See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/26768062 Metabolic systems maintain stable nonequilibrium via thermodynamic buffering Article in BioEssays · October 2009 DOI: 10.1002/bies.200900057 · Source: PubMed CITATIONS READS 19 240 2 authors: Abir U Igamberdiev Leszek A Kleczkowski 219 PUBLICATIONS 3,407 CITATIONS 108 PUBLICATIONS 3,304 CITATIONS Memorial University of Newfoundland SEE PROFILE Umeå University SEE PROFILE Some of the authors of this publication are also working on these related projects: Several research projects on the topics of plant science, bioenergetics and theoretical biology View project All content following this page was uploaded by Leszek A Kleczkowski on 11 January 2017. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately. DOI 10.1002/bies.200900057 Problems and paradigms Metabolic systems maintain stable non-equilibrium via thermodynamic buffering Abir U. Igamberdiev,1* and Leszek A. Kleczkowski2 1 2 Department of Biology, Memorial University of Newfoundland, St. John’s, NL, A1B 3X9, Canada Department of Plant Physiology, Umeå Plant Science Centre, University of Umeå, 901 87 Umeå, Sweden Here, we analyze how the set of nucleotides in the cell is equilibrated and how this generates simple rules that help the cell to organize itself via maintenance of a stable non-equilibrium state. A major mechanism operating to achieve this state is thermodynamic buffering via high activities of equilibrating enzymes such as adenylate kinase. Under stable non-equilibrium, the ratios of free and Mg-bound adenylates, Mg2R and membrane potentials are interdependent and can be computed. The adenylate status is balanced with the levels of reduced and oxidized pyridine nucleotides through regulated uncoupling of the pyridine nucleotide pool from ATP production in mitochondria, and through oxidation of substrates non-coupled to NADR reduction in peroxisomes. The set of adenylates and pyridine nucleotides constitutes a generalized cell energy status and determines rates of major metabolic fluxes. As the result, fluxes of energy and information become organized spatially and temporally, providing conditions for self-maintenance of metabolism. Keywords: adenylate kinase; metabolomics; regulated uncoupling; respiration; stable non-equilibrium; thermodynamic buffering Introduction The field of metabolism, complex as it is, can be hardly reduced to simple rules. However, the pathways of cellular metabolism are almost identical across widely divergent organisms, which may simplify finding the basic principles of metabolome organization and functioning.(1) Actual rates of metabolic fluxes depend on energy (stored and supplied mainly via ATP) and redox potentials (estimated by reduction levels of NAD for catabolic reactions and NADP for anabolic pathways). Several concepts of cell energy status have been developed, including adenylate energy charge [(ATP þ ½ADP)/(ATP þADP þAMP)],(2) catabolic redox charge [NADH/(NADH þ NADþ)] and anabolic redox charge [NADPH/(NADPH þ NADPþ)].(3) These ratios have some relation to the state of the cell and to the rate of metabolic *Correspondence to: A. U. Igamberdiev, Department of Biology, Memorial University of Newfoundland, St. John’s, NL, A1B 3X9, Canada. E-mail: igamberdiev@mun.ca BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. fluxes,(4,5) but they are too general and do not include major factors governing these processes, e.g., free magnesium concentration in the case of the adenylate energy charge.(6–8) Although living systems operate far from equilibrium,(9) non-equilibrium fluxes should be steady. This stable nonequilibrium state(10) can be reached via a ‘‘fitting function’’ of special thermodynamic buffering enzymes that equilibrate fluxes of load and fluxes of consumption of major metabolic components, e.g., ATP and pyridine nucleotides.(11) Thermodynamic buffering generates steady metabolic fluxes via local equilibrations and regulated uncoupling or ‘‘slippage’’.(12) Understanding the rules of such equilibration will help in developing a computational approach for calculating major parameters of metabolism. With such approach, the primary metabolomic data could be used for calculating metabolic parameters, similar to the deciphering of huge number of genes and proteins using the genomic and proteomic data. The equilibration of fluxes leading to stable non-equilibrium is established in seconds and results in the exchange of metabolic material between catabolic and anabolic pathways. In this process, the established near-equilibria manifest themselves as stoichiometric linkage between different pools of metabolites so that they cease to be independent variables.(13) The dynamic equilibrium reached via thermodynamic buffering results in the appearance of sustained oscillations in the systems.(14,15) The resulting ordered complexity of biological systems out-performs the simplicity of physical or chemical systems.(16) Here, we analyze how the set of nucleotides is equilibrated in the cell and how this generates simple rules helping the cell to organize itself via maintenance of a stable non-equilibrium state. This is an important condition for self-maintaining metabolism, which operates in a homeodynamic rather than homeostatic mode.(17,18) The stable non-equilibrium state is supported through thermodynamic buffering, in which the ratios of adenylates and pyridine nucleotides follow thermodynamically and kinetically defined computable principles. This allows determination of the rates and regulation of metabolic pathways from the basic nucleotide ratios governing energy metabolism. 1091 Problems and paradigms Buffering of intermediates The buffering of energy intermediates is the most widespread source of free energy in biological systems.(19) Filling buffer reservoirs corresponds to the accumulation of free energy. The most striking example is the bicarbonate buffering via carbonic anhydrase. This plays a significant role in many physiological processes from carbon fixation in photosynthesis to respiration in animals.(20) This buffer operates effectively at a pH close to neutral, as does phosphate buffer. The role of phosphate in biological systems is of particular importance, as the internal instability of the pyrophosphate (PPi) bond was used in early evolution for energy purposes. This type of bond became the basis of energy-rich compounds such as ATP and other nucleoside triphosphates. The breakage of terminal phosphoanhydride bond results in a local acidification, and protons stored in phosphate buffer are released for catalysis (enzymes) or mechanical movement (actin).(21) This reaction can then be coupled with thermodynamically unfavorable reactions to give an overall negative (spontaneous) DG for the reaction sequence. The actual value of DG for ATP hydrolysis varies, primarily depending on Mg2þ concentration, and under normal physiological conditions is close to –50 kJ/mol.(22–25) By using ATP (or even PPi) in enzymatic reactions or cytoskeleton movement, the release of energy from phosphate buffer becomes vectorized.(19) Vectorization is achieved by the use of ATP for local changes in pH during enzymatic reactions through the link between ATP hydrolysis (accompanied by the release of proton) and an endergonic metabolic process. Therefore, the splitting of energy-rich bonds leads to local (in time and space) pH changes accompanying hydrolysis of these bonds. This results in driving of the conformational changes in protein molecules and in the maintenance of mechanical macromolecular action. While simple inorganic buffers store energy without vectorization, the adenylate system itself needs buffering to make vectorization most efficient. This buffering is not possible with the PPi system, but the adenylate system includes three acting species, ATP, ADP, and AMP. This allows using the most dephosphorylated species (AMP) for fine regulation of the ATP consumption/production and for the maintenance of ADP concentration at the level sufficient to support ATP synthesis by ATP synthase.(26) The major process of ATP synthesis (oxidative phosphorylation) has a fluctuating load conductance, so it needs a maximal buffering of the ADP supply from the intermembrane space of mitochondria. It was experimentally shown that the oxidative phosphorylation obeys linear and symmetric relations between flows and forces, reminiscent of the classical Onsager relations.(27,28) It can operate at optimal efficiency only if the conductance of the load, i.e., the ATP utilizing 1092 A. U. Igamberdiev and L. A. Kleczkowski reactions in a living cell, is exactly matched by the output conductance of oxidative phosphorylation.(11,29) Because the load conductance in a living cell is rarely constant but fluctuates dramatically, oxidative phosphorylation could hardly ever operate at optimal efficiency due to a violation of conductance matching.(30) A way out of this dilemma is in the introduction of reversible ATP-utilizing reactions, such as the ones catalyzed by adenylate kinase (AK) or creatine kinase (CK), which have been proposed to act as thermodynamic buffers.(31) Thermodynamic buffer enzymes represent a bioenergetic regulatory principle for the maintenance of a regime that is far from equilibrium. The intermembrane space of mitochondria (containing a high activity of AK and, in animals, of CK) is a major compartment for such buffering, acting as a filter where the adenylate concentration is adjusted to a correct value before being handed over by adenylate translocator. It acts as a linear energy converter maintaining the linearity of oxidative phosphorylation within a physiological range. Due to thermodynamic buffer enzymes, linear non-equilibrium thermodynamics becomes applicable to biological systems, and metabolic fluxes can be derived from the concentrations of nucleoside phosphates and metal ions established under such equilibrium. Computability of the metabolic system due to thermodynamic buffering Thermodynamic buffering of adenylates Within the framework of the theory of dissipative structures,(9) thermodynamic buffering(31) represents the basic regulatory principle for the maintenance of a stable regime, i.e., far from equilibrium where the production of energy is minimal. Due to ATP production in oxidative phosphorylation, the phosphate potential is shifted far away from equilibrium, and, since ATP drives cell metabolism, this shifts all other potentials into a farfrom-equilibrium regime. The load of metabolism by ATP can fluctuate depending on the state of the cell and its environment. To satisfy the condition of stable non-equilibrium, it should be optimized by the efficiency of ATP production fluctuating in concert with fluctuations in ATP consumption. Deviations from optimal efficiency can be diminished by the action of enzymes that generate sensitive thermodynamic buffering.(31–34) Every reversible ATP utilizing reaction can act as a thermodynamic buffer, e.g., the AK reaction reversibly utilizes MgATP in the reaction with free AMP and produces one MgADP and one free ADP.(7) This type of buffering effectively equilibrates the pool of adenylates. Such an important function as thermodynamic buffering should have its own compartments associated with it. The BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. A. U. Igamberdiev and L. A. Kleczkowski compartment(s) should be in close proximity with the locations of ATP synthesis and consumption. This is satisfied for intermembrane space of mitochondria and (in plants only) for intermembrane spaces of chloroplasts. In fact, the mitochondrial compartment exhibits extremely high activity of AK, e.g., it is two to four times higher than the activity of ATP synthase in plant mitochondria,(35) and the activities of associated enzymes such as nucleoside diphosphate (NDP) kinase and (for animals) a major part of CK activity is also associated with the mitochondrial intermembrane space. Indeed, one can consider thermodynamic buffering as the major function for this compartment. Problems and paradigms ratios of magnesium-bound and unbound adenylates that govern various aspects of cell energetics. For instance, the MgATP/MgADP ratio (essential for the rates of energy metabolism), ATPfree/ADPfree ratio (important for the translocation of adenylates) and ADPfree/AMPfree ratio (a parameter in allosteric regulation) could all be derived from total adenylate contents.(38) The AK equilibrium generates the computational property of the adenylates-magnesium set. A similar computational set can be established for guanylates, cytidylates, and uridylates via corresponding buffering equilibria. Creatine kinase Adenylate kinase The AK activity generates simple linear relations between the equilibrated adenylate species,(7,36,37) so the system itself generates computation. This means that the adenylate metabolome is organized computationally due to the ‘‘stable non-equilibrium’’ principle maintained via thermodynamic buffering of opposite (e.g., catabolic and anabolic) nonequilibrium fluxes. The AK is therefore a ‘‘measuring device’’ in the biological system that makes it computable. Figure 1 shows how the mechanism of equilibration of fluxes via AK works. ATP in the cell provides the energy for anabolic reactions (load of cell biosynthesis) via two major mechanisms, one including hydrolysis to ADP (load 1) with corresponding phosphorylation of a substrate or Pi release, and another to AMP (load 2) with pyrophosphorylation or release of PPi. ATP is resynthesized via phosphorylation from ADP. These fluxes are controlled by AK in such a way that, upon reaching equilibrium, all concentrations of adenylates (free and magnesium-bound) plus free magnesium become computationally linked. This results in the establishment of CK equilibrium is coupled to AK equilibrium in certain biological systems, e.g., in muscle.(39) The large pool size of creatine and creatine phosphate in muscle represents an additional thermodynamic buffering system. Its high capacity is beneficial for muscle where the fluctuations of the extrinsic load conductance are bigger than, for instance, in liver. This results in specific conditions for adenylate equilibration essential for special needs, e.g., active muscle contraction, where concentrations of adenylates, Mg2þ and Ca2þ are strongly controlled. The link between the AK and CK equilibria results in high MgATP/MgADP ratios and also provides optimal Ca2þ and Mg2þ concentrations for muscle operation.(40) The resulting ATP/ADP ratios in the intermembrane space of animal mitochondria are up to 20 and thus mitochondria operate closer to state 4.(41) Disruption of AK and CK activities(42) results in severe reductions of the cellular capability to maintain total ATP turnover under muscle functional load. This indicates that AK and CK are major points in the coordinated network of complementary enzymatic pathways that serve in the maintenance of energetic homeostasis and physiological efficiency.(43) CK and AK operate as interrelated cellular high energy phosphoryl transfer systems from mitochondria to the cytosol(44): each creatine phosphate molecule produced undergoes about 50 unidirectional CK-catalyzed phosphotransfers in transit to an ATP consumption site in the intact muscles. CK can be considered as a functional analog of arginine kinase, the enzyme that fulfils thermodynamic buffering role in lower animals.(45) In these species, the arginine pool (but not creatine) is linked to adenylates. Essential components of the adenylate system in driving metabolism Figure 1. Scheme of the operation of the adenylate system in metabolism. Energy consuming metabolic reactions are fed by ATP hydrolysis to ADP (load 1) or AMP (load 2). AK buffer equilibrates three adenylate species (Mg2þ feedback is not shown). ATP is synthesized by phosphorylation, while AMP and ADP can also be produced by apyrase and other reactions. BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. The adenylate system in driving metabolism consists of a directing component (ATP and ATP synthase) and energysensing component (AMP and AMP kinase) so it has both energetic and informational constituents. The energy-sensing mechanism maintains the balance between ATP production 1093 Problems and paradigms and consumption in all eukaryotic cells. Without considering all factors (including free magnesium), the energy-sensing ratio (ATP/AMP) changes as a square of metabolic driving ratio (ATP/ADP).(26) In reality, the dependence is more complicated and based on the established concentration of free magnesium,(8) and the directing component is in fact the MgATP/MgADP ratio while the sensing component is the MgATP/AMPfree ratio. In a more detailed view, the adenylate system has also an activating or milieu component (free magnesium), i.e., essential to provide optimum condition for the activity of major anabolic and catabolic enzymes. Other ion concentrations (potassium, manganese) can be established or adjusted under AK equilibrium.(36) The system also includes a translocation component (ATPfree/ADPfree ratio) determining the exchange of adenylates across membranes. Thus, in approximation, the adenylate system can be represented as a matrix of the four components: driving, sensing, activating, and translocating. Adenylate species not included in these ratios can be counted in some specific cases, e.g., the MgATP-dependent phosphofructokinase is allosterically regulated by all free adenylates, with MgATP acting as an inhibitor of fructose 6-phosphate binding to the enzyme.(46,47) The knowledge of driving forces for metabolic fluxes is essential in metabolomic research. It is easy to get the data of total amounts of adenylates, i.e., ATP, ADP, and AMP. As they are not distributed evenly in the cell, it is more valuable to have the data on their subcellular amounts. From the total amounts of adenylates one can get information on whether they comply with AK equilibrium, i.e., whether the value of [ATP][AMP]/([ADP])2 is within the range of apparent AK constant.(7,8) Reliable data can be obtained for erythrocytes,(48) but also for chloroplast stroma and for intermembrane spaces of plant mitochondria and chloroplasts. The mitochondrial AK also effectively equilibrates adenylates in the cytosol of green leaves, but there is apparently no AK equilibrium in the cytosol of some heterotrophic tissues, e.g., roots.(7) Thus, provided that thermodynamic equilibria are established in the cell (which is detected via correspondence of the values of the substrate ratios to established constants of thermodynamic buffering enzymes), one can argue that major metabolic parameters are computationally interrelated and their balance is almost as firm as sequences of genes and proteins. Under AK equilibrium, ratios of free and Mg-bound adenylates and concentration of Mg2þ are established. This equilibrium is also linked to the ratios of guanylates, cytidylates, and uridylates and to concentrations of other metal ions. Specific equations for the determination of these parameters have to be established, but it is evident that, in the case of metabolomic research, we have a solid computational pattern that can be firmly analyzed. 1094 A. U. Igamberdiev and L. A. Kleczkowski Adenylate system governs organellar membrane potentials The ratios of adenylates are linked to membrane potential values by the set of formulae described in ref.(8) The measure of subcellular adenylate concentrations (AK-equilibrated) provides an effective tool for the estimation of in vivo mitochondrial and chloroplastic membrane potentials, which would be impossible or complicated by other methods. High respiratory rates generate membrane potential across the mitochondrial membrane. The membrane potential at established pH values inside and outside of mitochondria and chloroplasts is tightly linked to the ATP/ADP ratios inside and outside the organelles. The values of membrane potentials are tightly linked to total subcellular concentrations of adenylates, as well as internal [Mg2þ] values and ratios of free and/or Mg-bound adenylates.(8) The maintenance of electric potential is one of the main functions of membranes achieved by the balance of both non-equilibrium processes of proton pumping, ATP synthesis and ion transport, and equilibrium action of the AK enzyme. Other nucleotide systems governing metabolism Ratios of guanylates, uridylates, and cytidylates are responsible for driving metabolism of proteins, carbohydrates, and lipids, respectively.(3) They are directly linked to ATP availability since the essential phosphorylase/kinase reactions responsible for the production of the other triphosphate nucleotides in the cell transfer one phosphate from ATP to another nucleotide such as GDP, UDP, or CDP to create ADP and either GTP, UTP, or CTP, respectively, by NDP kinases.(49) Because the energies of the triphosphate bonds being created and destroyed in these reactions are essentially identical, these reactions are driven entirely by the concentration gradient between the reactants and products. Driving the ATP/ADP ratio to very low levels, also drives the ratios GTP/GDP, UTP/UDP, CTP/CDP, and nearly every other triphosphate/diphosphate pair in the cell, to even lower levels to maintain a favorable concentration gradient.(3) GTP is important for protein synthesis, UTP for carbohydrate synthesis and CTP for lipid synthesis, so these distinct groups of metabolites are ‘‘marked’’ by corresponding nucleotides. Buffering enzymes for the nonadenylic nucleotides are, correspondingly, guanylate, uridylate, and cytidylate kinases, all belonging to NMP kinase family.(50,51) Some of these enzymes have mixed substrate specificity, and it is likely that the nucleotide species are well equilibrated; AK itself is frequently nonspecific for nucleotides (e.g., corn AK).(37) On the other hand, the operation of guanylate, uridylate, and cytidylate systems depends on the adenylate system, because only ATP is synthesized by a direct mechanism linked to membrane potential. The enzymes that link ATP to other nucleotide systems are NDP kinases that catalyze BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. A. U. Igamberdiev and L. A. Kleczkowski Problems and paradigms active phosphorylation of GDP, UDP, CDP, and IDP.(49,52) These nucleoside triphosphates are not formed by the oxidative phosphorylation. Thermodynamic buffering of pyridine nucleotide system and regulated uncoupling Pyridine nucleotide system The pyridine nucleotide system is the most important, together with the adenylate system, in governing metabolism. It is composed of reduced and oxidized pyridine nucleotides: one (NAD) is used mainly in catabolic processes and its reduced form (NADH) supplies electrons for ATP synthesis, while another (NADP) used is mainly in anabolic processes. Its direct participation in ATP synthesis is limited to certain groups of organisms, e.g., in plants, where NADPH can serve as an electron donor for the mitochondrial electron transport chain (ETC) in the case of a high reduction level.(53) Some dehydrogenases use NADH or NADPH in a stereo a form, while others use the b form. Although it is not possible to distinguish the two forms as they are interconvertible, this different usage may make sense in metabolic channeling when nucleotide from the previous enzyme is supplied in corresponding orientation. Also, different dehydrogenases use NADH or NADPH species with different degrees of protonation, e.g., external mitochondrial NADH and NADPH dehydrogenases use protonated forms, which are related to low pH optimum of their operation. NAD (nicotinamide-ribose-ADP) and NADP (nicotinamideribose-phosphoADP) are both the derivatives of adenylates carrying a residue that can be in a reduced or oxidized form. We do not consider here covalently bound cofactors such as FAD, FMN, or pyridoxal phosphate that cannot be equilibrated freely by corresponding thermodynamic buffering systems. Equilibration of pyridine nucleotides is achieved via transhydrogenase (Fig. 2). The proton-pumping transhydrogenase equilibrates pyridine nucleotides together with the proton gradient and hence membrane potential.(54) Plant mitochondria lack a proton-pumping transhydrogenase.(55) The nonenergy-linked transhydrogenase reactions are generally important for NADH generation in conditions of high NADPH concentration, like in photosynthetic cells, while the energylinked transhydrogenase is important for NADPH formation using reducing power of NADH and proton motive force, which is important for animal cells.(56) Transhydrogenase equilibrium is an important regulatory principle in metabolic systems. It regulates the flux through the tricarboxylic acid cycle via regulation of NAD- and NADPdependent isocitrate dehydrogenases, one of which is irreversible (NAD) and another reversible (NADP).(57) A BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. Figure 2. Scheme of the operation of the pyridine nucleotide system in metabolism. Catabolism is fed by NADH while anabolism is driven by NADPH. Oxidized pyridine nucleotides are reduced by corresponding enzymes and anabolic and catabolic nucleotide pairs are equilibrated by transhydrogenase, TH, which is linked to proton pumping in bacteria and animals but not in plants. similar mechanism, but in the absence of proton-translocating transhydrogenase, may also function in plants, which equilibrate pyridine nucleotides via a different mechanism involving other transhydrogenase-like reactions.(55,58) When the concentration of NADH rises, it transhydrogenates with NADPþ forming NADPH. The consequence of this will be the activation of additional oxidation flow via the internal NADPH dehydrogenase of the ETC. Its capacity [up to 15% of the total capacity for NAD(P)H oxidation] can provide an additional power to increase flux through the ETC. The rise in NADPH also contributes to the activation of the so-called (cyanideinsensitive) alternative oxidase (AOX),(59) so the total flux through the ETC can increase even much more. It also stimulates the reverse reaction of NADP-dependent isocitrate dehydrogenase, leading to citrate efflux from mitochondria and to the activation of AOX gene.(60) Regulated uncoupling in the pyridine nucleotide system in mitochondria The irreversible exergonic reaction that is not coupled to ATP synthesis can be tightly regulated by shifts in balance between the reactions of load and consumption. It can be called ‘‘regulated uncoupling’’ and, in general, occurs if one of the two coupled reactions of a cyclic process proceeds without its counterpart(12); e.g., an enzyme passes a proton without ATP synthesis, e.g., the AOX in the mitochondrial ETC in plants.(61) Although this slippage decreases the efficiency of energy utilization, it enables controlling and regulating metabolic demands.(12) Another example is the activation of an uncoupling protein under certain conditions or regulated uncoupling caused by long-chain fatty acids.(62) This type of thermodynamic buffering equilibrates the ‘‘energy charge’’ of the ATP pool with the redox charge of pyridine nucleotides. It 1095 Problems and paradigms A. U. Igamberdiev and L. A. Kleczkowski allows uncoupling of a high redox charge from the imminent elevation of energy charge. As a result, metabolic processes may be less affected by high values of redox charge. Figure 3 shows the regulation of ATP synthesis (and hence of the value of energy charge after its equilibration through thermodynamic buffering) achieved via regulated uncoupling. This uncoupling can be achieved via different mechanisms, e.g., in plant mitochondria such a mechanism includes the AOX, rotenone-insensitive NAD(P)H dehydrogenases, and the uncoupling proteins. In animal mitochondria, although non-coupled dehydrogenases, and oxidases are absent, different types of uncoupling proteins, including those that may be involved in Ca2þ release,(63) regulate the level of coupling between NADH oxidation and ATP synthesis. Regulated uncoupling represents a link between redox charge and energy charge, and serves to establish the energy charge at a certain level prior to buffering. The form of metabolism is based on the fact that concentrations of the metabolites NADH, NADþ, and ADP are all under strict regulatory control. One purpose for maintaining the concentrations of these metabolites near the center of the thermodynamically feasible range could be to maximize flexibility.(3) The thermodynamic equilibration of pyridine nucleotides results in their mass action ratio (NADPH/NADPþ)/(NADH/ NADþ) of 10 or less in the absence of proton-pumping transhydrogenase, as in some bacteria depending on oxygen supply(56,64) and in plants.(58) For animal cells, this ratio has much higher values (from 100 to 500 depending on membrane potential).(65) Different species of pyridine nucleotides can participate in particular metabolic processes, e.g., external NADH and NADPH dehydrogenases in plant mitochondria use protonated forms of reduced pyridine nucleotides as substrates.(66) However, further details of thermodynamic equilibration of pyridine nucleotides and a possibility of local pH values as a feedback of this equilibrium need to be clarified.(67) The regulated uncoupling of NADH oxidation from ATP synthesis results in a buffered maintenance of the ATP/ADP and NADH/NADþ ratios under different physiological conditions. The ratio of ATP/ADP decreases with the decrease in oxygen supply, while NADH/NADþ increases. The balance between the catabolic and anabolic reduction charges is established at different levels depending on the state of the cell, e.g., in higher plants it depends on developmental stage (floral induction, flowering) and light/dark conditions.(68) Figure 3. Link between pyridine nucleotide and adenylate systems. The balance between NADH/NAD and ATP/ADP ratios is supported via switching to non-coupled pathways of NADH oxidation triggered at high reduction level. Figure 4. Generalized scheme of the peroxisomal metabolism, representing the mechanism of dissipation of high NAD(P)H/NAD(P)þ ratio, thus unlinking the transformation of the substance A to B from redox control. 1096 Peroxisomes in regulated uncoupling Whereas in mitochondria NADH oxidation can be unlinked with ATP synthesis, peroxisome may provide uncoupling at another level, i.e., when substrate oxidation is not coupled to NAD reduction. Therefore, peroxisome can be considered as a ‘‘regulated leak compartment’’.(69) Peroxisomes carry out reactions that may lower reduction charge, e.g., fatty acid oxidation in peroxisomes is not coupled to NADþ reduction as in mitochondria, and in higher plants glycolate oxidation occurs via an NAD-independent mechanism (contrary to that existing in algae containing glycolate dehydrogenase).(69) In this regard, peroxisomes can be considered as redox buffering organelles representing metabolic steps regulating NADH/NADþ ratio. Usually, peroxisomes contain metabolic cycles in which one (reductive) half is linked to the consumption of NADH and another (oxidative) dissipates energy in the reaction of hydrogen peroxide formation (Fig. 4). Since several oxidative metabolic processes can be linked or non-linked to NADH generation (such as b-oxidation of fatty acids in mitochondria versus in peroxisomes), this may provide a fine metabolic control of NAD reduction.(69) Plant photorespiration, a wasteful process, attributed to peroxisomes in several key points, is important for controlling the redox level in photosynthesizing cells.(70) Some metabolites linked to reduction by NADH and NADPH can be converted outside of the peroxisome,(71) while those that undergo flavin-dependent oxidation are exclusively located in the peroxisome. BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. A. U. Igamberdiev and L. A. Kleczkowski Pyrophosphate system PPi can be used as a macroergic molecule; however, the system of PPi-based energy transfer lacks the flexibility of the adenylate system, since it is not capable of thermodynamic buffering. This is because the PPi system cannot directly regulate the load and consumption of PPi through a buffered feedback compound such as the AMP in the adenylate system. PPi energy transfer may be important under conditions of ATP deficiency, e.g., during anoxia in plants or under uncoupling conditions when the ATP production is suppressed and low ATP level is not sufficient to fulfill energy demands of the cell. Lowering of ATP in the cell results in the increase in Mg2þ due to the shift in the AK equilibrium.(7) PPidependent enzymes such as PPi-dependent phosphofructokinase and pyruvate-Pi-dikinase thus become more active upon elevation of Mg2þ concentration, and in hypoxia-tolerant plants their expression is strongly increased.(72) The MgPPidependent phosphofructokinase needs higher [Mg2þ] than the MgATP-dependent phosphofructokinase because of the much lower affinity that Mg2þ has for complexation with PPi than with ATP (KMgPPi ¼ 1.2 mM
1 compared with KMgATP ¼ 73 mM
1).(73) We do not have information on the regulatory role of free PPi, so the system may lack a feedback in the regulation, such as that demonstrated for the adenylate system. However, PPi has a more stable pool at lower pH values characteristic for low oxygen conditions because the enzyme splitting it (inorganic pyrophosphatase) is less active at pH below 7 and is inhibited by Ca2þ accumulating under hypoxia.(74) This may make PPi an important autonomous salvage energy molecule when ATP production goes down. In plants, the vacuolar proton pumping can be also PPi dependent, and its role is increased upon elevation of Mg2þ concentration.(75) While the ratios of nucleotides such as ATP/ADP and UTP/ UDP change in a similar manner, PPi concentration changes independently, highlighting its importance as an autonomous energy donor.(76) Problems and paradigms which is equilibrated by thermodynamic buffering enzymes. There are several links between energy charge and RNA processing, including the supply of nucleoside monophosphates released during RNA cleavage for buffering with NTP. It is known, e.g., that NDPK cleaves DNA and RNA.(77) In addition, AMP kinase having key functions in regulation of energy metabolism also regulates transcription.(78) The level of ATP directly affects adenylation level of messenger RNA and hence its stability.(79) The generalized view on metabolism should, ideally, include the metabolic reproduction (via transcription and translation) of catalysts that govern metabolism. In fact, synthesizing and maintaining the catalysts is necessary for sustained operation of metabolism and this is a part of the metabolic network. This is based on the assumption that proteins themselves are products of metabolism, and thus metabolites.(80–82) The function of ATP, GTP, UTP, and CTP is two-fold: these nucleoside triphosphates are used both in general metabolism as free nucleotides and in nucleic acid metabolism via covalent polymerization to RNA molecules. Changes in their pools directly affect both energy metabolism and RNA turnover. Nucleotides serve as metabolites (free coenzymes) of enzymes and their association into nucleic acids generates matrices for the reproduction of enzymes themselves.(83) The main challenge here is the difficulty of formal joint representation of energetic and information fluxes in the frames of a whole system. It is important to mention that the operation of the two systems (the genetic and the metabolic) is based on simple computable rules that are generated within these systems. For the metabolic system, certain ratios between adenylates are established via thermodynamic buffering of enzymes such as AK, NDP kinases, and others. Deviations from such equilibrium will result in an unstable state that tends to return to the initial stable non-equilibrium state. Acknowledgments: This work was supported by the Natural Sciences and Engineering Research Council of Canada (to A. U. I.) and by the Swedish Research Council (to L. A. K.) Future challenges We have argued in this paper, that the stable non-equilibrium state is generated by steady fluxes established via thermodynamic buffering and regulated uncoupling. Under these conditions, fluxes of energy and information in the system operate in a constant steady-state way and are organized spatially and temporally. The thermodynamic buffer theory of Stucki(11,31) suggested a general explanation of steady operation of oxidative phosphorylation. This theory can be extended to include all major parameters related to cell metabolism. Both energy charge and genetic information use the components of the same pool (nucleoside triphosphates), BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. References 1. Caspi, R., Foerster, H., Fulcher, C. A., Hopkinson, R., Ingraham, J., et al. MetaCyc: a multiorganism database of metabolic pathways and enzymes. Nucleic Acids Res 2006. 34: D511–D516. 2. Atkinson, D. E., The energy charge of the adenylate pool as a regulatory parameter. Interaction with feedback modifiers. Biochemistry 1968. 7: 4030–4034. 3. Henry, C. S., Broadbelt, L. J. and Hatzimanikatis, V., Thermodynamicsbased metabolic flux analysis. Biophys J 2007. 92: 1792–1805. 4. Andersen, K. B. and Meyenburg, K. V., Charges of nicotinamide adenine nucleotides and adenylate energy charge as regulatory parameters of the metabolism in Escherichia coli. J Biol Chem 1977. 252: 4151–4156. 1097 Problems and paradigms 5. Russell, J. B. and Cook, G. M., Energetics of bacterial growth – balance of anabolic and catabolic reactions. Microbiol Rev 1995. 59: 48–62. 6. Fromm, H. J., Control by energy charge is an untenable theory. Trends Biochem Sci 1977. 2: N198–N200. 7. Igamberdiev, A. U. and Kleczkowski, L. A., Implications of adenylate kinase-governed equilibrium of adenylates on contents of free magnesium in plant cells and compartments. Biochem J 2001. 360: 225–231. 8. Igamberdiev, A. U. and Kleczkowski, L. A., Membrane potential, adenylate levels and Mg2R are interconnected via adenylate kinase equilibrium in plant cells. Biochim Biophys Acta 2003. 1607: 111–119. 9. Prigogine, I., Thermodynamics of irreversible processes. New York, Wiley, 1967. 10. Bauer, E. S., Theoretical biology. Moscow, VIEM, 1935. 11. Stucki, J. W., The thermodynamic-buffer enzymes. Eur J Biochem 1980. 109: 257–267. 12. Tomashek, J. J. and Brusilow, W. S., Stoichiometry of energy coupling by proton-translocating ATPases: a history of variability. J Bioenerg Biomembr 2000. 32: 493–500. 13. Reich, J. G. and Sel’kov, E. E., Time hierarchy, equilibrium and nonequilibrium in metabolic systems. Biosystems 1975. 7: 39–50. 14. Chance, B., Williamson, G., Lee, I. Y., Mela, L., DeVault, D., et al. Synchronization phenomena in oscillations of yeast cells and isolated mitochondria. New York, Academic Press, 1973. 15. Aon, M. A., Roussel, M. R., Cortassa, S., O’Rourke, B., Murray, D. B., et al. The scale-free dynamics of eukaryotic cells. PLoS One 2008. 3: e3624. 16. Lloyd, D., Oscillations, synchrony and deterministic chaos. Prog Bot 2009. 70: 69–91. 17. Lloyd, D., Aon, M. A. and Cortassa, S., Why homeodynamics, not homeostasis? Scientific World J 2001. 1: 133–145. 18. Lloyd, D. and Murray, D. B., Redox rhythmicity: clocks at the core of temporal coherence. BioEssays 2007. 29: 465–473. 19. Shnoll, S. E., Physico-chemical factors of the biological evolution. Moscow, Nauka, 1979. 20. Henry, R. P., Multiple roles of carbonic anhydrase in cellular transport and metabolism. Annu Rev Physiol 1996. 58: 523–538. 21. Khan, L. A. and Amin, M., A kinetic scheme for the early phase of ATP hydrolysis by actomyosin ATPase and its bioenergetic implications. Indian J Biochem Biophys 1995. 32: 37–43. 22. Alberty, R. A., Standard Gibbs free energy, enthalpy, and entropy changes as a function of pH and Mg for several reactions involving adenosine phosphates. J Biol Chem 1969. 244: 3290–3302. 23. Alberty, R. A., Thermodynamics of the hydrolysis of adenosine triphosphate as a function of temperature, pH, Mg, and ionic strength. J Phys Chem B 2003. 107: 12324–12330. 24. Shikama, K., Standard free energy maps for the hydrolysis of ATP as a function of pH, Mg and Ca. Arch Biochem Biophys 1971. 147: 311–317. 25. Tulub, A. A. and Stefanov, V. E., New horizons of adenosinetriphosphate energetics arising from interaction with magnesium cofactor. Eur Biophys J 2008. 37: 1309–1316. 26. Hardie, D. G. and Hawley, S. A., AMP-activated protein kinase: the energy charge hypothesis revisited. BioEssays 2001. 23: 1112–1119. 27. Rottenberg, H., Non-equilibrium thermodynamics of energy-conversion in bioenergetics. Biochim Biophys Acta 1979. 549: 225–253. 28. Lemasters, J. J. and Billica, W. H., Non-equilibrium thermodynamics of oxidative phosphorylation by inverted inner membrane vesicles of rat liver mitochondria. J Biol Chem 1981. 256: 2949–2957. 29. Kedem, O. and Caplan, S. R., Degree of coupling and its relation to efficiency of energy conversion. Trans Faraday Soc 1965. 61: 1897. 30. Kadenbach, B., Intrinsic and extrinsic uncoupling of oxidative phosphorylation. Biochim Biophys Acta 2003. 1604: 77–7794. 31. Stucki, J. W., The optimal efficiency and the economic degrees of coupling of oxidative phosphorylation. Eur J Biochem 1980. 109: 269–283. 32. Stucki, J. W., Lehmann, L. H. and Mani, P., Transient kinetics of thermodynamic buffering. Biophys Chem 1984. 19: 131–145. 33. Veuthey, A. L. and Stucki, J., The adenylate kinase reaction acts as a frequency filter towards fluctuations of ATP utilization in the cell. Biophys Chem 1987. 26: 19–28. 34. Demirel, Y. A. and Sandler, S. I., Nonequilibrium thermodynamics in engineering and science. J Phys Chem B 2004. 108: 31–43. 1098 A. U. Igamberdiev and L. A. Kleczkowski 35. Roberts, J. K. M., Aubert, S., Gout, E., Bligny, R. and Douce, R., Cooperation and competition between adenylate kinase, nucleoside diphosphokinase, electron transport, and ATP synthase in plant mitochondria studied by 31P-nuclear magnetic resonance. Plant Physiol 1997. 113: 191–199. 36. Blair, J. M., Magnesium, potassium, and the adenylate kinase equilibrium. Magnesium as a feedback signal from the adenine nucleotide pool. Eur J Biochem 1970. 13: 384–390. 37. Kleczkowski, L. A. and Randall, D. D., Maize leaf adenylate kinase: purification and partial characterization. Plant Physiol 1986. 81: 1110– 1114. 38. Igamberdiev, A. U. and Kleczkowski, L. A., Equilibration of adenylates in the mitochondrial intermembrane space maintains respiration and regulates cytosolic metabolism. J Exp Bot 2006. 57: 2133–2141. 39. Vincent, A. and Blair, J. M., The coupling of the adenylate kinase and creatine kinase equilibria. Calculation of substrate and feedback signal levels in muscle. FEBS Lett 1970. 7: 239–244. 40. Kemp, G. J., Manners, D. N., Clark, J. F., Bastin, M. E. and Radda, G. K., Theoretical modelling of some spatial and temporal aspects of the mitochondrion/creatine kinase/myofibril system in muscle. Mol Cell Biochem 1998. 184: 249–289. 41. Gellerich, F. N., The role of adenylate kinase in dynamic compartmentation of adenine nucleotides in the mitochondrial intermembrane space. FEBS Lett 1992. 297: 55–58. 42. Janssen, E., Terzic, A., Wieringa, B. and Dzeja, P. P., Impaired intracellular energetic communication in muscles from creatine kinase and adenylate kinase (M-CK/AK1) double knock-out mice. J Biol Chem 2003. 278: 30441–30449. 43. Dzeja, P. P., Zeleznikar, R. J. and Goldberg, N. D., Adenylate kinase: kinetic behavior in intact cells indicates it is integral to multiple cellular processes. Mol Cell Biochem 1998. 184: 169–182. 44. Dzeja, P. P., Zeleznikar, R. J. and Goldberg, N. D., Suppression of creatine kinase-catalyzed phosphotransfer results in increased phosphoryl transfer by adenylate kinase in intact skeletal muscle. J Biol Chem 1996. 271: 12847–128451. 45. Teague, W. E., Jr. and Dobson, G. P., Thermodynamics of the arginine kinase reaction. J Biol Chem 1999. 274: 22459–22463. 46. Kemp, R. G. and Krebs, E. G., Binding of metabolites by phosphofructokinase. Biochemistry 1967. 6: 423–434. 47. Li, Y. L., Rivera, D., Ru, W., Gunasekera, D. and Kemp, R. G., Identification of allosteric sites in rabbit phosphofructo-1-kinase. Biochemistry 1999. 38: 16407–16412. 48. Rose, I. A., The state of magnesium as estimated from the adenylate kinase equilibrium. Proc Natl Acad Sci U S A 1968. 61: 1079–1086. 49. Hasunuma, K., Yabe, N., Yoshida, Y., Ogura, Y. and Hamada, T., Putative functions of nucleoside diphosphate kinase in plants and fungi. J Bioenerg Biomembr 2003. 35: 57–65. 50. Fukami-Kobayashi, K., Nosaka, M., Nakazawa, A. and Go, M., Ancient divergence of long and short isoforms of adenylate kinase: molecular evolution of the nucleoside monophosphate kinase family. FEBS Lett 1996. 385: 214–220. 51. Alexandre, J. A. C., Roy, B., Topalis, D., Pochet, S., Périgaud, C. and Deville-Bonne, D., Enantioselectivity of human AMP, dTMP and UMPCMP kinases. Nucleic Acids Res 2007. 35: 4895–4904. 52. Ray, N. B. and Mathews, C. K., Nucleoside diphosphokinase – A functional link between intermediary metabolism and nucleic acid synthesis. Curr Top Cell Regul 1992. 33: 343–357. 53. Rasmusson, A. G., Geisler, D. A. and Møller, I. M., The multiplicity of dehydrogenases in the electron transport chain of plant mitochondria. Mitochondrion 2008. 8: 47–60. 54. Hatefi, Y. and Yamaguchi, M., Nicotinamide nucleotide transhydrogenase: a model for utilization of substrate binding energy for proton translocation. FASEB J 1996. 10: 444–452. 55. Bykova, N. V., Rasmusson, A. G., Igamberdiev, A. U., Gardeström, P. and Møller, I. M., Two separate transhydrogenase activities are present in plant mitochondria. Biochem Biophys Res Commun 1999. 265: 106– 111. 56. Voordouw, G., van der Vies, S. M. and Themmen, A. P. N., Why are two different types of pyridine nucleotide transhydrogenase found in living organisms? Eur J Biochem 1983. 131: 527–533. BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. A. U. Igamberdiev and L. A. Kleczkowski 57. Sazanov, L. A. and Jackson, J. B., Proton-translocating transhydrogenase and NAD-linked and NADP-linked isocitrate dehydrogenases operate in a substrate cycle which contributes to fine regulation of the tricarboxylic acid cycle activity in mitochondria. FEBS Lett 1994. 344: 109–116. 58. Igamberdiev, A. U. and Gardeström, P., Regulation of NAD- and NADPdependent isocitrate dehydrogenases by reduction levels of pyridine nucleotides in mitochondria and cytosol of pea leaves. Biochim Biophys Acta 2003. 1606: 117–125. 59. Gray, G. R., Villarimo, A. R., Whitehead, C. L. and McIntosh, L., Transgenic tobacco (Nicotiana tabacum L.) plants with increased expression levels of mitochondrial NADPR-dependent isocitrate dehydrogenase: evidence implicating this enzyme in the redox activation of the alternative oxidase. Plant Cell Physiol 2004. 45: 1413–1425. 60. Vanlerberghe, G. C. and McIntosh, L., Coordinate regulation of cytochrome and alternative pathway respiration in tobacco. Plant Physiol 1992. 100: 1846–1851. 61. Lambers, H., Cyanide-resistant respiration: a non-phosphorylating electron transport pathway acting as an energy overflow. Physiol Plant 1982. 55: 478–485. 62. Schönfeld, P. and Wojtczak, L., Fatty acids decrease mitochondrial generation of reactive oxygen species at the reverse electron transport but increase it at the forward transport. Biochim Biophys Acta 2007. 1767: 1032–1040. 63. Trenker, M., Malli, R., Fertschai, I., Levak-Frank, S. and Graier, W. F., Uncoupling-proteins 2 and 3 are elementary for mitochondrial Ca2R uniport. Nat Cell Biol 2007. 9: 445–452. 64. Matin, A. and Gottschal, J. C., Influence of dilution rate on NAD(P) and NAD(P)H concentrations and ratios in a Pseudomonas sp. grown in continuous culture. J Gen Microbiol 1976. 94: 333–341. 65. Hoek, J. B. and Rydström, J., Physiological roles of nicotinamide nucleotide transhydrogenase. Biochem J 1988. 254: 1–10. 66. Edman, K., Ericson, I. and Møller, I. M., The regulation of exogenous NAD(P)H oxidation in spinach (Spinacia oleracea) leaf mitochondria by pH and cations. Biochem J 1985. 232: 471–477. 67. Noctor, G., De Paepe, R. and Foyer, C. H., Mitochondrial redox biology and homeostasis in plants. Trends Plant Sci 2007. 12: 125–134. 68. Bonzon, M., Simon, P., Greppin, H. and Wagner, E., Pyridine nucleotides and redox-charge evolution during the induction of flowering in spinach leaves. Planta 1983. 159: 254–260. 69. Igamberdiev, A. U. and Lea, P. J., The role of peroxisomes in the integration of metabolism and evolution of land plants. Phytochemistry 2002. 60: 651–674. BioEssays 31:1091–1099, ß 2009 Wiley Periodicals, Inc. View publication stats Problems and paradigms 70. Igamberdiev, A. U., Bykova, N. V., Lea, P. J. and Gardeström, P., The role of photorespiration in redox and energy balance of photosynthetic plant cells: a study with a barley mutant deficient in glycine decarboxylase. Physiol Plant 2001. 111: 427–438. 71. Timm, S., Nunes-Nesi, A., Pärnik, T., Morgenthal, K., Wienkoop, S., et al. A cytosolic pathway for the conversion of hydroxypyruvate to glycerate during photorespiration in Arabidopsis. Plant Cell 2008. 20: 2848–2859. 72. Huang, S., Colmer, T. D. and Millar, A. H., Does anoxia tolerance involve altering the energy currency towards PPi? Trends Plant Sci 2008. 13: 221– 227. 73. O’Sullivan, W. J. and Smithers, G. W., Stability constants for biologically important metal-ligand complexes. Meth Enzymol 1979. 63: 294–336. 74. Hemalatha, K. P. and Prasad, D. S., Purification, physicochemical properties, and subcellular location of alkaline inorganic pyrophosphatase from sesame (Sesamum indicum L.) cotyledons. Biochem Cell Biol 2002. 80: 215–224. 75. Maeshima, M. and Yoshida, S., Purification and properties of vacuolar membrane proton-translocating inorganic pyrophosphatase from mung bean. J Biol Chem 1989. 264: 20068–20073. 76. Dancer, J., Veith, R., Feil, R., Komor, E. and Stitt, M., Independent changes of inorganic pyrophosphate and the ATP/ADP or UTP/UDP ratios in plant cell suspension cultures. Plant Sci 1990. 66: 59–63. 77. Hammargren, J., Salinas, T., Maréchal-Drouard, L. and Knorpp, C., The pea mitochondrial nucleoside diphosphate kinase cleaves DNA and RNA. FEBS Lett 2007. 581: 3507–3511. 78. McGee, S. L. and Hargreaves, M., AMPK and transcriptional regulation. Front Biosci 2008. 13: 3022–3033. 79. Eads, B. D. and Hand, S. C., Mitochondrial mRNA stability and polyadenylation during anoxia-induced quiescence in the brine shrimp Artemia franciscana. J Exp Biol 2003. 206: 3681–3692. 80. Montero, F., Nuno, J. C., Melendez-Hevia, E., Olasagasti, F., Vazquez, S. and Moran, F., Stoichiometric analysis of self-maintaining metabolisms. J Theor Biol 2008. 252: 427–432. 81. Rosen, R., Life itself. New York, Columbia University Press, 1991. 82. Cornish-Bowden, A., Cárdenas, M. L., Letelier, J. C. and SotoAndrade, J., Beyond reductionism: metabolic circularity as a guiding vision for a real biology of systems. Proteomics 2007. 7: 839–845. 83. Igamberdiev, A. U., Foundations of metabolic organization: coherence as a basis of computational properties in metabolic networks. BioSystems 1999. 50: 1–16. 1099