Download The Role of ATP in Metabolism and more Lecture notes Biological Sciences in PDF only on Docsity!
The Role of ATP in Metabolism
R W HANSON
Department of Biological Sciences Plymouth Polytechnic Drake Circus, Plymouth PL4 8AA, UK
Introduction The thesis presented in most textbooks of biology, cell biology and (to a lesser extent) biochemistry regarding the metabolic role of ATP is that the nucleoside triphosphate acts as the 'energy store' or as the 'energy currency' of the cell. The following passages from the book by Darnell, Lodish and Baltimore I are typical: Cells extract energy from foods through a series of reactions that have negative free energy changes. Much of the free energy released is not allowed to dissipate as heat, but is captured in chemical bonds formed by other molecules for use throughout the cell. In almost all organisms the most important molecule for capturing and transferring free energy is adenosine triphosphate, or A TP. The useful free energy in an A TP molecule is contained in high- energy phosphoanhydride bonds (p 42). The authors justify the designation of a phosphoanhydride bond as a high energy bond as follows (p 43). Although this bond is an ordinary bond it is referred to as a high energy bond because it releases about 7. kcal/mol of free energy when it is broken, as in hydrolysis. When summarising the involvement of ATP in energeti- cally unfavourable reactions (p 49) it is stated that, All of these reactions are fueled by the hydrolysis of A TP. Several aspects of this type of explanation of the role of ATP are, however, questionable. Most fundamentally, Banks 2 has claimed that classical thermodynamics, which applies only to closed systems, cannot be used to analyse metabolic reactions which occur in open systems and are under kinetic control. In contrast, Atkinson, 3 and Crab- tree and Taylor 4 have argued convincingly that thermo- dynamics can be used with advantage in discussions of metabolism. Each component reaction in a metabolic pathway must move towards equilibrium in the direction of the pathway, that is, the change in Gibbs function for every reaction in a pathway must be negative; endergonic reactions do not occur in metabolism, or indeed else- where, despite claims to the contrary. 5 Furthermore, each reaction must provide products at concentrations which allow the enzyme catalysing the next step in the pathway to maintain the pathway flux. Atkinson, 3 and Crabtree and Taylor 4 have given excellent accounts of how these two criteria can be assessed using the principles of equilibrium thermodynamics. It should be recognised, however, that data calculated assuming that a reaction is at equilibrium relate to a limiting situation because the reaction will, in fact, be in a steady state somewhat displaced from equilibrium. If it is accepted that thermodynamics is relevant to
discussions of the role of ATP in metabolism the use of free energy changes remains a source of much confusion to graduates and undergraduates alike. Free energy is not a conventional type of energy, like thermal or electrical energy, which is conserved according to the first law of thermodynamics. It is a function of state which was introduced by Gibbs as an indirect measure of the net production of entropy in a closed system and its sur- roundings. 6 In real, spontaneous processes the net entropy always increases and the Gibbs function (free energy) always decreases: neither is conserved. Further- more the change in Gibbs function which accompanies a reaction is determined by the difference between the initial and final states of the entire system. It cannot be ascribed to a single component in the system and still less to a single bond in any component. These points have been discussed at length by several authors 2-4'7-1°. It is surprising, therefore, that many other authors continue to treat free energy as a caloric substance and that the concept of the 'high-energy bond' continues to be propagated. This last particularly in view of the fact that experimental evidence has been available for some time which shows that in at least two enzyme-catalysed reactions cleavage of the terminal phospho-anhydride bond in ATP does not "liberate significant amounts of free energy". Bagshaw and Trentham 11 have made a detailed study of the hydrolysis of ATP catalysed by the myosin ATPase of muscle fibre. Their results show that the step in which enzyme-bound ATP and water are converted to enzyme-bound ADP plus Pi has a Gibbs function change of only - 5. 5 kJ mol -~ whereas non- enzymic hydrolysis in free solution (Eqn 1) has AG O' = -30.5 kJ mo1-1, Keq' = 1.3 x 105.
ATP + H20 ~:~ ADP + Pi (1)
Furthermore, Boyer and co-workers 12 have shown that ATP, ADP and Pi are essentially in equilibrium (AG' = 0 kJ mo1-1) when bound to the active site of the mitochondrial ATP synthetase complex. In both cases, the steps with large changes in Gibbs function are those involving substrate binding and product release. Perhaps when these facts become more widely known the high- energy bond will eventually be laid to rest! Finally, considerable confusion must arise from the failure of authors to distinguish between the thermo- dynamic and kinetic stability of ATP towards hydrolysis, and the frequent mis-use of the word hydrolysis. The moderately large change in Gibbs function accompanying the hydrolysis of ATP (Eqn 1) gives no information whatsoever about the kinetic stability of ATP in the presence of water. The nucleoside triphosphate is, in fact, stable in water for up to 6 months if stored at 4°C; 13 it is certainly not easily hydrolysed as claimed by Alberts et al. 14 Furthermore although ATP participates in many coupled reactions in which it is converted to ADP and Pi (eg Eqns 11 and 12), they do not involve hydrolysis, that is, bond cleavage as a result of reaction with water. 15
Indeed, the reaction catalysed by the myosin ATPase may
be the only significant metabolic reaction which involves
hydrolysis of the nucleotide. Here, the role of hydrolysis is not to "generate free energy" but is to displace strongly bound ATP from the protein so that another cycle of filament contraction can begin. The areas of confusion noted above make the metabolic role of ATP a very difficult topic to present to under- graduate students, particularly to those, such as biologists, who may have only a weak background in the physical sciences. Fortunately a coherent presentation, which avoids the pitfalls associated with interpretations of thermodynamic functions of state, can be given in terms of equilibrium constants and mass action quotients as described in this article. The concept of an equilibrium constant is simpler than that of 'free energy' and, unlike the latter, should be familiar to any student who has followed an introductory course in chemistry.
Energy, the Capacity for Change
As noted in the introduction, the role of ATP in metabolism is usually discussed in terms of energy transduction. Energy, however, is a difficult concept to grasp. It is defined as the capacity to do work, but since work itself is often-defined in terms of changes in energy the definition is somewhat circular. Progress can be made when it is realised that the performance of work always involves a change in the condition of the system and/or surroundings in which it occurs; energy can then be defined as the capacity to produce, or undergo, change. This definition is particularly apposite to biological systems, which are essentially networks of integrated chemical reactions, because energy transduction in such systems can then be discussed in terms of the capacity for chemical change. The capacity or potential for change in a chemical system is usually assessed in terms of the associated change in Gibbs function (AG). The latter is related to the equilibrium constant for the reaction (K'q) by Eqn 2 in which R is the gas constant, T the absolute temperature and Q' is the mass action quotient ([C] ¢ [D]d/[A]a [B] b)
AG' = - R T In (geq/Q') (2)
for a generalised reaction aA + bB x=~ cC + dD: note that the terms refer to initial concentrations, not to equilibrium values. Primes indicate that each concen- tration term has been divided by the appropriate standard state concentration: [H20] = 55.5 mol dm -3, [H ÷] = 1 x 10-7 mol dm -3, all others = 1 mol dm -3. It follows that any deduction regarding the potential for, and direction of, reaction made on the basis of AG' can be made equally
well on the basis of the value of the ratio K~q/Q'. If the
ratio is greater than unity, then the reaction will move towards equilibrium in the forward direction; the concen- trations of C and D increase at the expense of those of A and B until, at equilibrium, Q' = K~q and the larger the value of the ratio, the greater the potential for change
towards the equilibrium state. Conversely, if g'q/Q' is less than unity, the reaction will proceed in the reverse direction. Again, the value of K'q/Q' is a measure of the potential for change; in this case the smaller the ratio, the greater the potential. It is important to be familiar with a number of other properties of equilibrium constants 9'17'1s before using them to explain the metabolic role of ATP. These are: (1) The equilibrium constant of a reaction is equal to geq/Q' when Q' is unity; it is thus a measure of the potential for the reaction to move towards equilibrium when all of the reactants are present in their standard states. (2) Because, by convention, product concentrations always appear in the numerator of the expression for an equilibrium constant and starting material concentrations in the denominator, K'q for the reverse component of a reaction is the reciprocal of the equilibrium constant for the forward component.
(3) The ratio K~q/O', like AG', gives no information
whatsoever about the rate at which a chemical change
occurs.
(4) The potential for change, as measured by K'q/Q', is
entirely independent of the route by which the change occurs. This follows from the fact that the Gibbs function, G, is a function of state. (5) The equilibrium composition of a reaction mixture is unaffected by a second reaction, occurring in the same
phase, provided that there is no interaction between the two
sets of reactants. The situation changes radically, however,
if chemical coupling exists between the two reactions. Consider, for example, two reactions (Eqns 3 and 4) coupled by a common intermediate, D, the net effect of the coupled reactions being described by Eqn 5.
A + B x ~ C + D (3)
D + E ~ F + G (4)
A + B + E x ~ C + F + G (^) (5)
At equilibrium the concentration of D must be such that it satisfies the equilibrium expressions for both reactions 3 and 4 (Eqn 6). It then follows, via Eqn 7, that K~q for the overall reaction is equal to the product of the constants of the component reactions (Eqn 8).
[DI = K~q3 [A][B] [F][G] [C] = K~q4[E] (6)
K~q3 x K~q4 = [C][FI[G] [AI[B][E]
Keq3 × Keq4 = K~q5 (8)
Since AG' values are additive, this deduction also follows from the logarithmic relationship between K'q and AG'.
the equilibrium state) is independent of the route by which the reaction occurs. Enzymes have evolved which catalyse coupled reactions of the following type. In the first reaction of a coupled pair, ATP phosphorylates one starting material of the condensation, producing a product which, unlike water, can be recognised unambiguously by the enzyme catalysing the second reaction of the coupled pair; the material phosphorylated is usually a carboxylic acid or a reducing carbohydrate. In the second reaction of the coupled pair the condensation product is generated by the remaining starting material of the condensation displacing phosphate from the common, phosphorylated, intermediate. The coupled reactions of this type which are used to produce acetyl coenzyme A in certain bacterial cells are described by equations 11, (AK = acetate kinase), and 12 (PTA = phosphotransacetylase) and their net effect by Eqn 13.
CH3CO2H + ATP A__~g2+K/M CH3CO2P + ADP (11) acetyl phosphate
PTA CH3CO2P + H S C o A x=~ C H 3 C O S C o A + Pi (12)
C H 3 C O 2 H + ATP + HSCoA ~ C H 3 C O S C o A
The latter is identical with Eqn 10 and in keeping with this the equilibrium constants for Eqns 11 and 12 are 3.6 x 10 -3, and 74, respectively19'2° giving Keq'13 = 0.27*; as noted previously the maximum concentration of acetyl coenzyme A which can be produced by reaction (10/13) is of the order of 10-5 mol dm -3. Thus the high potential for an aqueous solution of ATP to be converted to an
aqueous solution of ADP and Pi is realised, not by direct
hydrolysis, but by a route which concomitantly produces a
condensation product. A plausible mechanism for coupled reactions of the type just described might be as shown in Eqns 14 to 17:
RIOH xo--R10-: + H + (14)
R10-: + M g A D P - P O ] - ~ R1OPO 2- + MgADP-
R ~ ~ R z ~_ R I ~ R z + PO34- I H ~:~ R1XR 2 + HPOa 2- (16)
HPO 2- + H + ~ H2PO 4- (17)
From this it can be seen that ATP is acting as a dehydrating agent as would be expected of an acid
*The small difference between this figure and that given previously for reaction 10 reflects the difficulties associated with the determination of very large, or very small, equilibrium constants.
anhydride. The inclusion of the nucleoside triphosphate as a reactant in a condensation allows the elements of water to be removed from the reactants, RIOH and HXR 2, and liberated, not as water, but as protons or as part of phosphate. The chemical potential of both of the latter species will be lower in the system than that of water and on this basis the potential for formation of RIXR 2, will clearly be increased in comparison with the direct con- densation. Some variation in the general theme described above is seen in as much as the common intermediate produced by ATP in coupled reactions is sometimes an adenylic acid derivative, as in the activation of amino-acids during protein biosynthesis, or a pyrophosphate derivative as in the biosynthesis of polysaccharides. In all cases, however, the immediate role of ATP in biosynthesis can be summarised as follows: the participation of ATP in biosynthesis allows cells to produce physiologically acceptable concentrations of condensation products. The latter are produced by coupled reactions, involving a common phosphorylated, pyrophosphorylated, or adenyl- ated intermediate, such that the high potential for ATP to be converted to ADP + Pi or to AMP + PPi, etc, is realised without the nucleotide being hydrolysed.
The Role of ATP in Solute Transport
One of the most thoroughly investigated solute transport systems involving ATP is the calcium pump of the sarcoplasmic reticulum. 16 In a resting muscle cell the pump maintains the concentration of calcium ions in the sarcoplasmic reticulum at a level of 1 x 10-3 mol dm - and in the cytoplasm at a level of 1 x 10 -6 mol dm -3. Unfortunately, quantitative analysis of the system in terms of Keqand Q' for the ion transfer (Eqn 18)
2Cacyto ~^ 2 + 2Ca~2+ (18)
is complicated by the existence of an electrical potential gradient across the membrane. 4 It is clear, however, that in the absence of the pump, the potential for change would be such that calcium would re-enter the cytoplasm (as it does when gated calcium channels are opened on receipt of a nerve impulse). The presence of the pump allows ion transfer to occur against the opposing chemical and electrical potential gradients by coupling it to the conversion of ATP to ADP and Pi. The pump, which spans the membrane, is thought to bind two calcium ions at a high affinity site (E in Eqns 19 to 24) on its cytoplasmic face (Eqn 19). ATP is then bound at a separate site and the 13-carboxyl group of an aspartic acid residue is phosphorylated (Eqn 20). It is suggested that the introduction of the phosphoryl group allows additional non-covalent bonds to be formed between it and the protein such that the conformation of the latter is changed; the calcium binding site is converted to a low affinity site (E* in Eqns 21 to 24) which is now accessible to the sarcoplasmic reticulum (Eqn 21). The calcium ion is released from this site (Eqn 22) and, finally, the pump is
returned to its original conformation by hydrolysis of the 13-aspartylphosphate group (Eqns 23 and 24).
E + 2Cacyto ~=~^ 2+ E.Ca22+ (19)
E.Ca 2+ + ATP x=~ E - P. C a 2+ + ADP (20)
E - P - C a 2+ x=~ E*-P.Ca22+ (21)
E*-P.Ca22+ x=~ E * - P + 2Ca~+ (22)
E * - P + H 2 0 Xr~ E* + Pi (^) (23)
E* x=~ E (24)
2Cacyto + ATP + H 2 0^ 2 + x=~ 2Ca2~++ ADP + Pi (25)
The overall function of the pump is described by Eqn 25, which is the sum of Eqns 18 and 1. As in the case of the synthesis of acetyl coenzyme A, the inclusion of ATP as a reactant in the process will increase the potential for change in the required direction by a factor of 6.5 × 108. Thus, the role of the nucleotide is exactly the same as in biosynthesis - - the high potential for it to be converted to ADP and Pi is realised, not by direct hydrolysis, but in such a way that a process which would otherwise have a low potential for change is facilitated.
ATP and Muscle Contraction Although the details remain elusive, Huxley's 'sliding filament' model is widely accepted as providing the best explanation of the mechanism of muscle contraction. 16' Muscle fibres contain interdigitating arrays of thin actin and thick myosin filaments. 'Cross-bridges' are present between the filaments where protrusions (headpieces) on the myosin filament make contact with binding sites on the actin filament. The essence of Huxley's model is that the cross-bridges go through an oar-like motion, sliding the thick and thin filaments past each other, thus shortening the muscle fibre. After an individual cross- bridge has undergone a power stroke, the myosin head detaches from the actin filament, returns to its original conformation (the back-stroke of the oar) and then attaches to an adjacent binding site on the actin filament (Fig. 1). The sequence of events is then repeated. In the absence of ATP ADP and Pi, the potential for the overall change depicted in Fig 1 and Eqn 26 to occur would be very small.
A1.M + A2 xr:~A2.M + A 1 (26)
Steps 1 and 3 involve mechanical work, while step 2 involves breaking strong protein-protein interactions. Only step 4, which involves bond formation, would be expected to have a favourable equilibrium constant. As in the cases of the biosynthesis of condensation products and of solute transport, evolution has overcome this problem by coupling the low potential process to the conversion of
M
1 2 1 2
r
A L ~p
M
1 ~ ' 2 A t )
M
1 2 A ' ~)~ADP'Pi
4 M
Figure 1 The cross-bridge model of muscle contraction
ATP to ADP and Pi. The steps in the process may be as shown in Fig 1 and Eqns 27 to 30.
A1.M.ADP.Pi ~ AI.M * + ADP + Pi (27)
AI.M * + ATP ~ M*.ATP + A 1 (28)
M*.ATP + H 2 0 ~=~ M.ADP.Pi (29)
M.ADP.Pi + A 2 ~ A2.M.ADP.Pi (30)
A1.M.ADP.Pi + A 2 + ATP + H20 ~:~ A2.M.ADP-Pi + A' + ADP + Pi (31)
In the equations, M represents a myosin head in a conformation such that it makes an angle of approxi- mately 90 ° to the actin filament to which it binds, while M* represents a myosin head in a 45° conformation with respect to the actin filament. As noted previously, Bagshaw and Trentham's 11 analysis of the hydrolysis of ATP catalysed by myosin showed that the steps with large equilibrium constants are those involving substrate bind- ing and product release. In keeping with this, it is suggested that the force generating step in muscle contraction is associated with product (specifically ADP) release from an actomyosin ADP.Pi complex (Step 1, Fig. 1; Eqn 27). As a result of product release, the conformation of the myosin head changes from the 90° state to the 45 ° state and the actin filament moves with respect to the myosin filament. In order for further movement to occur, the strongly-bound myosin must be displaced from the actomyosin complex. This is achieved by ATP binding when one set of strong bonds is replaced by another (Step 2, Fig 1; Eqn 28). Strongly bound ATP must then, in turn, be displaced from myosin; this is accomplished by hydrolysis (Step 3, Fig 1; Eqn 29). The free energy change associated with hydro-
lysis ( - 5. 5 kJ mol-1), corresponding to a value of K~q/Q'
of approximately 10, indicates that there is a small, but significant, potential for the change to occur. It is not
realised in such a way that low potential processes such as biosynthesis, active transport and muscle contraction are facilitated.
References l Darnell, J, Lodish, H and Baltimore, D (1986) 'Molecular Cell Biology', Scientific American Books, W H Freeman, New York, 42- /Banks, B E C (1969) Chemistry in Britain 5,514- 3Atkinson, D E (1977) 'Cellular Energy Metabolism and its Regu- lation', Academic Press, New York 4Crabtree, B and Taylor, D J (1979) in 'Biochemical Thermodynamics', Edited by Jones, M N, Elsevier, Amsterdam, 333- 5Watson, J D, Hopkins, N H, Roberts, J W, Steitz, J A and Weiner, A M (1987) 'Molecular Biology of The Gene', Vol 1, Fourth Edition, Benjamin-Cummings, Menlo Park, p 167 6Allen, A (1983) Trends in Biochem Sci 8, 81- 7Banks, B E C and Vernon, C A (1978) Trends in Biochem Sci 3, N156- 8Novick, S (1976) J Biol Educ 10, 116- 9Morris, J C (1974) 'A Biologist's Physical Chemistry', Second Edition, Edward Arnold, London l°Chappeli, J B (1977) 'ATP', Carolina Biology Reader No 50, Carolina Biological Supply Co, Burlington, NC, USA llBagshaw, C R and Trentham, D R (1983) Biochem J 133,323- ~2Reported by Hatefi, Y (1985) Ann Rev Biochem 54, 1075- ~3Bioehemica Information II (1975), Boehringer Mannheim GmbH, 21 a4Alberts, B, Bray, D, Lewis, J, Raft, M, Roberts, K and Watson, J D (1983) 'The Molecular Biology of the Cell', Garland, New York, p 67 15Hawsley, M G G (1981) in 'The Condensed Chemical Dictionary', Tenth Edition, Edited by Sobers, H A, Van Nostrand Reinhold, New York, p 546 a6Eisenberg, E and Hill, T L (1985) Science 227, 999- 17Harris, W F (1982) J Chem Educ 59, 1034- 18McPartland, A A and Segal, I H (1986) Biochem Educ 14, 137- ~9jencks, W P (1968) in 'Handbook of Biochemistry', Second Edition, CRC Press, Cleveland, OH, J181-J 2°Stadman, E R (1955) Meth Enzymol Vol 1,596- 21Hibberd, M G and Trentham, D R (1986) Ann Rev Biochem 15, 119- 22Lehninger, A L (1970) 'Biochemistry', Worth, New York, p 298 23Harold, F M (1986) 'The Vital Force: A Study of Bio-energetics', W H Freeman, New York
Does Free-energy Change of Individual
Alone Determine the Flux through a
Pathway?
ABRAHAM K ABRAHAM
Department of Biochemistry Faculty of Medicine University of Kuwait, Kuwait
Reactions
Metabolic
Introduction
One of the most important facts that needs to be known about a metabolic pathway is the flux through the pathway. This can be determined from the free energy change, AG, of its individual reactions. Most textbooks of biochemistry emphasise this point. Thus for any given metabolic pathway, the flux through the pathway in the forward direction is possible only if the AG values of all individual reactions are negative, except in the case of coupled reactions. In other words, it is not sufficient that the overall value of free-energy changes is negative. Yet, when this point is described in more detail later in the book by providing the AG values of glycolytic reactions, ~' the data presented seem to confuse the students.
Free Energy
Several text b o o k s 1-3 provide the standard free-energy change, AG °, and the actual free-energy change, AG, calculated from mass action ratios for individual reactions of the glycolytic pathway. One of the reasons for presenting these details is probably to illustrate that while the standard free-energy change, AG ° , of some reactions have a positive value, the actual free-energy change, AG, of all reactions are negative, thus re-emphasising the basic concept taught earlier under bioenergetics. However, the data presented show that the actual free-energy change, of several reactions of this pathway are positive. 1"2 This increase in the free-energy depicted for several reactions has been pointed out, and explained as deriving from errors in experimental measurements, z Another text book provides no explanation for the positive AG values.1 Two out of the three reaction steps with positive AG values do not confuse the majority of students. 1 These reactions are: (1) conversion of di- hydroxyacetone phosphate to glyceraldehyde 3-phosphate by triosephosphate isomerase, and (2) the conversion of 1,3-glycerate bisphosphate to 3-phosphoglycerate by phosphoglycerate kinase. In the first case, being a side reaction, it is easy to understand that this reaction does not prevent the flux of glucose along the glycolytic pathway. In the second case, if a student is able to recall that the reactions catalysed by glyceraldehyde 3-phos- phate dehydrogenase and phosphoglycerate kinase are coupled, they should be able to work out the overall free- energy change which is in fact negative. For the reaction catalysed by phosphoglyceromutase ie conversion of 3-phosphoglycerate to 2-phosphoglycerate, also shown with a positive AG value, 1 it is difficult for the students to understand how glycolysis can proceed beyond
