert with AMP and ADP stimulation of isocitrate dehydrogenase and with citrate stimulation of CoASAc carboxylase in partitioning CoASAc between oxidation by way of the citric acid cycle and storage as fat. It may also play a role in regulating liver ketone body formation.

Citrate synthase from yeast has been purified to apparent homogeneity. The reaction catalyzed by this enzyme followed normal kinetics with respect to both oxaloacetate and acetyl-CoA. The K m values for the two substrates were 1 to 4 μ m and 2 to 6 μ m , respectively. ATP (5 m m ) increased the K m for acetyl-CoA to about 100 μ m . Both ATP and pyrophosphate competitively inhibited the reaction with respect to acetyl-CoA. These properties are consistent with the suggestion that modulation by ATP of the catalytic properties of citrate synthase is a factor in regulation of the rate of oxidation of acetyl-CoA in the citrate cycle.

Summary The adenine nucleotides participate in every extended metabolic sequence, and thus serve as stoichiometric coupling agents linking all of the chemical activities that constitute life. Each metabolic sequence may be assigned an ATP coupling coefficient—the number of molecules of ATP regenerated (or used) when a mole of substrate is degraded or a mole of product formed. The ATP coupling coefficient for any metabolic conversion is an evolved phenotypic characteristic of the organism, and it determines the thermodynamically feasible direction of conversion under physiological conditions. Thus the unidirectionality of metabolic sequences is a direct consequence of the values of their ATP coupling coefficients. Using these coupling coefficients, “prices,” expressed in ATP equivalents, may be assigned to metabolites. Metabolism may thus be viewed as a problem in the allocation of resources. This approach may be pedagogically useful in focusing attention on metabolic interrelation-ships, rather than on individual sequences for their own sakes. It also explains the need for universal kinetic control of metabolic sequences by the energy charge of the adenine nucleotide pool.

Phosphoenolpyruvate synthetase from Escherichia coli B cells grown with lactate as sole carbon source responds to variation in the adenylate energy charge (the mole fraction of ATP in the adenylate nucleotide pool plus one-half the mole fraction of ADP) in the manner characteristic of enzymes in biosynthetic sequences. This response is appropriate, since conversion of pyruvate to phosphoenolpyruvate is prerequisite to all biosynthetic sequences in lactate-grown cells. The response to energy charge is modulated by feedback inhibition by intermediates in various biosynthetic sequences: α-ketoglutarate, oxalacetate, phosphoenolpyruvate, 3-phosphoglyceraldehyde, and ADP-glucose. In conjunction with the previously demonstrated oppositely-directed response of pyruvate dehydrogenase to energy charge, these effects should assure the proper partitioning of pyruvate between biosynthesis (for which conversion to phosphoenolpyruvate is the first step) and oxidation with concomitant regeneration of ATP (for which conversion to acetyl coenzyme A is the first step).

Activity of the pyruvate dehydrogenase complex (pyruvate + DPN+ + HSCoA → acetyl-SCoA + DPNH + CO2 + H+) of Escherichia coli is stimulated by glycolytic intermediates. Fructose-di-P is the most effective modifier, and activity is enhanced also by fructose-6-P, glyceraldehyde-3-P, phosphoenolpyruvate, dihydroxyacetone-P, and glucose-6-P. These effects are exerted on the pyruvate decarboxylase component of the complex. Fructose-di-P overcomes the inhibition resulting from a high value of the adenylate energy charge. The activity of the complex is modulated also by the oxidation level of the DPN+-DPNH pool, increasing steeply with increase in the DPN+ mole fraction in the physiological range around 0.8. The responses to the adenylate energy charge, the DPN+ mole fraction, and the acetyl coenzyme A concentration are in the right direction, and the interactions are of the right type, to suggest that modulation of pyruvate dehydrogenase activity contributes to the stabilization of these parameters in vivo.

INTRODUCTION • . ' " . . • • . . . . . • . . . . . . . • . . • • . • . . . . . . . . . • . . . . . . . . . . . . . . . . . 47 Experimental test in vivo of suggested regulatory effects and interactions. . . . . . 48 Molecular basis for regulatory phenomena .. '" . . . .. . . . . . . .. . . . . . . . . . . . . . 49 Unidirectional inhibition or stimulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 METABOLIC REGULATION-GENERAL CONSIDERATIONS. • • . . . . . . . . . . . . . . . . . . 51 Adenylate control. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 REGULATION OF GLYCOLYSIS AND THE CITRATE CYCLE.. ......... 56 Pyruvate kinase. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Lactic dehydrogenase. . .. .. .. ... . .. ... .. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Pyruvate dehydrogenase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Citrate synthetase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Isocitrate dehydrogenase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3-C to 4-C conversions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Glyoxylate cycle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Glutamate dehydrogenase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 BroSYNTHETIC FEEDBACK INHIBITION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Aspartokinase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

The adenosine diphosphate glucose synthase (pyrophosphorylase) of Escherichia coli responds very sharply to variation in the energy charge of the adenylate pool. The response is in the right direction to insure that this enzyme, which catalyzes the first step unique to glycogen storage in E. coli, will have appreciable activity only under conditions of energy excess (high energy charge). The positive response to high energy charge is strongly enhanced by reduced triphosphopyridine nucleotide, fructose 1,6-diphosphate, 3-phosphoglyceraldehyde, and phosphoenolpyruvate. High concentrations of these metabolites seem appropriate as signals for energy storage.

The acellular slime mold Physarum polycephalum releases a soluble adenosine 3', 5'-monophosphate phosphodiesterase into the growth medium. Although this enzyme resembles the particulate diesterase of the same organism in kinetic properties and in inhibition by methyl purines, its greater stability, its insensitivity to stimulation by imidazole and to inhibition by adenosine triphosphate,- and its selective release into the medium indicate a specific function, perhaps protection against exogenous cyclic nucleotide, for the soluble enzyme.

The citrate cleavage enzyme (EC 4.1.3.8) of rat liver is inhibited by adenosine diphosphate, which appears to compete with adenosine triphosphate. This effect may ensure that fatty acids are produced only when the ATP level is high. The "energy charge" of the adenylate system, defined as (ATP + ½ ADP)/(AMP + ADP + ATP), is proposed as a fundamental metabolic control parameter. Enzymes that utilize ATP and are inhibited by ADP or AMP will yield steep curves of velocity as a function of energy charge (resembling the steep curves of velocity as a function of substrate concentration that are characteristic of many regulatory enzymes) even in the absence of multiple sites and cooperative binding.