Cellular Energy Utilization and Molecular Origin of Standard … · 2017-10-17 · PHYSIOLOGICAL REVIEWS Vol. 77, No. 3, July 1997 Printed in U.S.A. Cellular Energy Utilization and

PHYSIOLOGICAL REVIEWS

Vol. 77, No. 3, July 1997

Printed in U.S.A.

Cellular Energy Utilization and Molecular Origin of Standard Metabolic Rate in Mammals

DAVID F. S. ROLFE AND GUY C. BROWN

Department of Biochemistry, University of Cambridge, Cambridge, United Kingdom

I. Introduction II. Quantitative Assessment of the Origin of Standard Metabolic Rate in Mammals

A. B.

C.

D. E. F. G. H.

Contribution of different tissues to standard metabolic rate Contribution of non-ATP-consuming reactions in the major oxygen-consuming tissues to standard

metabolic rate Contribution of ATP-consuming reactions in the major oxygen-consuming tissues to standard

metabolic rate Service functions Relative ATP use by cellular ATPases Reactions that uncouple standard metabolism Substrate utilization and glycolytic ATP production in standard metabolic rate Summary

III. Contribution of Feeding, Cold, and Muscle Use to Field and Maximal Metabolic Rate A. Cold-induced thermogenesis B. Feeding-induced thermogenesis C. Muscle use-induced thermogenesis

IV. Heat Production and Free Energy Dissipation A. Thermogenesis and enthalpy dissipation B. Free energy dissipation and efficiency C. Distribution of energy dissipation

V. Changes and Differences in Metabolic Rate A. Control of cellular respiration B. Causes of species differences in standard metabolic rate related to body size and phylogeny C. Summary

VI. Functions of Standard Metabolic Rate A. Maintenance of cellular composition against undesigned uncoupling reactions B. Standard metabolic rate as a means of heat production C. High standard metabolic rate enables high maximal and field metabolic rate D. Standard metabolic rate increases the potential for regulation E. Standard metabolic rate as a means for regulating harmful free radical production F. Summary

VII. Conclusions

732 733 734

735

737 741 742 742 743 743 744 744 745 745 745 745 747 748 749 749 750 751 751 751 752 752 753 753 753 753

Rolfe, David F. S., and Guy C. Brown. Cellular Energy Utilization and Molecular Origin of Standard Metabolic Rate in Mammals. Physiol. Rev. 77: 731- 758,1997. - The molecular origin of standard metabolic rate and thermogenesis in mammals is examined. It is pointed out that there are important differences and distinctions between the cellular reactions that 1) couple to oxygen consumption, 2) uncouple metabolism, 3) hydrolyze ATP, 4) control metabolic rate, 5) regulate metabolic rate, 6) produce heat, and ?‘) dissipate free energy. The quantitative contribution of different cellular reactions to these processes is assessed in mammals. We estimate that -90% of mammalian oxygen consumption in the standard state is mitochondrial, of which -20% is uncoupled by the mitochondrial proton leak and 80% is coupled to ATP synthesis. The consequences of the significant contribution of proton leak to standard metabolic rate for tissue P-to-O ratio, heat production, and free energy dissipation by oxidative phosphorylation and the estimated contribution of ATP-consuming processes to tissue oxygen consumption rate are discussed. Of the 80% of oxygen consumption coupled to ATP synthesis, -25-30% is used by protein synthesis, 19-28% by the Na’-K’-ATPase, 4-8% by the Ca2+-ATPase, Z-8% by the actinomyosin ATPase, 7- 10% by gluconeogenesis, and 3% by ureagenesis, with mRNA synthesis and substrate cycling also making significant contributions. The main cellular reactions that uncouple standard energy metabolism are the Na+, K’, H+, and Ca2’ channels and leaks of cell membranes and protein breakdown. Cellular metabolic rate is controlled by a number of processes

0031-9333/97 $15.00 Copyright 0 1997 the American Physiological Society 731

732 DAVID F. S. ROLFE AND GUY C. BROWN Volume 77

including metabolic demand and substrate supply. The differences in standard metabolic rate between animals of different body mass and phylogeny appear to be due to proportionate changes in the whole of energy metabolism. Heat is produced by some reactions and taken up by others but is mainly produced by the reactions of mitochondrial respiration, oxidative phosphorylation, and proton leak on the inner mitochondrial membrane. Free energy is dissipated by all cellular reactions, but the major contributions are by the ATP-utilizing reactions and the uncoupling reactions. The functions and evolutionary significance of standard metabolic rate are discussed.

I. INTRODUCTION imal to SMR are seen; for example, a trained athlete’s maximal metabolic rate may be 20 x SMR, whereas in an

Mammalian cells and organisms use a considerable amount of energy in the basal or standard state, when no net work is done and all the free energy is dissipated. This review examines where this energy goes, what controls the energy usage, and what possible functions this basal energy usage might serve.

Standard metabolic rate (SMR) is the steady-state rate of heat production by a whole organism under a set of “standard” conditions; in mammals, these conditions are that the individual is an adult and is awake but resting, stress free, not digesting food (prior food intake being at or around maintenance level), and maintained at a temperature that elicits no thermoregulatory effect on heat production (see Ref. 15 for review). Standard metabolic rate is measured either directly as heat production, or indi- rectly as oxygen consumption from which it can be accurately predicted (189). Basal metabolic rate is similar to SMR, although the term is now most usually applied to human metabolism only. Resting metabolic rate is also similar to SMR, except that the metabolic rate is measured while the animal is still digesting food.

Metabolic rate changes under different conditions, and (at least) three other conditions are generally distin- guished: minimal metabolic rate, field metabolic rate, and maximal metabolic rate. Minimal metabolic rate is the metabolic rate in various conditions that cause the rate to fall below that of SMR; causes of this depression include sleep, anesthesia, and continued starvation. Meta- bolic rate falls by -10% below SMR during sleep and may fall by 40% (a decrease of 20% in metabolic rate per kilogram body weight) during long-term starvation in humans (15). Field metabolic rate is the average rate over an extended period of time when the animal is living in its natural environment and may be assessed either by direct measurements of heat production (as is often the case in human studies) or by the doubly labeled water method. Field metabolic rate is higher than SMR as a result of the energy requirements made by feeding, cold exposure, and muscle use (Fig. 1). Each of these conditions can increase metabolic rate above the standard rate, and the extent depends on conditions. The maximal meta-

untrained human a value of 12 x SMR would be more usual (15).

The energy use in the standard state is sometimes regarded as a fixed requirement for maintenance of the organism, to which other energy uses may be added. How- ever, the extra energy use induced by cold, feeding, muscle use, growth, or reproduction is not necessarily addi- tive with SMR. This is because the metabolic processes underlying SMR may decrease or increase in the conditions of extra energy demand (see Ref. 216). For example, during exercise, the metabolic rate of some organs (e.g., kidney and gut) decreases, whereas gluconeogenesis increases in the liver, and protein synthesis may decrease in skeletal muscle. Thus the processes underlying SMR

cannot be regarded as fixed, but rather change with conditions.

The standard and field metabolic rates are important because they determine the calorific nutrient requirements and growth rates of animals and humans. Not all the nutrient energy intake of an individual organism is lost as heat, since a proportion is lost as feces and urine (-4 and 5%, respectively, in humans) (114), a proportion is saved as growth of the individual or passed on during birth or lactation, and a small proportion is saved as work on the environment (Fig. 1; see Refs. 15 and 114 for energy requirements). However, the majority of an individual’s calorific intake is lost as heat, and thus accounted for by metabolic rate. This heat production is due to the cellular metabolism of all the tissues of an individual organism. In an adult organism that is not growing, reproducing, lactating, or doing work on the environment, all the energy taken in and metabolized is lost as heat by cellular metabolic reactions. All the free energy is dissipated, and all the chemical potential energy is converted to heat.

In this article we address the following questions with regard to mammalian SMR. 1) What is the quantitative

contribution of the major organs and tissues to SMR? 2) What is the quantitative contribution of the major metabolic pathways and cellular processes to SMR? 3) What are the main ATP-using reactions in mammalian cells, and

what is their contribution to total ATP consumption in a

bolic rate is the maximum steady-state metabolic rate at- tainable during hard exercise over a defined period of time and is often considered to be simply 10 times SMR in mammals. However, large variations in the ratio of max-

at which reactions is the free energy dissipated? 5) Which cell? -4) At which cellular reactions is the heat lost, and

reactions control SMR, and which are responsible for physiological changes in SMR? 6) What is the function of

July 1997 ORIGIN OF STANDARD METABOLIC RATE 733

Total Energy Intake

Total

energy

used

w--B---

Urine & Feces

\ \

\ \

\ \

“4

Total

energy loss

-----I

-Growth -----

- Repro- duction

\ \ \ \ \ \ \ \

‘4

Standard metabolic

rate

---e-B--

Muscle

use

--------

Food

--------

Cold

Total Heat

output

FIG. 1. Schematic illustration of how energy taken into human body is used. First column represents total energy intake, second column total energy used, and third column total heat produced. Proportions given are for adult humans and are only roughly to scale. Note that field metabolic rate represents daily (24 h) energy expenditure and does not include energy required for reproduction. Contribution to total heat output arising from cold-induced thermogenesis will obviously depend on ambient temperature at which metabolic rate measurements are made (see text for explanation). [Data from Blaxter (15).]

SMR? We first assess the contribution of a tissue or process to SMR by quantifying its coupling to oxygen consumption. Standard metabolic rate cannot be equated with oxygen consumption, but oxygen consumption accurately predicts SMR, and the extent to which a process is coupled to oxygen consumption quantifies the energy flux through the process.

II. QUANTITATIVE ASSESSMENT OF THE

ORIGIN OF STANDARD METABOLIC RATE IN MAMMALS

Metabolism has often been conceptually divided into the reactions that produce energy or ATP (catabolism) and the reactions that use energy or ATP (anabolism). This has been a useful dichotomy, but is misleading in several ways. 1) The processes of ATP production are not energy producing (as according to the first and second laws of thermodynamics energy is conserved and free energy dissipated by all reactions), but rather function to distribute free energy. 2) Free energy is distributed to energy-requiring reactions using intermediates other than the ATP, for example, GTP, NADH, NADPH, the mitochondrial proton gradient, and the plasma membrane Na’ gradient. 3) The ATP-using reactions are wrongly assumed to be responsible for all the heat production and free energy dissipation of metabolism and to be solely responsible for control of metabolic rate.

Because ATP utilization has been equated with cellular energy utilization, the origin of SMR has been thought to be solvable simply by identifying the major reactions responsible for ATP utilization. However, for the purposes of locating the origin of SMR, it will be more useful to divide metabolism into coupling and uncoupling reactions. Because no net work is done by SMR, the fact that

oxygen consumption occurs at all in the standard state is related to processes that “uncouple” metabolism (Fig. a>. The cell consists of a number of low-entropy structures (for example, macromolecules and ion gradients) whose production requires free energy. In animals, this free energy comes from the oxidation of food molecules: carbo- hydrates, fats, and proteins. This oxidation is coupled to the reduction of NAD. Oxidation of NADH by the mitochondrial electron transport chain is coupled to the production of an electrochemical proton gradient across the mitochondrial inner membrane. Synthesis of ATP is coupled to the “downhill” channeling of these protons through the ATP synthase. Processes such as protein syn-

thesis, maintenance of sodium and potassium gradients, and muscular contraction are coupled to ATP hydrolysis (see Fig. 2). However, if this was all that there was to metabolism, then the whole system would come to equilibrium and there would be no metabolic rate. There is a

significant flux through the system because of reactions that uncouple metabolism, such as protein degradation, ion leaks, and muscle relaxation. Thus, in a sense, the uncoupling processes account for SMR.

As mentioned above, energy metabolism has previously been thought to be fully coupled to ATP production, but it is now recognized that 1) not all oxygen consumption is mitochondrial and therefore coupled to oxidative phosphorylation, and 2) a significant proportion of mitochondrial respiration is not coupled to ATP synthesis.

Hence, some of the reactions responsible for SMR occur before oxidative ATP synthesis. Thus energy metabolism can be regarded as a leaky hose pipe (Fig. Z), and in the standard metabolic state, all the energy must leak out somewhere, since none of the energy is conserved, rather

all is lost as heat. However, the site of heat production cannot simply be identified with either the uncoupling or


Proteins I

* Amino acids

Na+out

I v

Na+in x

Actinomyosin (contracted)

I Actinzmyosin

(relaxed)

0

NADH,

x x NAD

Substrate H’out

I I I I

V H+LIl

x out I

V X in

e2+H2O ADP

+ Pi H;O

FIG. 2. Simplified scheme of cellular energy metabolism in standard state. Curved unbroken arrows represent some of coupling reactions of energy metabolism. Broken arrows indicate some of uncoupling reactions. Depicted coupling reactions are dehydrogenation, mitochondrial proton pumping, and ATP synthesis and ATP consumption by protein synthesis and Na’-K’-ATPase plus Na’-linked transport of many different substances (X in this figure) across plasma membrane. Note that transport of X may be via Na+ antiport as shown (e.g., if X = Ca2’), or by Na+ symport (e.g., if X = amino acids). Uncoupling reactions are mitochondrial proton leak, protein degradation, and Na+ and X leaks at plasma membrane. Note that uncoupling of substrate dehydrogenation may also occur, for example, due to peroxisomal fatty acid oxidation.

coupling reactions, because both types of reaction produce heat. Also, heat production cannot be equated with free energy dissipation, because these are different processes that must be evaluated separately. In addition, we cannot equate the uncoupling reactions with the reactions that control metabolic rate, since control is also located

standard metabolism have been measured by various methods. In vitro measurements are made on perfused organs or tissue slices; in vivo estimates are made from measure- merits of arteriovenous differences in oxygen concentration and of blood flow. However, even in studies where tissue oxygen consumption rates were measured in vivo, the total oxygen consumption of the organs does not always add up to the measured body oxygen consumption, and this presumably reflects technical difficulties in making the measurements (1, 85, 183). An estimate of the contribution of various organs to the oxygen consumption of an adult human and rat in the standard metabolic state is given in Table 1. The relative contribution of these organs to body mass and SMR varies somewhat with species (183). No one organ is responsible for the majority of the minimal metabolic rate, but some organs (brain, kidney, heart, and gastrointestinal tract) contribute a much larger fraction of the total SMR than their fractional mass or volume of the body, whereas others (bone, white adipose tissue, skin, and muscle) contribute much less. Note, however, that some tissues (e.g., skeletal muscle) make a significant contribution to SMR by virtue of their size, even though their metabolic rate per gram is relatively low. Thus the pathways contributing to the metabolic rate of these tissues are often

in coupling reactions. Different tissue preparations from different animals,

in different metabolic conditions, have been used to estimate the relative contribution of processes to tissue SMR. Thus, in using these data to estimate the overall contribution of a process to SMR, care must be taken in extrapolat- ing to the in vivo metabolic rate. Furthermore, it is essen- tial to distinguish between the metabolic rate of isolated tissues or cells in a minimal medium and the rate under standard physiological conditions. The former rate may be lower than the latter, because, for example, in a minimal medium isolated brain and heart will be missing electrical excitation, kidney tissue will lack metabolites to transport, and liver will be missing substrates for glucose, urea, and protein synthesis. However, if care is taken, oxygen consumption rates measured on tissue slices in vitro are not dramatically different from those of the same tissues in vivo. For example, organ metabolic rates estimated from tissue slice data (e.g., 37 pmol O2 l min-’ l whole of more relevance to the study of SMR than those in more liver-’ and 0.606 ,umol O2 l min-’ l g skeletal muscle-‘; Ref. metabolically active tissues such as kidney and brain. 71) are similar to in vivo measurements of tissue meta- Further geographical localization of oxygen con- bolic rate (e.g., 61 pmol O2 l rnin l whole liver-‘, Ref. 116; sumption and thus heat production has been attempted and 0.787 pmol O2 l min-’ l g skeletal muscle-‘, Ref. 173). in some tissues by measuring oxygen consumption or Z-

A. Contribution of Different Tissues to Standard deoxyglucose uptake, or mitochondrial or cytochrome ox-

Metabolic Rate idase density (which are thought to be proportional to metabolic rate). The metabolic rate in brain as estimated

The oxygen consumption rates and thus heat produc- by Z-deoxyglucose uptake is 25 - 50% lower in white matter

tion by the various organs of the mammalian body during than in gray matter, and cytochrome oxidase, mitochon-


TABLE 1. Contribution of the major oxygen-consuming organs of the body to body mass and standard metabolic rate

Organ

Body Mass, %

Human Rat

Body Oxygen Use, %

Human Rat

Liver 2 5 17 20 Gastrointestinal tract 2 5 10 5 Kidney 0.5 0.9 6 7 Lung 0.9 0.6 4 1 Heart 0.4 0.5 11 3 Brain 2 1.5 20 3 Skeletal muscle 42 42 20 30 Total 49.8 55.5 88 69

Organ masses and oxygen consumption rates are mean values from data on humans (1, 3, 72, 116, 135, 183) and rats (10, 71, 116, 117, 173, 181, 199). All data are for adult individuals except for data of Field et al. (71), which are for 150-g animals. Data for oxygen consumption rates are taken from in vivo tissue respiration measurements except some data for rat skeletal muscle which are from tissue slices (71) and perfused hindlimb preparations (10, 181, 199). Data for skeletal muscle tissue slice respiration are corrected for any observed decline in slice respiration rate with time (71).

dria, and the Na’-Kf-ATPase colocalize to dendrites and presynaptic terminals (see Refs. 68, 193, 220). For example, in primate visual cortex, 62% of mitochondria are found within dendrites, 21% in presynaptic terminals, with relatively little in axons (12%) or glia (2%) (220). In skeletal muscle, oxygen consumption and mitochondrial content are higher in red muscle than white muscle and decrease in the order type I > type IIa > type IIb muscle fibers. In kidney, the density of mitochondrial volume (mitochondrial volume per unit volume of cytoplasm) is similar to that of mitochondrial membrane density and basolateral membrane surface density and decreases from 33 to 22% along the proximal tubule. Along the distal portion of the nephron, it decreases from 44% in the medullary thick ascending limb to 31% in the distal convoluted tubule. In the thin limbs, the density is only 6-8%, whereas the collecting tubule has a mitochondrial density of -20 and 10% in the cortical and medullary segments, respectively (161). The liver is fairly homogeneous: 80% of liver cells have the same or similar mitochondrial content, but the density of inner mitochondrial membrane is slightly lower in pericentral regions of the lobules (140).

B. Contribution of Non-ATP-Consuming Reactions in the Major Oxygen-Consuming Tissues to Standard Metabolic Rate

1. Nonmitochondrial oxygen consumption

Not all oxygen consumption by tissues is mitochondrial or coupled to proton pumping and, therefore, involved in ATP production. A large number of oxidases function throughout metabolism, e.g., in peroxisomal fatty

acid oxidation. Estimates of nonmitochondrial oxygen consumption can be made by adding specific inhibitors of the mitochondrial respiratory chain and measuring the residual oxygen consumption of the tissue. Nonmitochon- drial respiration accounted for -2 1% in isolated resting rat liver cells (20, 41), 20% in perfused liver (187), 14% in perfused resting rat skeletal muscle (63, 176), and 20% in rabbit reticulocytes (192), but only 2% in isolated thymocytes (134) and 3% of the estimated in vivo oxygen consumption of rat heart (48). If we assume that nonmitochondrial oxygen consumption is 20, 14, and 3% of in vivo oxygen consumption of liver, skeletal muscle, and heart, respectively, and multiply these fractions by the respec- tive contributions of each organ to the total oxygen consumption of rat and human (listed in Table l), then we can estimate that nonmitochondrial oxygen consumption within these organs alone contributes 8.3 and 6.5% of total oxygen consumption in rats and humans, respectively. Because these organs contribute -50% of total oxygen consumption, and other organs are likely to have a smaller but still significant proportion of nonmitochondrial oxygen consumption, then the total contribution of nonmitochondrial oxygen consumption to whole body oxygen consumption is likely to be -10%.

A small proportion of mitochondrial oxygen consumption is not coupled to proton pumping, due to leakage of electrons from the electron transport chain to re- duce oxygen to superoxide and hydrogen peroxide. This proportion has been estimated to be maximally l-2% of the oxygen consumption rate of isolated mitochondria by measuring the rate of hydrogen peroxide production (17, 205).

2. lWi tochondrial pro ton leak

As stated in the introduction to section II, not all mitochondrial oxygen consumption in the standard state is coupled to ATP synthesis. In brown adipose tissue, a specific protein (thermogenin) enhances the proton conductance of the mitochondrial inner membrane, uncoupling mitochondrial oxygen consumption from ATP synthesis (155). Brown adipose tissue (BAT) contributes to SMR in neonatal mammals and small mammals adapted to cold environments (105, 155). Foster and Frydman (74) found that over 60% of the excess heat produced in response to cold in rats is generated in BAT. Brown adipose tissue thermogenesis is also stimulated by overfeeding (105, 155). However, BAT does not make a significant contribution to SMR in adult rats and humans (73).

Thermogenin is restricted to BAT mitochondria; however, mitochondria isolated from other organs have a significant passive leak for protons (proton leak), which is apparently not mediated by a specific protein (reviewed in Refs. 20,35). Proton leak is not an artifact of mitochondrial isolation, since it has been demonstrated in mito-


chondria within isolated hepatocytes (22, 41, 97, 156) and thymocy-tes (44).

With the assumption that mitochondrial oxygen consumption and proton pumping are obligatorily coupled under physiological conditions (see section I&!?), oxygen consumption in the absence of phosphorylation may be used as an indirect measure of proton leak activity (38, 154). The contribution of proton leak to the resting oxygen consumption of a particular tissue may be thus estimated by completely inhibiting mitochondrial ATP production. This inhibition will result in a drop in the tissue oxygen consumption rate and an increase in the mitochondrial membrane potential. Because of the strong dependence of the mitochondrial proton leak on membrane potential, the mitochondrial potential must then be returned to its resting value by sequentially inhibiting the mitochondrial respiratory chain. Once the mitochondrial potential has been returned to its resting value, the remaining oxygen consumption is a measure of the magnitude of the proton leak flux (plus any nonmitochondrial oxygen consumption) that had been occurring under resting conditions. The nonmitochondrial oxygen consumption rate is determined by completely inhibiting mitochondrial oxygen consumption, and hence, the proton leak flux can be calculated.

Experiments (of the type described above) to estimate the proton leak rate have been carried out using isolated rat liver cells and perfused rat skeletal muscle and indicate that proton leak accounts for 26% of oxygen consumption in resting rat liver cells (22, 41, 97, 156; reviewed in Ref. 20) and 52% in resting rat skeletal muscle (20, 176). In the heart, Challoner (48) found that 84% of the estimated in vivo oxygen consumption was inhibited by arresting the heart; of the remaining 16%, 94% was insensitive to oligomycin, but only 20% was insensitive to cyanide (an inhibitor of the respiratory chain). Thus a maximum of lo- 13% of the in vivo oxygen consumption may be coupled to the proton leak in heart. This will probably overestimate the actual contribution of proton leak (as explained above), but the correction of this overestimate in liver and skeletal was not greater than 50%, so we may estimate the contribution of proton leak in heart to be -5-10% of SMR. Note that the experiments outlined in References 41 and 176 may give an upper limit

to proton leak activity in the intact animal, since the skeletal muscle and liver may have a greater ATP demand in the standard state in vivo than in vitro, for example, due to muscular contraction and additional liver work functions. However, the oxygen consumption rate of isolated hepatocytes may be scaled to allow comparison with the in vivo rates in whole liver (14). The oxygen consumption rate of whole liver in vivo is 2.84 pmol O2 l min-’ l g liver-’ (116). The oxygen consumption rates of resting hepato-

58%) (41, 97, 156, 165, 169). Similarly, the in vitro oxygen consumption rates of perfused muscle preparations are within the same range as estimates of the oxygen consumption rate of skeletal muscle in vivo. Of course, this argument does not preclude overestimates in proton leak measurements in vitro, since stimulations of ATP consumption would also cause decreased proton leak flux and therefore may not significantly alter the overall oxygen consumption rate while altering the balance of ATP consumption and proton leak. However, it is one indica- tion that perfused muscle and liver cells may be reasonable models for the whole tissue in vivo, and thus the experiments outlined above show that proton leak may be an important contributor to SMR, at least in rats.

With a knowledge of the contribution to rat SMR of skeletal muscle (estimates range from 13 to 42%; mean 30%) and liver (lo-20%; mean 15%), the contribution of proton leak in liver and skeletal muscle to rat SMR can be estimated to be, on average, 20% (30 X 0.52 + 15 x

0.26) [range 10 (13 X 0.52 + 10 x 0.26) to 27 (42 x 0.52 + 20 x 0.26)%]. If the same proportions of liver and skeletal muscle oxygen consumption were uncoupled in humans, then the proton leak in these tissues would make a smaller contribution to SMR (- 15%) due to the smaller contribution of skeletal muscle in humans. The contribution of proton leak in other tissues, apart from heart above, has not been estimated, although proton leak is present in mitochondria isolated from brain and kidney and has similar activity to that found in isolated liver mitochondria (178). It might be argued that, because these organs have a higher demand for ATP relative to that in skeletal muscle in the standard state, there may be a relatively smaller contribution of proton leak in these tissues, since the proton leak and ATP synthesis compete for the same intermediate. Thus a reasonable estimate of the overall contribution of proton leak to SMR is -20%.

3. P-to-O Ratio and H’ stoichiometries

The P-to-O ratio (P/O) is the ratio of ATP synthesized to oxygen consumed (expressed in moles of 0 rather than O,), but we may distinguish the mechanistic ratio from the effective ratio. The mechanistic P/O is the maximal P/O in the absence of proton leak, other uncoupling reactions, and oxygen consumption other than at cytochrome oxidase and is equal to the product of the Hf/O and ATP/ Hf ratios. The effective P/O of a tissue is the ratio of ATP synthesized (by mitochondria) to oxygen consumed by the whole tissue. The effective P/O is related to the mechanistic P/O of the mitochondrial ATP synthesizing reactions by multiplying the mechanistic ratio by the fraction of the total tissue oxygen consumption rate that is used to drive mitochondrial ATP synthesis. The effective P/O differs from the mechanistic P/O mainly because of nonmito-

cytes isolated from fed rats incubated at 37°C in Krebs- chondrial oxygen consumption and mitochondrial proton Henseleit bicarbonate buffer is between 33 and 79% (mean leak.


The mechanistic P/O was previously thought to be 1 per site of the mitochondrial respiratory chain, i.e., 3 for mitochondria respiring on NADH and 2 for mitochondria respiring on succinate. These estimates were based on measurements of the effective P/O in mitochondria but overestimated the effect of the proton leak on these measurements. More recent measurements quantitatively accounting for the proton leak estimate the mechanistic P/O to be 2.5 for mitochondria respiring on NADH and 1.5 respiring on succinate (19, 106). This fits well with the present consensus estimates of the Hf/O ratios of the respiratory chain (10 for NADH oxidation and 6 for succinate oxidation) and the H+/ATP ratio of ATP synthesis and transport (3 H+ for ATP synthesis and 1 Hf equivalent for transport of ADP and Pi; see Refs. 16, 106). With the use of these figures, the mechanistic P/O has been estimated (16) to be 2.42 when respiring on pyruvate (with 1 FADH and 4 NADH oxidized and 1 GTP synthesized per pyruvate), 2.58 with glucose as substrate (with an extra 2 NADH and 2 ATP produced per glucose by glycolysis), and 2.26 with palmitate as substrate (with 2 ATP equiva- lents required for activation to palmitoyl-CoA, 7 NADH, and 7 FADH produced by P-oxidation, 8 x 3 NADH, 8 FADH, and 8 GTP from the tricarboxylic acid cycle).

There has been some doubt as to whether the mitochondrial oxygen consumption and proton pumping have a fixed coupling ratio (i.e., the H+/O ratio might vary or “slip”) (reviewed in Refs. 35, 149). If the mitochondrial proton pumps did slip, this would result in large changes in thermogenesis, free energy dissipation, and efficiency of the mitochondrial proton pumps themselves. However, more recent evidence indicates that, under approximately physiological conditions, the ratio is fixed (20, 21, 32, 92, 125, 221).

We can estimate the effective P/O in vivo by combin- ing estimates of the mechanistic P/O (which we will assume to be -2.5) with estimates of the fraction of tissue oxygen consumption 1) used by nonmitochondrial oxygen consumption (O-20% depending on tissue) and 2) coupled to proton leak (O-50% depending on tissue). If the com- bined nonmitochondrial respiration rate of heart, skeletal muscle, and liver accounts for -10% of SMR (see sect. IIBI) and the proton leak in these organs accounts for -20% of SMR (see sect. 11B2), then this allows us to estimate the effective whole body P/O as 1.8 (0.7 X 2.5). If we exclude the contribution of nonmitochondrial oxygen consumption, then we can estimate the effective mitochondrial P/O as 2.0 (0.8 X 2.5).

Considerable interest has been shown in the effective P/O of intact tissues (e.g., Refs. 29, 46, 76, 127). The effective P/O of yeast (29, 46), perfused kidney (76), and heart (127) estimated from 31P-nuclear magnetic resonance

(NMR) saturation transfer measurements of Pi+ATP flux were close to 3, 2.45, and 2.36, respectively. These data imply that virtually all the oxygen consumption in these

systems is used to drive ATP synthesis. Thus it would appear that the data presented here for the contribution of proton leak to resting liver cell and skeletal muscle oxygen consumption contradict the current perceptions of the tightness of coupling between oxygen consumption and ATP synthesis in cells and organs. If the contribution of oxidative phosphorylation to resting oxygen consumption is 53% in liver cells (41,97, 156) and 34% in muscle (20, 176), then the effective P/O in these systems (assuming a mechanistic P/O of 2.5) would be 1.3 (0.53 x 2.5) for liver and 0.9 (0.34 X 2.5) for muscle. However, it has been pointed out (29, 27, 127) that 31P-NMR saturation transfer measurements of Pi+ATP flux may significantly overestimate the net Pi+ATP flux catalyzed by the mitochondrial ATP synthase because 1) the unidirectional forward rate of the ATP synthase may be much greater than the net rate because the backward rate is significant, and 2) a proportion of the measured Pi+ATP flux is catalyzed by glyceraldehyde-3-phosphate dehydrogenase and phosphoglycerate kinase in yeast (28) and perfused heart (127). Inhibition of the glyceraldehyde-3-phosphate dehydrogenase and phosphoglycerate kinase exchange reaction in beating heart lead to a drop in the estimated P/O from 6.4 to 2.2 (127). Overexpression of yeast adenine nucleotide translocase, the kinetic properties of which are thought to influence the backward (ATP+PJ rate of the mitochondrial ATP synthase, caused the apparent mitochondrial P/O to increase from 0.9 to 4.2 (26). Thus it seems likely that previous estimates of the effective P/O in resting cells and tissues by “lP-NMR saturation transfer were erron- eously high.

C. Contribution of ATP-Consuming Reactions in the Major Oxygen-Consuming Tissues to Standard Metabolic Rate

We now move on to consider the contribution of the different ATP-consuming processes within the cell to SMR. Estimates of the relative ATP use by a process are usually made either 1) by measuring the rate of the process and estimating the oxygen consumption required to account for that rate, or 2) by specifically inhibiting the process and measuring the resultant fractional change in oxygen consumption. Estimates based on the first method need to assume a value of the P/O that has in the past been overestimated (see sect. II.@. In the absence of significant glycolytic ATP production (see sect. IIG), 1) the ATP synthesized by the tissue = (the oxygen consumption of the tissue X the effective P/O of the tissue), 2) the fraction of tissue ATP production used by the particular ATP user = (the ATP use by a particular ATP user/the oxygen consumption of the tissue X the effective P/O), and 3) the fraction of tissue oxygen consumption coupled to the particular ATP user = (the ATP


use by a particular ATP user/the oxygen consumption of the tissue X the mechanistic P/O).

Estimates of the contribution of the different ATP- consuming processes to metabolic rate based on the use of inhibitors can be unreliable because 1) the inhibitors are rarely specific, 2) inhibition of functions such as ion transport may affect energy metabolism at multiple sites, and 3) inhibition of a major ATP user may raise the ATP/ ADP and thus stimulate other ATP consumers. The extent to which the latter is a problem depends on the relative sensitivity of the ATP-consuming process and ATP production to the ATP/ADP (see Refs. 34,43). If the sensitivity of the ATP consumers is insignificant relative to mitochondrial ATP production, as has been shown in conca- navalin A-stimulated thymocytes (43), then complete and specific inhibition of an ATP consumer will not lead to a significant drop in ATP/ADP and will not underestimate the contribution of that ATP-consuming process to tissue oxygen consumption.

1. Na’-K”-ATPase

The Na’-K+-ATPase is an important ATP consumer, although its importance may have been overestimated previously, due to the secondary effects on cellular oxygen consumption of ouabain, an inhibitor of the Naf-K’- ATPase widely used in studies of the ion pump (reviewed in Refs. 53, 124). The contribution of this pump has been estimated in a large range of tissues either by measuring ion fluxes or by determining the effect of ouabain on oxygen consumption. The estimated coupling to oxygen consumption of the Na’-K’-ATPase is - 14% in adult rats (calculated from data in Ref. 53 and data for rat organ metabolic rates shown in Table 1) and -20% in humans (calculated from data in Ref. 53 and data for human organ metabolic rates shown in Table 1). The contribution is particularly large in brain (4, 68) and kidney (98, 197), where the pump has been estimated to be coupled to 50- 60% of oxygen consumption.

2. Ca”+-ATPase

As for the Na+-K+-ATPase, the contribution of the calcium ATPase to the metabolic rate of the major oxygen- consuming tissues of many mammals has been exten- sively reviewed by Clausen et al. (53). The contribution of the Ca’+-ATPase is usually assessed by measurement of transmembrane ion fluxes and multiplying these fluxes by the ATP:Ca2+ stoichiometry of the ATPase. The contribution of the ATPase to total tissue ATP consumption is then calculated by assuming a P/O of 3 for the tissue (53). If the data of Clausen et al. (53) are reevaluated using a mechanistic P/O of 2.5, we can calculate that the Ca2’- ATPase accounts for the following fractions of tissue metabolic rate: 6% in resting skeletal muscle, 24-58% in contracting skeletal muscle, <l% for resting heart muscle,

B-36% for contracting heart muscle, and 2% in resting liver. A significant contribution of this ATPase in brain was also proposed (53), although a specific figure was not suggested. Using the estimates for resting skeletal muscle and liver and contracting heart, and the organ oxygen consumption rates of Table 1, we can estimate that calcium cycling in heart, liver, and muscle contribute 4-6% of SMR in humans and between 3 and 4% of SMR in rats. This figure may rise dramatically if there is any skeletal muscle contraction.

3. Muscle contraction

The actinomyosin ATPase probably accounts for a significant proportion of the ATP turnover even in the standard metabolic state, since heart work, lung work, vascular, gastrointestinal, and other smooth muscle tone, and skeletal muscle tone are still required in this state. There is no good estimate of the contribution of smooth muscle at present. The maintenance of muscle tone via the actinomyosin ATPase is often considered to be an important contributor to SMR. However, Webster (213) summarized results indicating that the energy costs of standing and of changing position were only -1% of the total energy expenditure of a confined, fed steer. Hence, the contribution of muscle tone maintenance to SMR may be insignificant. However, the electrical activity of resting skeletal muscle in anesthetized rats is high (190). Ac- cording to Rohracher (175), the roughly lo-fold greater metabolic rate in homeotherms compared with poikilo- therms of similar body weight is due to the existence of “muscular microtremor.” However, this theory has not been tested. It has been estimated that the isolated, beating but unloaded heart (doing no external work) consumes 50% of the oxygen consumption of the rat heart in vivo in the standard state, and the arrested heart consumes -15% (48, 80, ZOO). Thus -85% of the in vivo oxygen consumption of heart is due to activation and contraction. It is thought that 50-60% of this ATP requirement is for the actinomyosin ATPase (53, ZOO). Therefore, it seems likely that between 43% (0.85 x 0.5) and 51% (0.85 X 0.6) of the oxygen consumption of rat heart in vivo is due to the actinomyosin ATPase, and thus this ATPase accounts for a minimum of between 1.3% (3 X 0.43) and 1.5% (3 X

0.51) of the whole body SMR in rats and between 5% (11 x

0.43) and 6% (11 x 0.51) of that in humans (calculated using the organ metabolic rates given in Table 1).

4. Protein turnover

The contribution of protein synthesis and breakdown to SMR has been estimated either by 1) measuring rates of protein synthesis or 2) measuring the inhibition of oxygen consumption when protein synthesis is inhibited. Waterlow (210) and Waterlow and Millward (212) have reviewed manv measurements of whole bodv pro-


TABLE 2. Contribution of major oxygen-consuming processes to oxygen consumption rate of rat tissues in standard state

Tissue Protein Synthesis Na+-K+-ATPase Ca’+-ATPase Other Major Contributors

Liver 24 5-10 5

Gastrointestinal tract 74 60 Kidney 6 40-70 Heart 3 l-5 Brain 5 50-60 Skeletal muscle 17 5-10

15-30 Significant

5

Gluconeogenesis (15-40%) and urea synthesis (12%) substrate cycling (26%), oligonucleotide synthesis (5%), proton leak (26%), nonmitochondrial (20%)

Gluconeogenesis (5%) Actinomyosin ATPase (40-50%), proton leak (15% max)

Proton leak (50%), nonmitochondrial (14%)

Data for protein synthesis are calculated using data for whole body protein turnover and fractional contribution of each tissue to this figure (210) and assuming that the average molecular weight of a peptide bond is 110, ATP peptide bond stoichiometry is 4, and ATP/O is 2.5. Data for Na’-K+-ATPase and Ca’+-ATPase are from Clausen et al. (53). Data for substrate cycling are from Rabkin and Blum (169). Data for glucose and urea synthesis are from Kusaka and Ui (132), Berry et al. (14), and Haussinger et al. (100). Estimates of contribution of actinomyosin ATPase activity to ATP turnover are from data of Challoner (48), Suga (200), and Clausen et al. (53). Note that, where it was necessary to assume a value for P/O to calculate data presented here, a figure of 2.5 has been used. Data for rat tissue oxygen consumption rates in the resting state are taken from Table 1. Note also that because data used to compile this table are taken from different sources; the sum of energy-requiring processes in some organs may exceed 100%.

tein turnover using labeled amino acids and estimate that protein synthesis contributes 15-18% of SMR. This estimate is calculated essentially by multiplying the protein synthesis rate by the ATP requirement (4 ATP equiv- alents per peptide bond), dividing by the P/O (assumed to be 3), and then comparing it with the SMR. This calculation will give a small underestimate of the contribution of the SMR (using a mechanistic P/O of 2.5 the contribution is - 1822%).

Estimates of the contribution of protein synthesis using specific inhibitors and determining the change in oxygen consumption rate in various in vitro tissue preparations range from 2 to 30% in different organs, with an overall contribution to SMR of 12-25% (reviewed in Ref. 124). Estimates of the contribution in different organs are given in Table 2. Reeds and Harris (172) found that the rate of protein synthesis has the same relationship with the body mass of an animal as its SMR; thus protein synthesis makes the same percent contribution to SMR in a variety of animals regardless of body size.

The net energy costs of protein degradation are not clear, although various authors have claimed that this process is a significant energy sink (44, 45, 148, 182, 192). The energy cost of proteolysis was assessed as the difference in the amount of cellular oxygen consumption inhib- itable by emetine (an inhibitor of ATP-dependent proteolysis and protein synthesis) and cycloheximide (an inhibitor of protein synthesis only). With the use of this method, estimates of the coupling of ATP-dependent proteolysis to cell oxygen consumption have been 11- 15% in resting rabbit reticulocytes (191, 192), 58% in Ehrlich mouse ascites tumor cells (148, 182), 7% in resting lymphocytes (45), and <l% in resting thymocytes (41). However, these data are difficult to interpret, since the effect of emetine only becomes significantly different from that of cycloheximide when the cells are incubated for over an hour in the

presence of the inhibitors despite the fact that proteolysis (measured as radiolabeled lysine release from cell protein) was abolished by emetine in <30 min. Thus the overall contribution of protein degradation is impossible to estimate at the moment, although the ATP requirement of ubiquitin-dependent proteolysis, -4 ATP/protein (102), would not appear to predict a significant role for ATP- dependent proteolysis as a contributor to SMR.

5. Substrate cycles

Newsholme (152) estimated that if the phosphofruc- tose-fructose bisphosphate cycle was fully active, it could account for -50% of an adult human’s daily expenditure. He concluded that stimulation of many such cycles could result in the conversion of significant amounts of chemical energy into heat. Rabkin and Blum (169) estimated the activity of five substrate cycles (glucose/glucose-6- phosphate, glycogen/glucose-6-phosphate, fructose-6-phos- phatelfrmctose 1,6-bisphosphate, phosphoenolpyruvatel pyruvate/oxaloacetate, and acetate/acetyl CoA) using fluxes calculated by a computer model that gave the best fit to data from radioisotopic incorporation experiments. The results indicated that the contribution of these five substrate cycles to resting metabolic rate in isolated hepatocytes was -26% (although the contribution to ATP turnover may be higher due to low effective P/O in hepatocytes). However, the potential for substrate cycling is probably larger in liver than most other tissues. Modeling of energy metabolism in young pigs suggests that substrate cycling in glycolysis, triglyceride turnover, and the Cori cycle account for 7.5% of SMR (171).

6. RNA and DNA turnover

Several attempts have been made tribution of oligonucleotide turnover

to quantify the resting cell to

con- oxy-


gen consumption using inhibitors of RNA and DNA synthesis such as actinomycin and amanitin. With the use of the inhibitor method, the coupling of oligonucleotide synthesis to cell oxygen consumption has been estimated as 12% in resting lymphocytes (45) and Ehrlich mouse ascites tumor cells (182) but was insignificant in resting thymocytes (44).

Another way of estimating the energy cost of oligonucleotide turnover is to directly measure the rate of synthesis by incubating cells or tissues with radiolabeled nucleotides. Although most RNA in the cell is rRNA and tRNA, the turnover rate of mRNA (and its precursor hnRNA) is much higher, so that the rate of RNA turnover in the cell is mostly due to mRNA (or hnRNA). Different types of mRNA also turnover at widely different rates, so that estimating the overall rate of RNA turnover is not easy. How- ever, with the use of such methods, the rate of synthesis of RNA was estimated as 0.105 mg l ml cells-’ l h-l in rabbit reticulocytes (158) and as 0.80 mg l 100 g-’ l h-l in perfused rat liver (136). With the use of an ATP:RNA nucleotide stoichiometry of 2 and with the assumption that the average molecular weight of a nucleotide in an RNA mole- cule is 320, the total ATP consumed by RNA synthesis in these systems can be calculated as 0.65 [(O. 105 x 10B3)/ 320 x Z] pmol ATP l ml reticulocytes-’ l h-’ and 5 [(O.S x

10-3)/320 x Z] pmol ATP l 100 g liver-’ l h-‘. The ATP consumption rate of reticulocytes and whole liver can be calculated as 95 pmol ATP l ml reticulocytes-’ l h-l (from Refs. 191, 192 and using an effective P/O of 1.9) and 44 pmol ATP l 100 g liver-’ l h-l (from Ref. 116 and using an effective P/O of 1.3). Thus the contribution of RNA turnover to tissue ATP consumption in reticulocytes and liver can be estimated as -1% (0.65/95 x 100) and 10% (5/44 x loo), respectively. Because the oxygen consumption driving ATP synthesis represents 50% (20) and a minimum of -70% (191) of the total oxygen consumption rate of isolated hepatocytes and reticulocytes, respectively, then the contribution of RNA synthesis to the overall oxygen consumption rate may be estimated to be 1% in thymocytes and -5% in liver.

Whole body RNA turnover has been estimated in adult humans from the excretion rate of rare bases or their metabolites in urine, as 0.5 pmol l kg body wt-’ l day-’ for mRNA, 0.5 pmol l kg-’ l day-’ for tRNA, and 0.03 pmol l

kg-’ l day-’ for rRNA (186). If the average size of primary mRNA transcripts being turned over was 20,000 nucleotides, then the mRNA turnover rate would correspond to 10 mmol nucleotides l kg-’ l day-‘, equivalent to 20 mmol ATP l kg-’ l day-‘, equivalent to 1.4 mol ATP. day-’ l human, equivalent to 0.56 mol0 l day-’ l human-’ (assuming a P/O of 2.5), equivalent to 1.8% of basal metabolic rate (with a basal oxygen consumption of 32 mol O/day). However, the average size of primary transcript being turned over is very uncertain, and the accuracy of the estimated mRNA turnover rate is also uncertain (186), so

this estimate of the contribution to SMR cannot be used with any confidence.

The contribution of DNA synthesis to SMR is thought to be much smaller than RNA, since the turnover is much lower (147).

7. Enzyme phosphorylation and signal transduction

Goldbeter and Koshland (84) have made rough estimates of the cellular energy expenditure on protein phosphorylation. They suggest that glycogen phosphorylase kinase could use a substantial proportion of cellular ATP turnover (> lo%), while adenosine 3’,5’-cyclic monophos- phate (CAMP)-dependent protein kinase and protein kinase C could make a significant contribution. They further suggest that since one in six mammalian proteins is re- versibly phoshorylated, the total cost of protein phosphor- ylations may be substantial. However, these conclusions were based on measured maximal rates rather than actual rates in cells, and there is no good estimate of the contribution of protein phosphorylation to ATP turnover.

Phospholipids, particularly those involved in signal transduction, undergo rapid energy-dependent synthesis and breakdown (12). Goldbeter and Koshland (84) state that phosphatidylinositol turnover in thrombin-stimulated platelets may transiently account for a significant proportion of the ATP consumption in these cells. However, Verhoeven et al. (208) concluded that the markedly en- hanced phospholipid metabolism during secretion in platelets involved only a small fraction of the total energy usage of the cell. Similarly, it was calculated (by radioisotopic measurement of inositol phospholipid turnover and knowing the ATP requirement of inositol phospholipid synthesis) that a loo-fold stimulation in inositol phospholipid turnover (compared with the basal turnover rate) in contracting canine trachealis smooth muscle would only cause an increase in ATP consumption equivalent to 1% of the basal ATP requirement (31). Thus the contribution of phospholipid turnover to SMR is unclear, but probably small.

The ATP requirement for CAMP turnover in platelets has been estimated to be 3-6% of the total ATP turnover (83, 84), and maximal stimulation of adenylate cyclase can completely deplete cell ATP in adipocytes (52). G proteins can hydrolyze GTP at a rapid rate, but again, there is no good estimate of this contribution to ATP turnover.

8. Gluconeogenesis

Gluconeogenesis requires ATP and occurs at a high rate under standard conditions. The flux through this pathway is generally measured in vivo by infusing small quanti- ties of a radiolabeled substrate of de novo glucose synthesis such as alanine and observing the rate of transfer of the radiolabel to glucose. This method may underestimate


the total rate of gluconeogenesis, since only glucose arising from radiolabeled alanine itself (or other gluconeogenetic substrates such as lactate which might also become labeled as a result of the alanine infusion) will be mea-

sured as gluconeogenesis (170). Furthermore, exchange may occur between radiolabeled carbon and carbon from sources (e.g., intramitochondrial oxaloacetate), which make no net contribution to gluconeogenesis (170). In overnight-fasted humans (143), the rate of gluconeogenesis was estimated, using radiolabeled alanine, to be -0.72 mol glucose/day. This is equivalent to 4.3 mol ATP/day (assuming 6 ATP/glucose) or 1.7 mol O/day assuming a P/O of 2.5. The average adult human basal oxygen consumption is 32 mol O/day (116, 184), so this represents 5% of human SMR or -28% of human liver SMR assuming liver accounts for 17% of human SMR (Table 1) and that -10% of gluconeogenesis occurs in the kidney cortex. In young (- 170 g) rats starved for 20 h (132), the rate of gluconeogenesis was estimated to be 14 mmol/day by measuring the rate of conversion of radiolabeled lactate to glucose. This is equivalent to -34 mmol O/day assuming 6 ATP/glucose and a P/O of 2.5. This is roughly equivalent to 7% of rat SMR assuming that a 170-g rat consumes 490 mmol O/day (calculated from data in Ref. 71) or 32% of liver SMR if liver accounts for 20% of rat SMR (Table 1) and 90% of gluconeogenesis occurs in liver.

Data from isolated hepatocytes may enable us to crudely determine the significance of any underestimate of gluconeogenesis in the above data. Because glycogen stores are completely depleted in 18 to 24-h starved rats (14), all glucose production in hepatocytes isolated from these rats will be from gluconeogenesis. In hepatocytes isolated from 18 to 24-h fasted rats, the rate of glucose production under conditions designed to maximize the rate of gluconeogenesis has been estimated to be between 0.5 and 1.6 pmol l g wet wt cells-’ l mir-? (14,86). Knowing the weight of cells in whole liver (14), we can estimate that the rate of gluconeogenesis in rat liver is - lo-28 mm01 l liver-l l day-‘, which is equivalent to 24-67 mmol O/day driving gluconeogenesis in the whole rat assuming 6 ATP/glucose and a P/O of 2.5. An adult rat (-300 g) consumes -840 mmol O/day (116), and hence, gluconeogenesis in the liver can be estimated to account for between 3 and 8% of rat SMR and between 15 and 40% of rat liver SMR. However, the above calculation of the contribution of gluconeogenesis to SMR depends on the proportion of gluconeogenesis that is produced from glycerol (stoichiometry = 2 ATP/glucose) compared with the other major gluconeogenetic substrates alanine, glutamine, and lactate (6 ATP/glucose). Glycerol carbon has been estimated to account for -5% of the total carbon used for gluconeogenesis in 16- to 18-h fasted adult dogs (94). If this is also true for rats, then the above calculation for hepatocyte gluconeogenesis would overestimate the true gluconeogenetic flux by -3%. Using the inhibitor per-

fluorosuccinate, Gregory and Berry (86) found that a 72% inhibition of glucose production resulted in a 21% reduction in oxygen consumption rate, and thus we might guess

that a 100% inhibition would produce a 30% reduction in oxygen consumption rate, i.e., that gluconeogenesis accounts for -30% of liver metabolic rate under conditions designed to maximize de novo glucose production. How- ever, it is not clear what other effects perfluorosuccinate has on hepatocyte metabolism, so the results of this study are hard to interpret. Thus it appears that a variety of techniques and systems for measuring gluconeogenesis give approximately the same answer: that de novo glucose synthesis accounts for between 5 and 8% of SMR in humans and between 3 and 8% in rats.

9. Urea synthesis

Urea synthesis requires ATP and continues in the standard state. Total urea synthesis is 25 g/day equivalent to 0.42 mol/day in an average human, requiring - 1.7 mol ATP/day (assuming 4 ATP/urea) and thus 0.67 mol O/day, and thus 2% of the total oxygen consumption in humans. Almost all this urea synthesis occurs in liver, so that this represents 12% of the SMR of human liver. In perfused liver from fed rats or hepatocytes prepared from these livers, urea production can account for a significant proportion (up to 30%) of liver SMR (14, 86, 100). Total nitrogen excretion in 48-h starved rats was measured (55) to be 73 mg N l day-’ l 100 g body wt-‘, with 75% of urinary N in urea this is equivalent to 3.9 mmol urea l day-’ l 100 g body wt’ (73 x 0.75/14), which would require 15.6 mmol ATP l day-’ l 100 g-l (with 4 ATP required per urea synthesized), and thus 3.12 mmol O2 l day-’ l 100 g-l (with a mechanistic P/O of 2.5). The same rats had an oxygen consumption of 4.43 g O2 l day-’ l 100 g-’ (55), equivalent to 138 mmol OZ l day-’ l 100 g-‘, and thus the oxygen requirement for urea synthesis represents 2.3% of rat SMR or 12% of rat liver SMR. Thus it appears that urea synthesis may also make a small but significant contribution to SMR in rats and humans (-2%).

10. Other ATP-consuming processes contributing to standard metabolic rate

Exocytosis and endocytosis allow the movement of large molecules into and across cell membranes and the modulation of cell responses to hormones by altering re- ceptor availability. Energy is required for fibroblast endocytosis (160), exocytosis in amoeba (157), and for synaptic vesicle cycling in nerve terminals. However, the energy cost of these processes would be minor if one or a few ATP molecules were expended per vesicle or large mole- cule moved (147).

D. Service Functions

Another way of looking at the contribution of various processes to whole body oxygen consumption in the stan-

742 DAVID F.S. ROLFE AND GUY C. BROWN Volume 77

dard state is to divide metabolism into service and nonservice functions (7). Service functions are supracellular functions that serve the whole body, such as kidney excretion, nervous function, and pressure-volume work done by the heart and lungs, whereas nonservice functions are for maintenance of the cell itself (e.g., maintenance of ion gradients). Service functions have been estimated (see references in Refs. 7, 71) to be between 25 and 50% of the SMR of an animal. However, in practice, it is not clear to what extent the service and nonservice functions are distinguishable (e.g., kidney and neuronal work is mainly ion transport), and hence, an assessment of the relative contribution of the service and nonservice functions to SMR is difficult.

One way to estimate the contribution is to compare the oxygen consumption of tissues incubated in a minimal medium in vitro (where service functions are likely to be minimal) to the oxygen consumption measured in vivo in the standard state. Measurements of summed tissue slice oxygen consumption in rats (71), mice (142), and dogs (142) indicate that an average of -80% of the SMR of these animals can be accounted for by resting tissue oxygen consumption. Thus on the order of 20% of SMR may be concerned with service functions by this estimate. How- ever, the difference between in vitro and in vivo oxygen consumption rates might have many other causes unre- lated to service functions, so this is unlikely to be a reli- able estimate.

A potentially better estimate may be obtained by con- sidering the individual organs. It has been estimated that the arrested heart consumes -15% of the oxygen consumption of the rat heart in vivo (48, 80, ZOO), thus -85% of the oxygen consumption is concerned with the service function. In the brain, light anesthesia or sleep reduces brain oxygen consumption by lo-30%, whereas deep anesthesia, removing almost all electrical activity, reduces oxygen consumption by -4O-50% (4, 68); thus -40% of oxygen consumption is concerned with service functions in brain. In the kidney, 50-60% of oxygen consumption is thought to be coupled to sodium transport (53, 98, 197), and up to 5% is coupled to gluconeogenesis (88,197); most of this probably has a service function. Resting skeletal muscle performs no service function (unless we include heat production, which all tissues produce). The gut performs no obvious service function in the standard state, since this state is postabsorptive. However, in the liver, gluconeogenesis and urea synthesis (and protein synthesis for export) may be classified as service functions and have been estimated above to require between 40 and 50% of liver SMR in humans (see sects. IIC?? and IIC9). If 40% of brain, 60% of kidney, 50% of liver, and 85% of heart oxygen consumption was used for service functions in humans, the sum of the oxygen consumption driving these functions would account for -30% of the total SMR in humans and -20% in rats.

E. Relative ATP Use by Cellular ATPases

Section IIC estimated the extent to which the major ATP-consuming processes were coupled to cellular oxygen consumption, which closely approximates their contribution to SMR. Clearly, if all the oxygen consumption of a tissue was used to produce ATP (i.e., the effective P/O = 25), then the contribution of these processes to ATP consumption would be the same as their coupling to oxygen consumption or SMR. However, as mentioned in section II&?, the effective P/O may be less than maximal in liver and skeletal muscle and perhaps in other tissues also. We can obtain an estimate of the contribution of a particular ATP consumer to whole body ATP turnover by multiplying the coupling of the ATP consumer to oxygen consumption by the maximum P/O (2.5), then dividing by the whole body oxygen consumption rate multiplied by the effective whole body P/O (1.8; see sect. rB3). In this way we can estimate the contribution to whole body ATP consumption of protein synthesis (25-30%), Na’-K+- ATPase (19~ZS%), Ca2+-ATPase (4-8%), actinomyosin ATPase (2-s%>, gluconeogenesis (7-lo%), and ureagenesis (3%). The sum of these estimated contributions is 60- 87%. Other major contributors are substrate cycling and RNA synthesis. Obviously, the contribution of different ATP users differs dramatically in different organs.

F. Reactions That Uncouple Standard Metabolism

We have, until now, discussed many of the contributors to SMR in terms of coupling reactions such as the Na’-K’- and Ca2+-ATPases and protein synthesis. How- ever, it is important to consider which reactions uncouple these processes, since without the uncoupling reactions there would be no SMR.

In the brain, about one-half of the ATP production is coupled to the Na+-K’-ATPase and ion cycling at the pre- and postsynaptic membranes (4, 68, 220), and thus the reactions uncoupling most of brain metabolism are the voltage- and neurotransmitter-gated ion channels in these membranes. In the kidney, most of energy metabolism (50-60%; Refs. 53, 98, 197) is estimated to be involved in sodium reabsorption, so the uncoupling reactions of this organ are the sodium permeability of the luminal membranes of the tubule and glomerular filtration membranes. In heart and muscle, a smaller but still significant proportion of metabolism is uncoupled by voltage- and ligand- gated ion channels in their membranes. Protein degradation uncouples protein synthesis in all tissues and thus accounts for -20% of SMR (see sect. IIC~). Muscle contraction in heart alone may account for between 1 and 6% of SMR (see sect. 11c3), and thus muscle relaxation, cross- bridge cycling of actinomyosin complexes, must account for a small but significant uncoupling of metabolism.

J2dy mu’ ORIGIN OF STANDARD METABOLIC RATE 743

The mitochondrial proton leak uncouples very roughly 20% of SMR and may be particularly important in skeletal muscle (see sect. IBZ). This proton flux appears to be mediated by the bilayer (39). Around Z-50% of the proton leak of isolated mitochondria is apparently due to the protonophoric activity of free fatty acids in the bilayer (39). However, it is unclear whether this free fatty acid effect is present in vivo.

Nonmitochondrial oxygen consumption in heart, liver, and skeletal muscle alone may account for -10% of rat SMR (see sect. IIN). This oxygen consumption is catalyzed by a heterogeneous group of enzymes and so is uncoupled by a large number of different reactions.

Cellular energy metabolism may be uncoupled not just by uncoupling processes in the same cell, but also by uncoupling processes elsewhere in the body or outside the body in the environment. These uncoupling processes are often transport processes, where cellular energy metabolism is coupled to a transcellular transport process and the transport is reversed either in the environment or by other transport processes in the body. For example, Na+ is transported from one extracellular space to another through intervening cells in the gut and kidney. This process requires a continuous input of free energy, but no free energy is stored, because Na’ derived from the environment (as food) is returned to the environment (as urine). The pumping of the blood around the body by the heart is another process that does not come to equilibrium because the transport is in a loop (and thus the pump is short circuited), which requires a continuous input of energy. The movement of limbs and the body by the alter- nate contraction and relaxation of antagonistically ar- ranged skeletal muscles is a process that again conserves little free energy, but results in transport against frictional forces (although this is not involved in SMR). Gluconeo- genesis in liver and kidney is uncoupled by glycolysis in other tissues, and this cycle functions to transport energy between tissues. These uncoupling processes at the supracellular level are related to the service functions discussed in section ID.

G. Substrate Utilization and Glycolytic ATP Production in Standard Metabolic Rate

The relative contributions of carbohydrate, fat, and protein oxidation to SMR can be estimated from the respiratory quotient (ratio of COB produced to O2 consumed) and urea production (189). In humans in the postabsorptive state, the respiratory quotient is -0.82 (189), and the relative contributions can be estimated to be -60,30, and 10% for fat, carbohydrate, and protein oxidation, respectively. With longer fasting, the respiratory quotient falls further and has been measured to be 0.7 in 48-h starved rats, indicating that 90% of SMR is derived from fat oxida-

tion, an insignificant amount from net carbohydrate oxidation, and -10% from amino acid oxidation (55).

What proportion of total ATP production is produced by glycolysis rather than mitochondria in the standard state? During glucose oxidation, 2 ATP/glucose are produced by glycolysis and -23 ATP/glucose by the mitochondria (assuming a maximal ATP production of 31 ATPI glucose with 20% uncoupling by the proton leak, see Ref. 19); during glycogen oxidation, the glycolytic ATP production would be 4 ATP/glucose unit. Thus, if 30% of SMR is derived from carbohydrate oxidation, then 2.4-4.4% (0.3 X 2/25 or 0.3 X 4/27) of total ATP production is derived from glycolysis during glucose or glycogen oxidation. However, glycolysis may also produce lactate, which may recycle to glucose via gluconeogenesis in the liver. Because no lactate (or alanine) accumulates in the standard state, the rate of this (anaerobic) glycolysis can be estimated from the rate of gluconeogenesis. In section 11c8, we saw that the rate of glucose turnover in the stan-

dard state has been measured to be 0.72 mol glucose/day in humans and 14 mmol glucose/day in rats, and this is equivalent to 1.44 mol glycolytic ATP/day in humans and 28 mmol glycolytic ATP/day in rats (assuming 2 glycolytic ATP/glucose). The total oxygen consumption is 32 mol O/ day in humans and 490 mmol O/day in rats (see sect. 11C8), and with the assumption of an effective P/O of 1.8, this is equivalent to 57.6 mol mitochondrial ATP/day in humans and 882 mmol mitochondrial ATP/day in rats. Thus we can estimate that anaerobic glycolytic ATP production is 2.4% of total (glycolytic and mitochondrial) ATP production in humans and 3.1% in rats in the standard state. The total contribution of glycolytic ATP production (aerobic and anaerobic) is thus roughly 4.6-7.7%. Of course, the proportion varies with tissue; most glycolysis is occurring in the brain in the standard state.

H. Summary

Contributions to SMR can be quantified in terms of coupling to oxygen consumption, ATP turnover, or uncoupling. In the standard state, -90% of mammalian oxygen consumption is used by the mitochondria; of this 90%, -20% is uncoupled by the mitochondrial proton leak, whereas the other 80% is coupled to ATP synthesis. Of this ATP production, -28% is used by protein synthesis, 19-28% by the Na+-Kf-ATPase, 4-8% by the Ca’+-ATPase, 2-8% by actinomyosin ATPase, 7- 10% by gluconeogenesis, and 3% by ureagenesis, with mRNA synthesis and substrate cycling also contributing. Protein synthesis is uncoupled by protein degradation, the Na’ pump by Na+ channels and Na+-coupled transport, the calcium pump by Ca” channels, and gluconeogenesis by glycolysis. Con- tributions differ in different tissues (Table 2 and Fig. 3).

DAVID F. S. ROLFE AND GUY C. BROWN Volume 77 744

Total oxygen

consumption in the

Standard State

Mitochondrial

Non- Mitochondrial

Total mitochondrial

oxygen consumption

Coupled to ATP synthesis

-------a

Uncoupled by proton leak

\

\ \

\

\ \

4

III. CONTRIBUTION OF FEEDING, COLD, AND

MUSCLE USE TO FIELD AND MAXIMAL METABOLIC RATE

In this section, we very briefly summarize the major processes involved in field and maximal metabolic rate. This section is intended to put the above discussion in its context rather than to review the subject in detail, a task which has alrea.dy been fulfilled by others (e.g., Refs. 15, 110, 133). The value of the field metabolic rate (assessed, using the doubly labeled water method, as the 24-h metabolic rate of nonbreeding free-living animals) has been calculated to be roughly three times greater than SMR (measured under laboratory conditions) for mammalian species in the wild (151) and roughly twice SMR for do- mestic animals (15). However, in humans, the field metabolic rate (based on measurements of heat production made while the subject carried out normal daily activities) may be between 1.5fold (for office workers) and 2. l-fold (for heavy occupational work) greater than the SMR (15). The major components of these increases in metabolic rate are muscular activity, shivering and nonshivering thermogenesis, and the thermogenic effect of food. The relative contributions of these processes will obviously depend on the level of locomotion and exposure to cold.

Total ATP

consumption in the

Standard State

Protein synthesis

Na+/K+ ATPase

m e - - - - - -

Ca2 +

ATPase

Gluconeo- genesis m-B-----

Ureagenesis -----a--

Actinomyosin AVasie - - --ew

Others (including RNA synthesis and

substrate cycling)

FIG. 3. Estimated contribution of processes to energy utilization in standard state. First column shows contribution of mitochondrial (80%) and nonmitochondrial (10%) oxygen consumption to total respiration rate in standard state. Second column shows proportion of mitochondrial respiration used to drive ATP synthesis (80%) and proton leak (20%) in standard state. Data shown in first and second columns were calculated using data for proton leak and nonmitochondrial oxygen consumption from liver, heart, and skeletal muscle only and ignoring possible contribution of these processes in other tissues (see section II@.

Third column represents contribution of ATP-consuming processes to total ATP consumption in standard state calculated as outlined in section HE.

A. Cold-Induced Thermogenesis

Exposure to temperatures below thermoneutrality causes increases in metabolic rate (105), and increases of up to five times SMR (16) have been noted in cold- acclimatized mammals in response to catecholamines, the principal hormones involved in cold-induced thermogenesis. In addition, long-term cold exposure may also alter standard metabolism; for example, the rate of oxygen consumption of cold-acclimatized rats is >25% higher at thermoneutral temperature than that of warm-acclimated rats (estimated from the data of Ref. 75). The response to cold is divided into shivering and nonshivering thermogenesis. Shivering thermogenesis results from muscle tremor and thus increased activity of the actinomyosin ATPase, Na+- Kf-ATPase, and Ca2+-ATPase, leading to increased oxidative phosphorylation in skeletal muscle. However, if cold exposure occurs for longer periods (several weeks), heat production by shivering is gradually replaced by nonshivering thermogenic processes. Although a large proportion of this nonshivering thermogenesis is the result of increased activity of BAT (75), other tissues, particularly skeletal muscle, may contribute (54, 75, 117). Cold accli- matization studies in mammals indicate that there is a general increase in the ion permeability of membranes in


a variety of tissues, resulting in increased activity of the Na’-K’-ATPase and Ca’+-ATPase (47,53). It has been estimated that -4O-50% of the response to cold (6°C) in cold- acclimatized rats is due to nonshivering thermogenesis in BAT (73). The increase in oxygen consumption of adult cold-acclimatized rats in response to cold (6°C) is around two times SMR (75, 117). Adult humans lack BAT and may have relatively little cold exposure depending on conditions so that the contribution of cold-induced thermogenesis to field metabolic rate is probably smaller in humans than other mammals, although this obviously depends on environmental conditions.

B. Feeding-Induced Thermogenesis

Feeding has both short-term (i.e., after a meal) and long-term (over- or underfeeding) effects on metabolic rate. The short-term effect (formerly known as specific dynamic action) typically stimulates metabolic rate by -25%, with peak stimulation l-2 h after a meal in humans. This thermogenic effect of feeding contributes - 15% of field metabolic rate in omnivores and up to 30% in ruminants (15). The mechanisms of this stimulation are still not entirely clear but include 1) the stimulation of pathways leading to the temporary synthesis of storage compounds, 2) the use of the ingested food as substrate for generating ATP, and 3) hormonal changes, in particular, activation of the sympathetic nervous system. One important component of the thermogenic response is the metabolism of ingested amino acids in the liver resulting in glucose, fat, urea, and protein synthesis, accompanied by amino acid oxidation (119). Swallowing, hydrolysis/ fermentation and absorption of the food, and enzyme secretion all make very small additional contributions (15). Baldwin and Smith (6) estimated that after a carbohydrate meal the increased heat production could be attributed to conversion of glucose to fat (30% of the increase), conversion of glucose to glycogen (14%), hydrolysis of the carbohydrate in the gut (5%), synthesis of digestive proteins (IS%), and active transport during absorption (24%). Milligan and Summers (147) and Kelly and McBride (124) attributed most of the increase in energy turnover to increases in protein turnover and Na’-Kf-ATPase after feeding. Nutritional status (as distinct from the thermogenic effect of food) also affects SMR (179). Fasting lowers metabolic rate by up to 40% (15), and overeating may increase it (179, HO), although this effect is more contro- versial (15, 103, 180).

C. Muscle Use-Induced Thermogenesis

Maximal metabolic rate attained by exercise for short periods is roughly 10 times SMR (204), although this figure varies considerably between and even within species (see

sect. I). The extra oxygen consumption and heat production is the result of muscle contraction in heart and skeletal muscle. About 87% of body oxygen consumption is consumed by skeletal muscle during heavy work in hu-

mans and 5% by the heart (145). The estimated relative ATP consumption by different ATP users in isometrically contracting muscle is 6580% by the actinomyosin ATPase, lo-25% by the Ca’+-ATPase, 5-10% by the Nat- K’-ATPase, and -5% by myosin phosphorylation (53, 109, 110). In shortening muscle, the oxygen consumption is higher, and calcium cycling may require 20-50% of the ATP turnover (53, 109). During isometric contraction of the heart, 50-60% of ATP turnover has been attributed to the actinomyosin ATPase, 15-35% to the Ca2+-ATPase, and 15% to noncontractile functions (48, 53, 200). Muscle contraction also does some work including the compres- sion and acceleration of blood in the heart and limbs, and air in the lungs, and this energy is dissipated as heat in the blood vessels and lungs/external air, respectively. The overall efficiency of the heart is -20% but rises to 40% or higher if the energy requirement for activation and noncontractile functions is subtracted (80,200). Up to 20- 30% of the calorific input can be used for external work by skeletal muscle during walking, running, and cycling in humans (144). The efficiency of performing external work apparently increases with the body size of species (15). The uncoupling reactions of maximal metabolic rate are thus mainly 1) calcium and sodium channels, 2) muscle relaxation, 3) cross-bridge cycling of the actinomyosin during maintained contraction, and 0) dissipation of the physical work either inside or outside the organism.

IV. HEAT PRODUCTION AND FREE ENERGY DISSIPATION

A. Thermogenesis and Enthalpy Dissipation

The cellular production of heat and dissipation of free energy have sometimes been confused. The free energy change of a reaction @G) is made up of two components: the enthalpy change (LH) and the entropy change @S), where AG = LW-TAS (and T is the absolute temperature). For cellular reactions and transport processes, the entropy component (-TLS) of AG is often large so that the enthalpy change cannot be equated with the free energy dissipation (AG). The heat production of an enzyme or transport reaction is given by -AH of the overall reaction so that the heat production by a cellular reaction is given by -AH X J, where J is the net rate of the reaction.

Prusiner and Poe (168) pointed out that most of cellular heat production must occur in the process of ATP production as opposed to ATP usage. They showed this by comparing the enthalpy of substrate (glucose and fatty acid) oxidation to that of ATP hydrolysis (after accounting


TABLE 3. Estimated standard molar enthalpy change of cellularr reactions

IV (succinate oxidation) evolves -62 kJ/mol 0 (subtracting the AH of 6 Hf from the AH of succinate oxida-

Reaction

a: Glucose + 12 0 + 6 CO2 + 6 HZ0 b: Palmitate + 46 0 -+ 16 COP + 16 HZ0 c: Lactate + 6 0 --) 3 CO2 + 3 HZ0 d: NADH2 + 0 + NAD’ + HZ0 e: ATP + Hz0 + ADP + Pi

f: H,=, + H,+, (mitochondria)

-AH” tion, assuming 6 H+/O for succinate), the ATP synthase evolves -24 kJ/mol ATP (subtracting the AH for ATP

kJ/mol kJ/mol 0 hydrolysis from the AH for 3 Ht assuming 3 H+/ATP)

2,803 234 equivalent to 48 kJmo1 0 (assuming an effective mito-

10,014 218 chondrial P/O of Z), the adenine nucleotide translocator 1,367 228 evolves

256 256 -21 kJ/mol ATP (assuming one net charge trans-

20 40 located) equivalent to 42 kJ/mol 0 (assuming an effective

15 150 mitochondrial P/O of Z), and the proton leak evolves -21 g: Succinate + 0 -+ fumarate + HZ0 \ 152 152 kJmo1 H’ equivalent to 42 kJ/mol 0 (assuming 10 H’ are

Enthalpy changes (m) at 25°C are calculated for reactions as pumped per 0 and 20% of these protons return via the

shown and per mole of atomic oxygen. All enthalpies are negative. Val- proton leak). These estimates are crude because the mito- ues do not include buffer ionization or metal binding. Enthalpy changes chondrial membrane potential in vivo is not known accu- for reactions a-c are from Blaxter (15), and those for reaction d are from Poe et al. (163) and Burton (42). Standard enthalpy changes for

rately. A total of -80% of cellular heat production is

reaction e are from Podolsky and Morales (162) and are corrected for evolved by the proton pumps and transporters of the mito- buffer ionization. Enthalpy of ATP hydrolysis per atom of oxygen is chondrial inner membrane, as pointed out by Prusiner calculated assuming a ATP/O of 2. AH including buffer ionization has been estimated to be 47 kJ/mol ATP in muscle (58) and 15 kJ/mol (11). For reaction f, values for mitochondrial proton transport are enthalpy equivalent of mitochondrial membrane potential taken as 150 mV, and value per mole of atomic oxygen is calculated assuming a H’/O of 10 (see text).

for the ATP/O). Table 3 gives literature values of some enthalpy changes mated for protons

and the crossing

potential the mitoc

energy change esti- hondrial inner mem-

and Poe (168). However, expressing cellular heat production in percentage terms can be misleading because many cellular reactions may be endothermic, and considerable amounts of heat may be taken up or produced during binding or release of metabolites, protons, and metal ions by other molecules (58, 209).

lent of one charge transfer (6 kJ) from the enthalpy equivalent of ATP hydrolysis (21 kJ)], equivalent to 7 kJ/mol 0 [multiplying the heat production per ATP (15 kJ/mol ATP)

If we assume that the average plasma membrane potential is 60 mV (51, 107, 156), then the Na’-Kf-ATPase

by the fraction of total ATP consumption used by the Na+

evolves 15 kJ/mol ATP [subtracting the enthalpy equiva-

pump (0.23; see sect. IL%‘) and by the effective mitochondrial P/O (2; see sect. II&?)]. The return of Na’ back into the cell is associated with -6 kJ/mol Na’ (one charge

brane (excluding enthalpies of buffer ionization).

ever, high ratios of glycolysis to oxidative metabolism can result in a significant heat production from glycolysis in isolated cells, partly due to the net pH change (82). Glycol-

The enthalpy change for oxidation of a particular sub-

ysis from glucose to lactate produces 63-80 kJ/mol lactate when buffer ionization is included (82) (equivalent to lo-

strate by NAD can be estimated by comparing the en-

14 kJ/mol 0). The heat production by individual steps in

thalpy of combustion of the substrate to that of NADH

glycolysis has been considered in Reference 58. In contrast to NAD reduction, the oxidation of NADH evolves a

per oxygen atom consumed. From Table 3 we can see

considerable amount of heat, mostly produced mitochondrial respiration, oxidative phosphorylation, and proton

that little or no heat production occurs during glycolysis

leak. The enthalpy of combustion of NADH is about -250 kJ/mol 0, whereas the equivalent potential heat production from the mitochondrial membrane potential is 150 kJ/

or the oxidation of glucose or fatty acids by NAD. How-

mol0 (assuming an H’/O of 10 and membrane potential of 150 mV), and the enthalpy of ATP hydrolysis is roughly -40 kJ/mol 0 (assuming an effective mitochondrial P/O of 2). The missing enthalpy is lost as heat. From Table 2 and the estimated H+ stoichiometries, we can estimate that complex I evolves 20-40 kJ/mol 0 (subtracting the AH of succinate oxidation from the AH of NADH oxidation and then subtracting the AH of 4H+/O assuming an H’/2e- of 4, see Ref. 37 and sect. IBM), complex III and

moved through 60 mV) equivalent to 8 kJmo1 0 [multiplying the AHINa’ (6) by the Na’/ATP stoichiometry of the pump (3) and the fraction of total ATP consumption used by the pump (0.23) and the effective mitochondrial P/O (a)] if the return is not coupled to other reactions, and -5.5 kJ/mol 0 (6 x 2 x 2 x 0.23) with the return of K’ to the exterior of the cell. These estimates are crude because the contribution of the Na’-K+-ATPase to ATP turnover and the level of the plasma membrane potential varies from tissue to tissue. The ATP (and GTP) consumption associated with protein synthesis can be estimated to produce 12 kJ/mol 0 [multiplying the AH of ATP hydrolysis (21 kJ/mol ATP) by the fraction of total ATP consumption used by protein synthesis (0.28; see sect. IL%‘)

and the effective mitochondrial P/O (a)]. Because relatively little energy is stored in the peptide bond (-5%), most of this heat will be evolved in protein synthesis rather than protein breakdown. The ATP (and GTP) is used in several different steps of peptide chain elongation, and heat evolution is also likely to be spread through several steps, but the enthalpy values of the intermediates are not generally known. Heat evolution during muscle


contraction has been studied in some detail, and the enthalpy of a number of actinomyosin intermediates has been estimated (58, 109). If Z-7% of total ATP turnover is via the actinomyosin ATPase in the standard state (as very roughly estimated in sects. IL!? and 11c3), then the steady-state heat production by the actinomyosin ATPase and cross-bridge cycling of intermediates is small in this state. Of course, this proportion is much higher during maximal metabolism when 50-80% of the oxygen consumed is coupled to the actinomyosin ATPase, and thus up to 40 kJ/mol 0 may be generated by the actinomyosin ATPase. For shortening muscle, a proportion of the enthalpy change may not be lost as heat but rather passed on as internal or external work. The heart obviously performs work in accelerating blood through the vessels, and this work is dissipated as heat within the blood vessels as the blood is decelerated by the vessel walls or by turbulence. The ratio of work done to AH of ATP or PCr has been estimated to be 40- 100% in heart so that in some conditions the heat evolved by the cross-bridges themselves may be smaller than expected (80). Also, the work of intestinal or uterine smooth muscle is dissipated inter- nally. In contrast, skeletal muscle may perform external work (up to ZO-30% of calorific input) (144, 219).

Many of the reactions of metabolism produce or consume protons, and these protons are buffered by intracellular buffering groups. The binding and debinding of protons on these various groups produce or take up heat. The amounts of heat involved are large and vary with the type of buffer (58, 209). The M of ATP hydrolysis corrected for buffer ionization is about -20 kJ/mol ATP, whereas the observed LW of ATP hydrolysis including buffer protonation varies from -47 to -15 kJ/mol ATP and depends on which intracellular buffers are present (11, 58, 168). If we used the largest of these values (LW = -47 kJ/mol ATP) to quantify where the heat production occurs, we would estimate that -40% of heat production (2 x 47/230, where we assume an effective P/O of 2 and

a LUY of glucose or NADH oxidation of 230 kJ/mol 0) occurs in the processes of ATP utilization (plus coupled and uncoupling processes and proton binding to the buffer), whereas -60% occurs during the processes of ATP synthesis (including heat uptake due to ionization of the buffer). If the intracellular buffers take up a large amount of heat on ionization (as appears to happen in skeletal muscle), then part of the heat released during net ATP synthesis is taken up by the buffers as they ionize to provide the 0.8 Hf required per ATP synthesized, and this heat is released again during net ATP hydrolysis. How- ever, in the steady state, there is no net uptake or release of these protons, and therefore, the heat exchanges involved in this buffering are not relevant to the steady state. Similarly, the binding and unbinding of metal ions with metabolites and proteins is associated with heat uptake or release. For example, calcium binding and release

TABLE 4. Estimated free energy change of selected cellular reactions

Reaction -AG"', kJ/mol

-AG

kJ/mol kJ/mol 0

a: Glucose + 12 0 + 6 COz + 6 Hz0 b: Glucose + 2 ADP -+ lactate +

2 ATP c: NADH:! + 0 + NAD+ + Hz0 d: H,=, -+ HL (mitochondria) e: ATP + Hz0 + ADP + Pi

2,880 2,808 234

136 92 8 219 202 202

15 18 180 31 63 126

Free energy changes given are standard at pH 7 (AGO’), actual (AG), and actual per atom of oxygen associated with reaction, at 25°C. All AG values are negative. Reaction a: AGo’ value from Blaxter (15); AG calculated for cellular concentrations: 10 mM glucose, 1 mM CO2 and 20 PM Oz. Reaction b: AGo’ value from reactions a and e; AG calculated for 10 mM glucose, 2 mM lactate and AG ATP from reaction e. AG per 0 is nG/12. Reaction c: values were calculated using E”’ for oxygen/water couple of +815 mV and Eh of 788 mV for 20 ,uM O2 (37); an E”’ for NADIUNAD couple of -320 mV and Eh of -260 mV (Eh may be as low as -320 mV in heart) (69). AE,, was converted to AG by multiplyng by 2F. Reaction d: values quoted are median values (see text); AG per 0 is calculated on basis of 10 H+/O (see text). Reaction e: cytoplasmic AG for ATP has been estimated to be 65 kJ/mol in muscle (go), 63 kJ/mol in liver (113), 60-65 kJ/mol in heart (62), and 62 kJ/mol in brain (68). AG per 0 is calculated assuming ATP/O of 2.

in muscle may involve significant transient heat changes (109) that do not contribute to the steady state. The reactions of Table 3 have been corrected for buffer ionization and metal binding.

B. Free Energy Dissipation and Efficiency

Most of the internal energy (U) or enthalpy (H) of substrates is dissipated by the processes of ATP production. In contrast, most of the free energy (G) is dissipated by the processes of ATP consumption and the uncoupling reactions. Table 4 gives literature estimates of the free energy dissipation by various reactions in energy metabolism. These estimates depend on the measured metabolite concentrations and thus potentially vary with tissue and metabolic state; however, the measured variation is small in terms of this analysis.

If a (driving) reaction A+C is coupled to another (driven) reaction B+D, then the efficiency of the coupling can defined as

JBmD x AGDmB

JAeC x AGAmC

where JASC is the rate of conversion of A to C and JBwD is the rate of conversion of B to D, and nGAsC is the difference in free energy between A and C and aGDmB is the free energy difference between D and B. From this definition, it fol- lows that an efficiency of less than one can arise either from 1) an uncoupling reaction or slip reaction from A to


C, so that some of A is converted to C without B being converted to D, or 2) a fully coupled reaction displaced from equilibrium. Efficiency is not directly related to heat production, since efficiency is defined in terms of free energy rather than enthalpy.

The efficiency of mitochondrial oxygen consumption and oxidative phosphorylation depends on the effective P/O and is maximally 78% (assuming a P/O of 2.5 from NADH, a AG of 202 kJ/mol 0 for oxidation of NADH, and a AG of 63 kJ/mol ATP for ATP hydrolysis, see Table 4, then efficiency = 2.5 X 63/202) and minimally 62% (assuming an effeetive mitochondrial P/O of 2.0). Most of this inefficiency and free energy dissipation is due to the proton leak, whereas the component processes of oxidative phosphorylation are relatively efficient (see below).

In contrast to ATP production, the ATP-utilizing reactions are relatively inefficient. The efficiency of the Na’- K’-ATPase has been estimated to be -57% in brain neurons (35 kJ/mol ATP transferred to the Na+, Kf, and electrical gradients of the plasma membrane with AGATP of 62 kJ/mol) and 67% in glial neurons (42 kJ/mol transferred for the same assumed AGATP) (68). Similar efficiencies have been calculated for the Na+-K’-ATPase of squid axons and mammalian erythrocytes (202). However, an efficiency of 85% has been estimated for resting heart cells (59). The efficiency of the plasma membrane Ca”-ATPase has been estimated to be -42% in brain based on a ATP:Ca?H+ stoichiometry of 1:l:Z and a plasma membrane AG(Ca2’) of 26 kJ/mol(68). During protein synthesis, 4 mol ATP (AG -63 kJ/mol) are hydrolyzed per mole of peptide bonds formed (AGO -13 kJ/mol) (15). Thus the efficiency of protein synthesis is -5%. This does not take into account the relative concentrations of amino acids and protein or the sequence of amino acids in the peptide chain, but this is unlikely to alter the conclusion that protein synthesis is very inefficient. The gross efficiency of muscle contraction (that is, mechanical work output divided by AG of substrate oxidation) is maximally lo-30% in heart (SO, 200) and 20-30% in skeletal muscle (80, 219), whereas the contractile efficiency (that is, total mechanical energy output divided by the AGATP) has been estimated to be 40-80% in cardiac muscle (80, 200) and 45-77% in skeletal muscle (219).

We can calculate the proportions of the total free energy available for the oxidation of glucose dissipated by various cellular processes using the data in Table 4. Thus glycolysis dissipates -3% of the available free energy, substrate dehydrogenation (pyruvate transport, pyruvate dehydrogenase, and the tricarboxylic acid cycle) dissipates -lo%, the respiratory chain dissipates -lo%, the proton leak dissipates -15%, and ATP synthesis and transport dissipate -4%, the Na’-Kf-ATPase plus coupled and uncoupling transport of Na+ and Kf dissipates - 11% (assuming ATP consumption dissipates 54% of the total AG, and 20% of the total ATP consumption is used by

the Na’ pump), and protein synthesis dissipates -14% (assuming protein synthesis uses 28% of total ATP at 5% efficiency). This leaves -33% of the free energy dissipation unaccounted for, part of which will be due to the Ca’+-ATPase, RNA turnover, and substrate cycling. These calculations assume that fatty acids or glucose is being oxidized. Oxidation of amino acids is a little less efficient (15, 119), but the relative contribution of amino acid oxidation to energy metabolism is small (see sect. IIG).

Most of the free energy dissipation by glycolysis is thought to occur at hexokinase, phosphofructokinase, and pyruvate kinase (129). Most of the free energy dissipation of substrate dehydrogenation is thought to occur at pyruvate, isocitrate, and oxoketoglutarate dehydrogenase (24, 129). Most of the free energy dissipation by the respiratory chain occurs at cytochrome oxidase (36, 37, 150). Mitochondrial ATP synthesis and transport appears to be close to equilibrium so that little free energy is dissipated by these reactions. If 20% of the respiratory protons return via the proton leak, then the leak acts as a major free energy dissipating reaction. Gunter and Jensen (89, 118) have measured the efficiencies of component steps of oxidative phosphorylation in isolated mitochondria and estimated the power losses in state 3 (i.e., the maximal rate of ATP synthesis) to be 12% by electron transport, 10% by proton leakage, 19% by substrate and product transport, and 10% by phosphorylation, resulting in the transfer of 48% of the initial free energy to ATP. However, as we have seen, the in vivo efficiencies are probably higher at least during standard metabolism, because the AGATP is higher than in state 3. The efficiency may be lower during maximal metabolism.

C. Distribution of Energy Dissipation

Most (but not all) of the internal energy dissipation and heat production from metabolism occurs at the reactions of the mitochondrial inner membrane, whereas most (but not all) of the free energy dissipation occurs at the ATP utilizing and uncoupling reactions (Fig. 4). Why is this? One way to rationalize the distribution of heat production is to note that virtually all the free energy available from the oxidation of NADH or glucose is enthalpic, with a very small entropy production, whereas roughly two-thirds of the free energy available from ATP hydrolysis is entropic (see Tables 3 and 4). Thus, even if oxidative phosphorylation were at equilibrium, the transfer of the free energy from the NADH2 + 0, -+ NAD+ + H20 reaction to the ADP + Pi -+ ATP reaction must involve the loss of two-thirds of the internal energy of the former reaction (note, however, that the enthalpy and free energy values used here do not include buffer ionization). It is conceiv- able that reactions that produce a large amount of heat are confined to membranes (or other relatively rigid struc-

July kxu’

A H loss Heat Production

Respiratory chain

Proton leak

ATP synthesis

ATP usage

ORIGIN OF STANDARD METABOLIC RATE 749

A G loss

Free Energy dissipation

Substrate supply

Respiratory chain

Proton leak

ATP synthesis

ATP usage

FIG. 4. Estimated contribution of processes to heat production (MY) and free energy dissipation (AG). Contribution of each process to total LW loss and total LIG loss is represented by size of correspond- ing box in bar chart. ATP synthesis consists of mitochondrial ATP synthase and adenine nucleotide and phosphate translocators. ATP usage includes ATPases and all coupled processes and uncoupling reactions downstream of ATP production. Substrate supply includes all processes from substrates to mitochondrial NADH. Heat production is open ended, because although most processes produce heat, some processes may also take up heat.

tures) to prevent the proteins involved from denaturation by the released thermic energy.

Why is such a large proportion of the free energy of metabolism dissipated by the ATP-utilizing reactions, i.e., why are these reactions apparently so inefficient? One possible explanation is that the products of these reactions, for example, proteins, the Na+, Kf, and Ca2’ gradients, and cell membrane potentials, need to vary in level to control cell function, and this variation must be independent of ATP/ADP. An alternative explanation is that the relatively low enthalpic contribution of ATP hydrolysis means that there is relatively little energy available per reaction, and this may cause relatively slow kinetics if the ATP-utilizing reactions were endothermic.

V. CHANGES AND DIFFERENCES IN METABOLIC RATE

Differences in metabolic rate may be either within an organism or species or between different species or phvla. Changes in metabolic rate within an animal can be

classed as short term or long term. Short-term changes include those due to cold, feeding, and muscle use (discussed in sect. III). Long-term changes in SMR include those due to nutritional status (under- or overfeeding) or age. Differences in SMR between species are mainly due to differences in body mass or phylogeny. Standard metabolic rate is fairly constant for a particular individual (ignoring daily and seasonal variations; see Ref. 15) and is similar for different individuals of the same species, age, and weight, although small intraspecies differences do exist and several authors have proposed differences that account for these, such as variations in lean body mass (78, 79, 146). Standard metabolic rate changes with age; it increases from birth to maturity and falls slightly in old age. The mass-specific SMR (metabolic rate per kg mass of organism) declines with age (15). Standard metabolic rate varies with thyroid status. Individuals with lower or higher than normal thyroid levels have lower or higher than normal metabolic rates, respectively (e.g., Refs. 91, 203). Standard metabolic rate varies with body mass. Big- ger animals have lower metabolic rates per unit body mass; for example, the metabolic rate per gram of mouse is -25 times that of an elephant (30, 183). There are also differences in the SMR of species of the same mass but different phyla, although note that these differences may be due to differences in thyroid status (112). Warm- blooded mammals have roughly 4-5 times the SMR of cold-blooded reptiles of the same body mass and preferred body temperature (112).

The latter part of this section deals with the molecular origin of the differences in SMR between species of different body size and phylogeny. However, before that, we examine which cellular reactions control SMR, because only changes in reactions that control SMR will actually affect SMR.

A. Control of Cellular Respiration

The concepts of “control” and “regulation” are often confused. Ambiguity can be avoided by taking the proposition “x control y” to mean that a change in x causes a significant change in y. Whereas “x regulates y” means that in vivo changes in y are brought about by changes in x. Of course, x can only regulate y if x has significant control over y in vivo. In the metabolic control theory, the extent to which an enzyme (or transporter) controls a steady-state rate is quantified as a flux control coefficient (see Refs. 40, 120, 121). This coefficient is the percent change in the steady-state rate divided by the (small) percent change in the concentration of the enzyme. For a particular system in a particular condition, the sum of the control coefficients for all enzymes over a particular flux is equal to 1.

It had previously been assumed that all metabolic


pathways would have a single, rate-limiting step, but extensive investigations of control within isolated mitochondria have shown that control is shared by a number of reactions, and the distribution of control changes with conditions (see Refs. 34, 87, 93, 138, 217). Within cells, it is often assumed that metabolic rate is solely controlled by the ATP-utilizing processes, to which the ATP-producing processes passively respond. However, recent evidence suggests that processes other than those involved in ATP consumption have significant control over oxygen consumption rate. Control over cellular oxygen consumption has been examined in isolated liver cells (41, 97) and perfused rat skeletal muscle (177) in “resting” conditions, which may approximate the standard state. Control over resting mitochondrial oxygen consumption in rat liver cells has been shown to be shared between the “substrate oxidation” (29-35% of the control), “proton leak” (ZZ- 25%) and “ATP turnover” (41-49%) pathways (41, 97). Note here that, because these experiments were carried out in isolated cells, any control by substrate supply to the tissue is not included in the analysis. If significant control was invested in tissue substrate supply, as some consider to be the case (50, 61), then the amount of control invested in the cellular pathways would be less significant. In perfused skeletal muscle, the same analysis gave a similar control distribution: substrate oxidation was found to have 44 t 28% of the control, proton leak had 38 t 21%, and ATP turnover had 21 t 9% (177). There are indications in many other mammalian tissues that the ATP-utilizing reactions do not have all the control over cellular oxygen consumption (34, 101). This is indicated by the fact that respiratory substrates or increased mitochondrial free calcium can stimulate oxygen consumption in these tissues. In many tissues, a large amount of control does lie in the ATP-utilizing processes, but this control is not necessarily in the ATP-utilizing reactions themselves. For example, the rate of the Na+-K+-ATPase is usually determined by the Na+ permeability, and the actinomyosin ATPase by Ca” permeability, whereas protein synthesis and the associated ATP usage are controlled by a large number of processes (34). Thus it is clear that the control of oxygen consumption rate is widely distributed among metabolic pathways and is not solely a property of the ATP-consuming processes.

B. Causes of Species Differences in Standard Metabolic Rate Related to Body Size

and Phylogeny

What changes underlie these differences in metabolic rate between species and phyla? Part of the differences in SMR mentioned above are due to differences in the mass of the metabolically most active organs (e.g., liver, kidney, and brain) as a proportion of body mass (15, 183),

and part may be due to differences in the ratio of relatively active to inactive cells in organs (e.g., neurons/glia in brain; Ref. 193). However, the differences in SMR between animals of varying body mass, thyroid status, and phylogeny have been shown to correlate with differences in the metabolic rate of liver cells isolated from those animals (23, 97, 165) so that the decrease in SMR per kilogram in larger animals is partly due to 1) relatively more bone, skin, and muscle and relatively less of the more active internal organs; 2) a reduced proportion of relatively active cell types within organs; and 3) for a given cell type less active metabolism.

What metabolic changes underlie the differences in cellular metabolism in different species and phyla? It is unlikely that change in the level of a single cell reaction or process is sufficient to explain a large change in SMR, because the relevant reactions would need to have high control coefficients over SMR over a large range of rates. This is unlikely to be true, because what evidence we have (see sect. VA) suggests that none of the relevant reactions has a control coefficient over oxygen consumption approaching 1. If an enzyme has a control coefficient of < 1, raising the enzyme’s activity will bring the reaction closer to equilibrium and the enzyme will lose control. It has been shown that large increases in the rate of metabolic pathways can only be brought about by proportional activation of all the enzymes with significant control over the pathway (70). Thus it is unlikely that a large difference in metabolic rate could result from a change in the level of a single enzyme.

A more likely hypothesis for large differences in metabolic rate is that all the major metabolic pathways change in proportion, i.e., that all enzymes and transporters change in concentration to the same extent as the minimal metabolic rate. This hypothesis is discussed below with particular reference to differences in metabolic rate related to body mass, thyroid status, and phylogeny.

1. Body mass-related changes in metabolic rate

The SMR of adult mammals of different species is related to their body weight (M,) by the equation SMR cx(MJ”.75. The following components have been shown to scale in proportion to species metabolic rate in species of different body mass: total number of mitochondria in liver [ cy (Mb)0*72; Ref. 1961; total mitochondrial inner membrane area from liver, heart, kidney, brain, and skeletal muscle [c~(M~)‘*~~; Ref. 651; total liver and skeletal muscle cytochrome oxidase [cY(M~)‘*~‘-‘*~~; Refs. 115, 1301; whole body cytochrome c content [a(Mb)Oa7; Ref. 641; whole body protein turnover [cY(M~)‘-~~; Ref. 1721; whole body RNA turnover [ (x(MJ’*~~; Ref. 1851; and Na’-K+-ATPase activity in liver and kidney slices [cY(MJ’*~~; Ref. 571. This tends to support the proposition that changes in SMR are brought about by changes in all major cellular reactions.


In liver cells isolated from mammals of widely rang- various cellular processes to be maintained is that this ing body mass (20-750 kg), the proportion of hepatocyte would preserve a particular pattern of control distribu- respiration used to drive proton leak (-ZO%), ATP turn- tion. Analysis of the distribution of control over oxygen over (-65%), and nonmitochondrial oxygen consumption consumption in mitochondria isolated from the livers of (- 15%) was the same in all cases (167), although the different animals (from data in Ref. 167), from different proton permeability of liver mitochondria isolated from tissues of the same animal (178), and even from plants the same mammals varied considerably at the same pro- (60, 126) indicates that the distribution of control over ton-motive force (164). The relationship between the oxy- mitochondrial oxygen consumption (and other variables) gen consumption rate (per unit mass) of the liver cells appears to be a conserved property in different tissues

lated from [(x(MJ~*‘*] was virtually identical to the mass- specific SMR [cY(MJ -‘*17] of the animals from which they

and the body mass (Mb) of the mammals they were iso- tance of this pattern of control. Indeed, the need to con- serve a certain control pattern presumably reflects the

and species. This might be taken to indicate the impor-

were taken (165).

we would expect all metabolic intermediate levels to not If the whole of metabolism sealed with body size,

importance of certain key points in a pathway that are

in metabolic rate that are produced by alteration of the subject to in vivo regulation, to effect short-term changes

change with body size. However, it has been reported that activity of only some points in a metabolic pathway. the AGATP (estimated by NMR) of heart in vivo decreases with body size in mammals (62). If confirmed in other tissues, this would suggest that ATP production was de- C. Summary

creased more than ATP utilization with increasing body size. In apparent contrast to this, one of the driving forces for ATP synthesis, the mitochondrial membrane potential, increases with increasing body size in isolated liver cells (167), perhaps implying that ATP utilization was decreased more than ATP production in liver.

In summary, there is evidence that most of metabolism involved in SMR scales roughly in proportion to SMR in mammals of different body size, but the finding that metabolite levels do not remain constant suggests that the changes in metabolism are not exactly proportionate.

In conclusion, it appears that differences in SMR seen in animals of different body mass and phylogeny are the result of a proportional change in the activity of most cellular processes. Large-scale and long-term activation of metabolic rate may require such proportional activation because control over metabolic rate is distributed throughout metabolism so that a large activation of metabolic rate cannot be achieved by activating only a single or small number of metabolic reactions.

2. Metabolic rate and phylogeny

VI. FUNCTIONS OF STANDARD METABOLIC RATE

Comparisons of energy metabolism in homeotherms and poikilotherrns have shown that there is a roughly fivefold higher SMR in homeotherms than in poikilo- therms of the same body size and preferred body temperature, whereas the total mitochondrial inner membrane area of homeotherms is three- to fourfold higher (65, 66), the Na’ and K+ permeabilities of the cell membrane are five- to sevenfold higher (67), and the proton permeability of the mitochondrial inner membrane is four- to fivefold

Several different functions could be suggested for the processes underlying SMR. These include 1) maintenance required due to undesigned uncoupling reactions, 2) production of heat to maintain body temperature, 3) maintenance and heat production for maximal metabolic rate, -4) endowment of increased sensitivity of key metabolic reactions to effecters, and 5) reduction of harmful free radical production.

greater in the homeotherm (23). A comparison of the isolated liver cells of rat and a poikilotherm of equivalent A. Maintenance of Cellular Composition Against

body mass has shown that although the cells of the ho- Undesigned Uncoupling Reactions

meotherm had a fourfold greater oxygen consumption

leak, and total ATP turnover are similar in the two cell

rate, the proportions of oxygen consumption attributed to nonmitochondrial oxygen use, the mitochondrial proton

types (23). The contribution of the Na’-K+-ATPase was

Some authors (e.g., Ref. 188) have suggested that the low entropy structures of cells (e.g., macromolecules, ion gradients) are continually degrading and dissipating due to uncatalyzed or undesigned reactions, e.g., protein and

also found to be the same (23). Again, this suggests that all the components of metabolism are increased in proportion to metabolic rate.

It is interesting to note that one of the reasons why one might expect the balance between the activity of the

fat oxidation reactions and ion leaks, so that a continual input of free energy is required to maintain the low entropy states. It is important to distinguish between the function of coupling and uncoupling reactions here. If the uncoupling reactions are undesigned reactions, then the


coupling reactions could be regarded as functioning in if we accept that a body temperature above ambient is maintaining the levels and organization of components advantageous, then we must also accept that heat produc- (e.g., proteins and DNA). tion is functional.

We have seen above that the mitochondrial proton leak is a significant uncoupling reaction and might be regarded as uncatalyzed and/or undesigned (35,39). How- ever, the fact that the activity of this leak can change so dramatically with body mass and phylogeny suggests that the leak is partly under the control of the organism and thus presumably designed. The Nit+, Kf, and Ca2’ conduc- tances at the plasma membrane are almost certainly mainly mediated by ion channels rather than undesigned leaks (35). The extent to which protein degradation is

ultimately caused by uncatalyzed reactions leading to protein denaturation is not known but is generally thought to be small (102). Overall, the contribution of undesigned leaks to metabolic rate is probably small, at least in mammals. Thus most of the uncoupling reactions involved in SMR must have functional roles. The fact that poikilother- mic animals of the same size and essentially the same structure and metabolism as homeotherms have roughly fivefold lower SMR at the same body temperature also tends to suggest that most of the uncoupling (and coupling) processes in homeotherms are designed.

C. High Standard Metabolic Rate Enables High Maximal and Field Metabolic Rate

The argument that the function of SMR in homeother- mic animals is to generate heat may appear somewhat circular, since it is generally assumed that the major ad- vantage of homeothermy is a higher metabolic rate. How- ever, the argument is not circular if the higher body temperature allows a higher growth rate, faster reproductive cycle, and higher maximal metabolic rate.

B. Standard Metabolic Rate as a Means of Heat Production

It is possible that the level of SMR in mammals is determined by the necessity of maintaining a constant body temperature, and thus a rate of heat production which matches the rate of heat loss. Mammals need to maintain a constant body temperature of between 33 and 39°C and at thermoneutral environmental temperatures, the required heat is provided by standard metabolism. However, if the need for heat far exceeded the need for free energy, then we might expect heat production to be localized to a particular organ and/or a particular uncoupling reaction early in metabolism (cf. thermogenesis in brown adipose tissue). We have seen that heat production during standard metabolism is not localized to one organ, and significant amounts of heat are generated both before and after the production of ATP. This might be taken to suggest that in mammals the function of SMR is not only to produce heat, but also to supply free energy to drive cellular reactions whose function is not solely to generate heat. The production of heat is compatible with the production of free energy by the very same reactions, so that these are not mutually exclusive functions. Conversely, it could be argued that the sole function of SMR is to produce and use free energy and that heat is an inevitable waste product of this process. However, if this heat production were removed, then the body temperature of mammals would fall to that of the environment. Thus,

A high SMR may enable an organism to have a higher growth rate, faster reproductive cycle, and higher maximal metabolic rate not just because it provides a higher body temperature, but also because it maintains the metabolic machinery necessary for a high maximal metabolic rate. Thus, for example, fast running requires high levels of muscle actinomyosin to cause contraction, high rates of muscle calcium and sodium transport, high levels of mitochondria to supply the ATP, increased vasculariza- tion, increased heart and lung capacity, increased glycogen and fat storage, and increased capacity to interconvert metabolites. Thus a higher maximal metabolic rate or growth rate requires more metabolic machinery (i.e., more enzymes, transporters, membranes, metabolites, organ- elles, and cells), and the maintenance of more machinery may require more energy (e.g., for protein turnover) even in the standard state. Some support for this idea comes from the fact that in most animals maximal metabolic rate is always roughly 10 times SMR even though the absolute value of SMR varies greatly. Thus we might infer that a high SMR enables a high maximal metabolic rate. How- ever, this explanation echoes the argument of section VIA that the function of SMR is maintenance, which we have already dismissed, since most of the leaks that underlie SMR appear to be designed rather than undesigned. A stronger argument might be that a high maximal metabolic rate requires a high SMR to maintain a high body

temperature (see sect. VI@ or increase potential for regulation (see sect. VrD).

Even in the standard state, there are a number of organism functions that require a continuous supply of free energy. For example, the processing of information in the brain continues in the standard state and is of obvious survival value. Brain energy metabolism appears to be uncoupled mainly by plasma membrane ion channels, and these uncoupling processes participate in the information processing. Thus some uncoupling processes are functional in the standard state.

July 1997 ORIGIN OF’ STANDARD METABOLIC RATE 753

D. Standard Metabolic Rate Increases the Potential for Regulation

An alternative, but not mutually exclusive, explanation for the presence of a large SMR in mammals is that the presence of futile cycles (coupling and uncoupling processes) in cellular metabolism confers on metabolism 1) decreased transition times to new steady states, 2) a higher sensitivity to effecters, and 3) more sites at which control is exerted, as first proposed for substrate cycles by Newsholme and Underwood (153). The rapid synthesis and degradation of proteins, mRNA, phospholipids, metabolites, and signal molecules does not affect the steady- state levels but does decrease the transition times to new steady states and thus enables more rapid regulation of levels. High activity of the processes producing and consuming an intermediate relative to the net rate of production also means that relatively small changes in the production or consumption result in large changes in the net rate, and thus increase the sensitivity of control. Mito- chondrial proton leak, which uncouples mitochondrial ATP synthesis, may increase sensitivity and decrease response time to changes in ATP utilization in the cell. The transmembrane Na+, Kf, and Ca2+ fluxes must similarly respond rapidly and sensitively to demand, and here again, the high level of cycling of these ions across cellular membranes may facilitate this response. Thus the uncoupling of metabolism may provide mechanisms for increasing the sensitivity and rate of response of metabolic pathways to effecters.

E. Standard Metabolic Rate as a Means of Regulating Harmful Free Radical Production

A further function of SMR might be to control the production of harmful free radicals, either by the mitochondria themselves or by nonmitochondrial processes (195). There is ample evidence that mitochondria are a

major source of cellular superoxide and hydrogen peroxide (95). Experiments using isolated heart mitochondria have shown that the production of these reactive oxygen species (ROS) is dramatically increased when mitochondrial state 4 oxygen consumption is inhibited at certain point(s) in the respiratory chain. The site of free radical production is probably the semiquinone of the Q cycle (205), but complex I may also be involved (2). It has been shown that uncoupling or the availability of ADP prevents mitochondrial free radical production (49, 139), and thus increased ATP utilization or proton leak can decrease ROS production. Nonmitochondrial ROS production also appears to be important, particularly via electron transport chains in the endoplasmic reticulum (95). The rate of ROS production is roughly proportional to ambient oxygen concentrations (96).

Thus it appears that the rate of generation of superoxide free radicals is related to the concentration of oxygen within the cell and the degree of reduction of the species that donate electrons to O2 to form superoxide. One might therefore suppose that ROS production would become most problematic in the resting cell, since such a cell would consume less oxygen and have a higher concentration of reduced electron-transport-chain components that could donate an electron to oxygen to form superoxide. This might be overcome by 1) increasing tissue oxygen consumption to decrease tissue oxygen levels and 2) increasing mitochondrial proton leak and cellular ATP utilization to decrease the reduction level of the mitochondrial respiratory chain. Thus SMR might function to decrease ROS production (195). However, this would seem to be a very expensive way to decrease ROS production, given that there appear to be many other ways of achieving the same ends. For example, tissue oxygen levels can be and are decreased by arterial and arteriolar resistance to blood flow, and the respiratory chain can be and is maintained partially oxidized by restriction of electron entry into the chain. Also, a high SMR to decrease oxygen levels is accompanied by a high level of mitochondria, and thus an increased potential for producing ROS. Overall, it seems unlikely that decreasing ROS production is a major function of SMR, but it might play a minor role.

F. Summary

Thus the functions of SMR are not fully understood. Undesigned uncoupling reactions probably do not make a significant contribution to SMR in mammals. The greater metabolic rate of homeotherms compared with poikilo- therms points to an important role for SMR in heat generation, but this is also compatible with SMR providing free energy for functional coupling reactions (such as protein synthesis) and uncoupling reactions (such as information transfer and processing). The uncoupling of metabolism may also serve to increase the sensitivity and rate of response of metabolic processes to effecters.

VII. CONCLUSIONS

We have seen that there are important differences and distinctions between the cellular reactions that 1) couple to oxygen consumption, 2) uncouple metabolism, 3) hydrolyze ATP, -4) control metabolic rate, 5) regulate metabolic rate, 6) produce heat, and r) dissipate free energy. The quantitative contribution of different cellular reactions to these processes is difficult to estimate accurately at present, because of the vast amount of metabolic information required. However, we believe a useful first estimate can now be made, and this will help to focus attention and research on outstanding problems. We esti-


mate that -90% of mammalian oxygen consumption in the standard state is mitochondrial, of which -20% is uncoupled by the mitochondrial proton leak and 80% is coupled to ATP synthesis. Of the total ATP synthesis, -21530% is used by protein synthesis, 19-28% by the Na+-K+-ATPase, 4-S% by the Ca2+-ATPase, 2-8% by the actinomyosin ATPase, 7- 10% by gluconeogenesis, and 3% by ureagenesis, with mRNA synthesis and substrate cycling also making significant contributions. The main cellular reactions that uncouple standard energy metabolism are the Nit+, Kf, H’, and Ca2’ channels and leaks of cell membranes and protein breakdown. Cellular metabolic rate is controlled by a number of processes including metabolic demand and substrate supply. The differences in SMR between animals of different body mass and phylogeny appear to be due to proportionate changes in the whole of energy metabolism. Heat is produced by some reactions and taken up by others but is mainly produced by the reactions mitochondrial respiration, oxidative phosphorylation, and proton leak on the inner mitochondrial membrane. Free energy is dissipated by all cellular reactions, but the major contributions are by the ATP- utilizing reactions and the uncoupling reactions. Several functions for SMR can be proposed, but the relative importance of these is unclear.

12.

13.

14.

15.

16.

17.

18.

19.

20<

21

22.

23.

24.

REFERENCES 25.

1. ALTMAN, P. L., AND D. S. DITTMER. Biologic& Handbooks: Metab- olism. Bethesda, MD: FASEB, 1968.

2. AMBROSIO, G., J. L. ZWEIER, C. DUILIO, P. KUPPSAMY, G. SANT- ORO, P. P. ELIA, I. TRITTO, P. CIRILLO, M. CONDORELLI, M. CHIARELLO, AND J. T. FLAHERTY. Evidence that mitochondrial respiration is a source of potentially toxic oxygen free radicals in intact rabbit hearts subjected to ischemia and reflow. J. Biol. Chem. 268: 18532- 18541, 1993.

26.

27.

28.

3. ASCHOFF, J., B. GUNTHER, AND K. KRAMER. Energiehaus- falt und Temperatur-regulation. Munich, Germany: Urban & Schwarzeberg, 1971.

4. ASTRUP, J., P. M. SORENSEN, AND H. R. SORENSEN. Oxygen and glucose consumption related to Na’-K’ transport in canine brain. Stroke 12: 726-730, 1981.

29.

5. BACHELARD, H. S. Energy utilisation by neurotransmitters. In: Brain Work: T#he Coupling of Function, Metabolism and Blood Flow in the Brain, edited by D. H. Ingvar and N. A. Lassen. Copen- hagen: Academic, 1975, p. 79-81.

30. 31.

6. BALDWIN, R. L., AND N. E. SMITH. Molecular control of energy metabolism. In: The Control of Metabolism, edited by J. D. Sink. University Park: Pennsylvania State Univ. Press, 1974, p. 17-34.

7. BALDWIN, R. L., N. E. SMITH, J. TAYLOR, AND M. SHARP. Manipu- lating metabolic parameters to improve growth rate and milk secretion. J. Anim. Sci. 51: 1416-1428, 1980.

8. BEAVIS, A. D. Upper and lower limits of the charge stoichiometry of mitochondrial electron transport. J. Biol. Chem. 262: 6165-6173, 1987.

32.

33.

9. BEAVIS, A. D., AND A. L. LEHNINGER. The upper and lower limits of the mechanistic stoichiometry of mitochondrial oxidative phosphorylation. Eur. J. Biochem. 158: 315-322, 1986.

10. BERGER, M., S. A. HAGG, AND N. B. RUDERMAN. Glucose metabolism in perfused skeletal muscle. Interaction of insulin and exercise on glucose uptake. Biochem. J. 146: 231-238, 1975.

11. BERNARD, S. A. Ionisation constants and heats of tris(hydroxy-

34.

35.

36.

37.

methyl)aminomethane and phosphate buffers. J. Biol. Chem. 218: 961-969, 1956. BERRIDGE, M. J. Inositol triphosphate and diacylglycerol as sec-

ond messengers. Biochem. J. 220: 345-360, 1984. BERRY, E. A., AND P. C. HINKLE. Measurement of the electrochemical proton gradient in submitochondrial particles. J. Biol. Chem, 258: 1474- 1486, 1983. BERRY, M. N., A. M. EDWARDS, AND G. J. BARRITT. Isolated He- patocytes: Preparation, Properties and Applications. Amsterdam: Elsevier, 1991, p. 121-178. BLAXTER, K. Energy Metabolism in Animals and Man. Cam- bridge, UK Cambridge Univ. Press, 1989. BLOCK, B. A. Thermogenesis in muscle. Annu. Rev. Physiol. 56: 535-577, 1994. BOVERIS, A., N. OSHINO, AND B. CHANCE. The cellular production of hydrogen peroxide. Biochem. J. 128: 617-630, 1972. BRAND, M. D. The contribution of the leak of protons across the mitochondrial inner membrane to standard metabolic rate. J. Theor. Biol. 145: 267-286, 1990. BRAND, M. D. The stoichiometry of proton pumping and ATP synthesis in mitochondria. Biochemist 16: 20-24, 1995. BRAND, M. D., L.-F. CHIEN, E. K. AINSCOW, D. F. S. ROLFE, AND

R. K. PORTER. The causes and functions of mitochondrial proton leak. Biochim. Biophys. Acta 1187: 132- 139, 1994. BRAND, M. D., L.-F. CHIEN, AND P. DIOLEZ. Experimental discrim- ination between proton leak and redox slip during mitochondrial electron transport. Biochem. J. 297: 27-29, 1994. BRAND, M. D., L.-F. CHIEN, AND D. F. S. ROLFE. Control of oxidative phosphorylation in liver mitochondria and hepatocytes. Bio- them. Sot. Trans. 21: 757-762, 1993. BRAND, M. D., P. COUTURE, P. L. ELSE, K. W. WITHERS, AND

A. J. HULBERT. Evolution of energy metabolism. Biochem. J. 275: 81-86, 1991. BRAND, M. D., AND M. P. MURPHY. Control of electron flux through the respiratory chain in mitochondria and cells. Biol. Rev. 62: 141-193, 1987. BRINDLE, K., S. FULTON, J. SHELDON, AND S. WILLIAMS. Probing the properties of enzymes in vivo using NMR. Biochem. Sot. Trans. 23: 376-381, 1994. BRINDLE, K., S. FULTON, J. SHELDON, AND S. WILLIAMS. Probing the properties of enzymes in vivo using NMR. Biochem. Sot. Trans. 23: 376-381, 1995. BRINDLE, K. M., M. J. BLACKLEDGE, R. A. J. CHALLISS, AND G. K. RADDA. 31P NMR magnetisation transfer measurements of ATP turnover during steady state isometric contraction in the rat hind limb in vivo. Biochemistry 28: 4887-4893, 1989. BRINDLE, K. M., J. E. FULTON, A. J. KINSMAN, AND G. K. RADDA. 31P NMR magnetisation transfer measurements of flux between inorganic phosphate and ATP in yeast cells over-producing phosphoglycerate kinase. Biochem. Sot. Trans. 14: 1265, 1986. BRINDLE, K. M., AND S. KRIKLER. ‘lP NMR magnetisation transfer measurements of phosphate consumption in S. cerevisiae. Bio- chim. Biophys. Acta 847: 285-292, 1985. BRODY, S. Bioenergetics and Growth. New York: Reinhold, 1945. BROWN, C. B., M. PRING, AND R. F. COBURN. Inositol lipid turnover and compartmentation in canine trachealis smooth muscle. Am. J. Physiol. 256 (Cell Physiol. 25): C375-C383, 1989. BROWN, G. C. The relative proton stoichiometries of the mitochondrial proton pumps are independent of the proton motive force. J. Biol. Chem. 264: 14704-14709, 1989. BROWN, G. C. Total cell protein concentration as an evolutionary constraint on the metabolic control distribution in cells. J. Theor. Biol. 153: 195-203, 1991. BROWN, G. C. Control of respiration and ATP synthesis in mammalian mitochondria and cells. Biochem. J. 284: I- 13, 1992. BROWN, G. C. The slips and leaks of bioenergetic membranes. FASEB J. 6: 2961-2965, 1992. BROWN, G. C., AND M. D. BRAND. Thermodynamic control of electron flux through mitochondrial bcl complex. Biochem. J. 225:

399-405, 1985. BROWN, G. C., AND M. D. BRAND. Proton/electron stoichiometry of mitochondrial complex I estimated from the equilibrium thermodynamic force ratio. Biochem. J. 252: 473-479. 1988.

July 1.w?’ ORIGIN OF STANDARD METABOLIC RATE 755

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54 .

55 .

56.

57.

58.

59.

60.

61.

62.

CHUNG, F.-Z., H. W. WEBER, AND M. M. APPLEMAN. Extensive reversible depletion of ATP via adenylate cyclase in rat adipocytes. Proc. Natl. Acad. Sci. USA 82: 1614-1617, 1985. CLAUSEN, T., C. VAN HARDEVELD, AND M. E. EVERTS. Signifi- cance of cation transport in control of energy metabolism and thermogenesis. Physiol. Rev. 71: 733-774, 1991. COLQUHOUN, E. Q., M. G., AND CLARK. Open question: has thermogenesis in muscle been overlooked and misinterpreted? News Physiol. Sci. 6: 256-259, 1991. CORI, C., AND G. T. CORI. The fate of sugar in the animal body. II. The relation between sugar oxidation and glycogen formation in normal and insulinized rats during the absorption of glucose. J. Biol. Chem. 70: 557-576, 1926. COTTLE, W. H., AND L. D. CARLSON. Regulation of heat production in cold-adapted rats. Proc. Sot. Exp. Biol. Med. 92: 845-849, 1956. COUTURE, P., AND A. J. HULBERT. Relationship between body mass, tissue metabolic rate and sodium pump activity in mammalian liver and kidney. Am. J. Physiol. 268 (Regulatory Integrative Comp. Physiol. 37): R641-R650, 1995. CURTIN, N. A., AND R. C. WOLEDGE. Energy changes and muscular contraction. Physiol. Rev. 58: 690-761, 1978. DAUT, J. The living cell as an energy transducing machine. A minimal model of myocardial metabolism. Biochim. Biophys. Acta 895: 41-62, 1987. DIOLEZ, P., A. K. KESSELER, F. HARAUX, M. VALERIO, K. BRINK- MANN, AND M. D. BRAND. Regulation of oxidative phosphorylation

in plant mitochondria. Biochem. Sot. Trans. 21: 769-773, 1993. DIPRAMPERO, P. E. Metabolic and circulatory limitations to VozMAx at the whole animal level. J. Exp. Biol. 115: 319-331, 1985. DOBSON, G. P., AND J. P. HEADRICK. Bioenergetic scaling: metabolic design and body-size constraints in mammals. Proc. Natl.

Acad. Sci. USA 92: 7317-7321, 1995. 63. DORA, K. A., S. M. RICHARDS, S. RATTIGAN, E. Q. COLQUHOUN,

AND M. G. CLARK. Serotonin and norepinephrine vasoconstriction in rat hindlimb have different oxygen requirements. Am. J. Physiol. 262 (Heart Circ. Physiol. 31): H698-H703, 1992.

BROWN, G. C., AND M. D. BRAND. The ratio of the H+/O and H’/ ATP stoichiometries in mitochondria over a range of np values and the thermodynamic H+/ATP stoichiometry. EBEC Short Report 5: 123, 1988. BROWN, G. C., AND M. D. BRAND. On the nature of the mitochon-

drial proton leak. Biochim. Biophys. Acta 1059: 55-62, 1991. BROWN, G. C., R. P. HAFNER, AND M. D. BRAND. A “top-down” approach to the determination of control coefficients in metabolic control theory. Eur. J. Biochem. 188: 321-325, 1990. BROWN, G. C., P. L. LAKIN-THOMAS, AND M. D. BRAND. Control of mitochondrial respiration and ATP synthesis in isolated rat liver cells. Eur. J. Biochem. 192: 355-362, 1990. BURTON, K. The enthalpy change for the reduction of nicotin- amide-adenine dinucleotide. Biochem. J. 143: 365-368, 1974. BUTTGEREIT, F., AND M. D. BRAND. A hierarchy of ATP-consuming processes in mammalian cells. Biochem. J. 312: 163-167, 1995. BUTTGEREIT, F., M. D. BRAND, AND M. MULLER. ConA-induced changes in energy metabolism of rat thymocytes. Biosci. Rep. 12: 381-386, 1992. BUTTGEREIT, F., M. MULLER, AND S. M. RAPOPORT. Quantifica- tion of ATP-producing and consuming processes in quiescent pig spleen lymphocytes. Biochem. Int. 24: 59-67, 1991. CAMPBELL, S. L., K. A. JONES, AND R. G. SHULMAN. In vivo 31P NMR magnetisation transfer measurements of phosphate exchange reactions in the yeast S. cerevisiae. FEBS Lett. 193: 189- 193, 1985. CHAFFEE, R. R. J., AND J. C. ROBERTS. Temperature acclimation in birds and marnmals. Annu. Rev. Physiol. 33: 155-202, 1971. CHALLONER, D. R. Respiration in myocardium. Nature 217: 78- 79, 1968. CHANCE, B., H. SIES, AND A. BOVERIS. Hydroperoxide metabolism in mammalian organs. Physiol. Rev. 59: 527-605, 1979. CHINET, A. Control of respiration in skeletal muscle at rest. Expe- rientia 46: 1194-1196, 1990. CHINET, A., T. CLAUSEN, AND L. GIRARDIER. Microcalorimetric determination of energy expenditure due to active Na/K transport in the soleus muscle and brown adipose tissue of the rat. J. Physiol. (Lond.) 265: 43-61, 1977.

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79

80

81

82.

83.

84.

DRABKIN, D. L. The distribution of the chromoproteins, haemoglo- bin, myoglobin and cytochrome c in the tissues of different species

and the relationship of the total content of each chromophore to body mass. J. Biol. Chem. 182: 317-333, 1950. ELSE, P. L., AND A. J. HULBERT. An allometric comparison of the mitochondria of mammalian and reptilian tissues: the implications for the evolution of endothermy. J. Comp. Physiol. 156: 3-11, 1985. ELSE, P. L., AND A. J. HULBERT. Mammals: an allometric study of

metabolism at the tissue and mitochondrial level. Am. J. Physiol. 248 (Regulatory Integrative Comp. Physiol. 17): R415-R421, 1985. ELSE, P. L., AND A. J. HULBERT. Evolution of mammalian endothermic metabolism: “leaky” membranes as a source of heat. Am. J. Physiol. 253 (Regulatory Integrative Comp. Physiol. 22): Rl- R7, 1987.

ERECINSKA, M., AND I. A. SILVER. ATP and brain function. J. Cereb. Blood Flow Metab. 9: 2-19, 1989. ERECINSKA, M., D. F. WILSON, AND K. NISHIKI. Homeostatic regulation of regulation of cellular energy metabolism: experimental characterisation in vivo and fit to a model. Am. J. Physiol. 234 (Cell Physiol. 3): CBZ-C89, 1978. FELL, D. A., AND S. THOMAS. Physiological control of metabolic flux: the requirement for multisite modulation. Biochem. J. 311:

35-39, 1995. FIELD, J., H. S. BELDING, AND A. R. MARTIN. An analysis of the relation between basal metabolism and summated tissue respiration in the rat. J. Cell. Comp. Physiol. 14: 143-155, 1939. FOLKE, B., AND E. NEIL. Circulation. New York: Oxford Univ. Press, 1971. FOSTER, D. 0. Quantitative role of brown adipose tissue in thermogenesis. In: Brown Adipose Tissue, edited by P. Trayhurn and D. G.

Nicholls. London: Arnold, 1986, p. 31-52. FOSTER, D. O., AND M. L. FRYDMAN. Nonshivering thermogenesis in the rat. II. Measurements of blood flow using microspheres point to brown adipose tissue as the dominant site of the calorigenesis induced by noradrenaline. Can. J. Physiol. Pharmacol. 56: llO- 112, 1977. FOSTER, D. O., AND M. L. FRYDMAN. Tissue distribution of cold- induced thermogenesis in conscious warm and cold-acclimated rats re-evaluated from changes in tissue blood flow: the dominant role of brown adipose tissue in the replacement of shivering by nonshivering thermogenesis. Can. J. Physiol. Pharmacol. 57: 257-270, 1979.

FREEMAN, D., S. BARTLETT, G. K. RADDA, AND B. ROSS. Energet- its of sodium transport in the kidney: saturation transfer 31P NMR. Biochim. Biophys. Acta 762: 325-336, 1984. GALE, C. C. Neuroendocrine aspects of therrnoregulation. Annu. Rev. Physiol. 35: 391-430, 1973.

GARBY, L., AND 0. LAMMERT. Estimation of the sources of the between-subjects variation in energy expenditure. Eur. J. Clin. Nutr. 48: 303-304, 1994. GARBY, L., AND 0. LAMMERT. Between-subjects variation in energy expenditure: estimation of the effect of variation in organ size.

Eur. J. Clin. Nutr. 48: 376-378, 1994. GIBBS, C. L., AND C. J. BARCLAY. Cardiac efficiency. Cardiovasc. Res. 30: 627-634, 1995. GILL, M., J. FRANCE, M. SUMMERS, B. W. MCBRIDE, AND L. P.

MILLIGAN. Simulation of the energy costs associated with protein turnover and Na+, K’ transport in growing lambs. J. Nutr. 119: 1287- 1299, 1989. GNAIGER, E., AND R. B. KEMP. Anaerobic metabolism in aerobic

mammalian cells: information from the ratio of calorimetric heat flux and respirometric oxygen flux. Biochim. Biophys. Acta 1016: 328-332, 1990. GOLDBERG, N. D., T. F. WALSETH, S. J. EIDE, T. P. KRICK, B. L.

KUCHN, AND J. E. GANDER. Cyclic AMP metabolism in intact platelets determined by “0 incorporation into adenine nucleotide alpha- phosphoryls. Adv. Cyclic Nucleotide Protein Phosphorylation Res. 16: 363-379, 1984. GOLDBETER, A., AND D. E. KOSHLAND. Energy expenditure in

the control of biochemical systems by covalent modification. J. Biol. Churn. 262: 4460-4471, 1987. GRANDE. F. Energv expenditure of organs and tissues. In: Assess-


ment of Energy Metabolism in Health and Disease, edited by S. Calvert. Columbus, OH: Ross Laboratories, 1980, p. 88-92.

86. GREGORY, R. B., AND M. G. BERRY. The characterisation of perfluorosuccinate as an inhibitor of gluconeogenesis in isolated rat hepatocytes. Biochem. Pharmacol. 38: 2867-2872, 1989.

87. GROEN, A. K., R. J. A. WANDERS, H. V. WESTERHOFF, R. VAN DER MEER, AND J. M. TAGER. Quantification of the contribution of various steps to the control of mitochondrial respiration. J. Biol. Chem. 257: 2754-2757, 1982.

88. GULLANS, S. R., P. C. BRAZY, V. W. DENNIS, AND L. J. MANDEL. Interactions between gluconeogenesis and sodium transport in rabbit proximal tubule. Am. J. Physiol. 246 (Renal Fluid Electrolyte Physiol. 15): F859-F869, 1984.

89. GUNTER, T. E., AND B. D. JENSEN. The efficiencies of the component steps of oxidative phosphorylation. I. A simple steady state model. Arch. Biochem. Biophys. 248: 289-304, 1986.

90. GUPTA, R. K., AND R. D. MOORE. 31P NMR studies of intracellular Mg”’ in intact frog skeletal muscle. J. Biol. Chem. 255: 3987-3993, 1980.

91. GUSTAFSSON, R., J. R. TATA, 0. LINDBERG, AND L. ERNSTER. The relationship between the structure and activity of rat skeletal muscle mitochondria after thyroidectomy and thyroid hormone treatment. J. Cell Biol. 26: 555-578, 1965.

92. HAFNER, R. P., AND M. D. BRAND. Effect of the proton motive force on the relative proton stoichiometries of the mitochondrial proton pumps. Biochem. J. 275: 75-80, 1991.

93. HAFNER, R. P., G. C. BROWN, AND M. D. BRAND. Analysis of the control of respiration rate, phosphorylation rate, proton leak rate and proton motive force in isolated mitochondria using the top- down approach of metabolic control theory. Eur. J. Biochem. 188: 313-319, 1990.

94. HALL, S. E. H., R. GOEBEL, I. BARNES, G. HETENYI, AND M. BER- MAN. The turnover and conversion to glucose of alanine in new- born and grown dogs. Federation Proc. 36: 239-244, 1977.

95. HALLIWELL, B. Reactive oxygen species and the central nervous system. In: Free Radicals in the Brain, edited by L. Packer, L. Prilipko, and Y. Christen. Berlin: Springer-Verlag, 1992, p. 21-40.

96. HALLIWELL, B., AND J. M. C. GUTTERIDGE. Free Radicals in Biol- ogy and Medicine. Oxford, UK: Clarendon, 1989.

97. HARPER, M.-E., AND M. D. BRAND. The quantitative contributions of mitochondrial proton leak and ATP turnover reactions to the changed respiration rates of hepatocytes isolated from rats of different thyroid status. J. Biol. Chem. 268: 14850-14860, 1993.

98. HARRIS, S. I., R. S. BALABAN, L. BARRETT, AND L. J. MANDEL. Mitochondrial respiratory capacity and Na+- and K’-dependent adenosine triphosphatase-mediated ion transport in the intact renal cell. J. Biol. Chem. 256: 10319-10328, 1981.

99. HASSELBACH, W., AND H. OETLIKER. Energetics and electrogeni- city of sarcoplasmic reticulum calcium pump. Annu. Rev. Physiol. 45: 325-339, 1983.

100. HAUSSINGER, D., W. GEROCK, AND H. SIES. Hepatic role in pH regulation: role of the intercellular glutamine cycle. Trends Bio- them. Sci. 9: 300-302, 1984.

101. HEINEMANN, F. W., AND R. S. BALABAN. Control of mitochondrial respiration in the heart in vivo. Annu. Rev. Physiol. 52: 523-542, 1990.

102. HERSHKO, A., AND A. CIECHANOVER. The ubiquitin system for protein degradation. Annu. Rev. Biochem. 61: 761-807, 1992.

103. HERVEY, G. R., AND G. TOBIN. Luxuskonsumption, diet induced thermogenesis and brown fat. Clin. Sci. 64: 7- 18, 1983.

104. HIMMS-HAGEN, J. Cellular thermogenesis. Annu. Rev. Physiol. 38: 315-351, 1976.

105. HIMMS-HAGEN, J. Brown adipose tissue metabolism and thermogenesis. Annu. Rev. Nutr. 5: 69-94, 1985.

106. HINKLE, P. C., M. A. KUMAR, A. RESETAR, AND D. L. HARRIS. Mechanistic stoichiometry of mitochondrial oxidative-phosphorylation. Biochemistry 30: 3576-3582, 1991.

107. HODGKIN, A. L., AND P. HOROWICZ. The influence of potassium and chloride ions on the membrane potential of single muscle fi- bres. J. Physiol. (Land.) 148: 127-160, 1959.

108. HOEK, J. B., D. G. NICHOLLS, AND J. R. WILLIAMSON. Determina- tion of the mitochondrial protonmotive force in isolated hepatocytes. J. Biol. Chem. 255: 1458-1464, 1980.

109. HOMSHER, E. Muscle enthalpy production and its relationship to actinomyosin ATPase. Annu. Rev. Physiol. 49: 672-690, 1987.

110. HOMSHER, E., AND C. J. KEAN. Skeletal muscle energetics and metabolism. Annu. Rev. Physiol. 40: 93-131, 1978.

111. HULBERT, A. J. The evolution of energy metabolism in mammals. In: Comparative Physiology: Primitive Mammals, edited by K. Schmidt-Nielsen, L. Bolis, and C. R. Taylor. Cambridge, UK Carn- bridge Univ. Press, 1980, p. 129-139.

112. HULBERT, A. J. A comparative study of thyroid function in reptiles

and mammals. In: The Endocrine System and the Environment, edited by B. K. Follit, S. Ishii, and A. Chandola. Tokyo: Japan Sci. Sot. Press, 1985, p. 105-115.

113. ILES, R. A., A. N. STEVENS, G. R. GRIFFITHS, AND P. G. MORRIS. Phosphorylation status of liver by 31P-NMR spectroscopy and its implications for metabolic control. Biochem. J. 229: 141- 151, 1985.

114. JAMES, W. P. T., AND E. C. SCHONFIELD. Human Energy Require- ments. New York: Oxford Univ. Press, 1990.

115. JANSKY, L. Total cy-tochrome oxidase activity and its relation to

basal and maximal metabolism. Nature 189: 921-922, 1961. 116. JANSKY, L. Adaptability of heat production mechanisms in ho-

meotherms. Acta Univ. Carol. Biol. 1: 1-91, 1965. 117. JANSKY, L., AND J. S. HART. Participation of skeletal muscle and

kidney during nonshivering thermogenesis in cold-acclimated rats. Can. J. Biochem. Physiol. 41: 953-964, 1963.

118. JENSEN, B. D., K. K. GUNTER, AND T. E. GUNTER. The efficiencies of the component steps of oxidative phosphorylation. II. Experi- mental determination of the efficiencies and examination of the equivalence of membrane potential and pH gradient in phosphorylation. Arch. Biochem. Biophys. 248: 305-323, 1986.

119. JUNGAS, R. L., M. L. HALPERIN, AND J. T. BROSMAN. Quantitative analysis of amino acid oxidation and related gluconeogenesis in humans. Physiol. Rev. 72: 419-448, 1992.

120. KACSER, H., AND J. A. BURNS. The control of flux. Symp. Sot. Exp. Biol. 32: 65-104, 1973.

121. KACSER, H., AND J. A. BURNS. Molecular democracy: who shares the control? Biochem. Sot. Trans. 7: 1149- 1160, 1979.

122. KAUPPINEN, R. A. Proton electrochemical potential of the inner mitochondrial membrane in isolated perfused rat hearts as measured by exogenous probes. Biochim. Biophys. Acta 725: 131- 137, 1983.

123. KAUPPINEN, R. A., J. K. HILTUNEN, AND I. E. HASSINEN. Mito- chondrial membrane potential, transmembrane difference in NAD+ redox potential and the equilibrium of the glutamate-aspartate translocase in the isolated perfused rat heart. Biochim. Biophys. Acta 725: 425-433, 1983.

124. KELLY, J. M., AND B. W. MCBRIDE. The sodium pump and other mechanisms of thermogenesis in selected tissues. Proc. Nutr. Sot. 49: 203-215, 1990.

125. KESSELER, A., AND M. D. BRAND. The mechanism of stimulation of state 4 respiration by cadmium in potato tuber (Solanum tubero- sum) mitochondria. Plant Physiol. Biochem. 33: 519-528, 1995.

126. KESSELER, A. K., P. DIOLEZ, K. BRINKMANN, AND M. D. BRAND. Characterisation of the control of respiration in potato tuber mitochondria using the top-down approach of metabolic control analysis. Eur. J. Biochem. 210: 775-784, 1992.

127. KINGSLEY-HICKMAN, P. B., E. Y. SAKO, P. MOHANAKRISHNAN,

P. M. L. ROBITAILLE, A. H. L. FROM, J. E. FOKER, AND K. UGUR- BIL. 31P NMR studies of ATP synthesis and hydrolysis kinetics in the intact myocardium. Biochemistry 26: 7501-7510, 1987.

128. KREBS, H. A. Body size and tissue respiration. Biochim. Biophys.

Acta 4: 249-269, 1950. 129. KREBS, H. A., AND H. L. KORNBERG. Energy Transformations in

Living Matter. Berlin: Springer-Verlag, 1957. 130. KUNKEL, H. O., J. F. SPALDING, G. FRANCISCIS, AND M. F. FU-

TRELL. Cytochrome oxidase activity and body weight in rats and three species of large animals. Am. J. Physiol. 186: 203-206, 1956.

131. KUNZ, W., F. N. GELLERICH, L. SCHILD, AND P. SCHONFELD. Kinetic limitations in the overall reaction of mitochondrial oxida-

tive phosphorylation. FEBS Lett. 233: 17-21, 1988. 132. KUSAKA, M., AND M. UI. Tracer kinetic analysis of Cori cycle activ-

ity in the rat: effect of feeding. Am. J. Physiol. 232 (Endocrinol. Metab. Gastrointest. Physiol. 1): E136-E144, 1977.

133. KUSHMERICK, M. J. Energetics of muscle contraction. In: Hand-


boolc of Physiology. Skeletal Muscle. Bethesda, MD: Am. Physiol. Sot., 1983, sect. 10, chapt. 7, p. 189-236.

134. LAKIN-THOMAS, P. L., AND M. D. BRAND. Stimulation of respiration by mitogens in rat thymocytes is independent of mitochondrial calcium. Biochem. J. 256: 167-173, 1988.

135. LAMBERTSON, C. J. Elements of normal renal function. In: Medi- cal Physiology (11th ed.), edited by P. Bard. St. Louis, MO: Mosby, 1961.

136. LARDEUX, B. R., AND G. E. MORTIMORE. Amino acid and hormonal control of macromolecular turnover in perfused rat liver: evidence for selective autophagy. J. Biol. Chem. 262: 14514- 14519, 1987.

137. LEMASTERS, J. J. The ATP-to-oxygen stoichiometries oxidative phosphorylation by rat liver mitochondria. J. Biol. Chem. 259: 13123- 13130, 1984.

138. LETELLIER, T., M. MALGAT, AND J.-P. MAZAT. Control of oxidative phosphorylation in rat muscle mitochondria: implications for mitochondrial myopathies. Biochim. Biophys. Acta 1141: 58-64, 1993.

139. LOSCHEN, G., L. FLOHE, AND B. CHANCE. Respiratory-chain- linked Hz02 production in pigeon heart mitochondria. FEBS Lett. 18: 261-264, 1971.

140. LOUD, A. D. A quantitative stereological description of the ultra- structure of normal rat liver parenchymal cells. J. Cell Biol. 37: 27-46, 1968.

141. MAGNUSSON, I., D. J. ROTHMAN, B. JUCKER, G. W. CLINE, R. G. SHULMAN, AND G. I. SHULMAN. Liver glycogen turnover in fed and fasted animals. Am. J. Physiol. 266 (Endocrinol. Metab. 29): E796-E803, 1994.

142. MARTIN, A. W., AND F. A. FUHRMAN. The relationship between summated tissue respiration and metabolic rate in the mouse and dog. Physiol. 2001. 26: 18-34, 1955.

143. MARTINEAU, A., L. LECAVALIER, P. FALARDEAU, AND J.-L. CHI- ASSON. Simultaneous determination of glucose turnover, alanine turnover and gluconeogenesis in humans using a double stable- isotope-labelled tracer infusion and gas chromatography-mass spectrometry analysis. Anal. Biochem. 151: 495-503, 1985.

144. McARDLE, W. D., K. I. KATCH, AND V. L. KATCH. Exercise Physiol- ogy (2nd ed.). Philadelphia, PA: Lea & Febiger, 1986, p. 149.

145. McGILVERY, R. W. Biochemistry: A Functional Approach. Phila- delphia, PA: Saunders, 1979.

146. MILLER, A. T., AND C. S. BLYTH. Lean body mass as a metabolic reference standard. J. Appl. Physiol. 5: 311-316, 1953.

147. MILLIGAN, L. P., AND M. SUMMERS. The biological basis of maintenance and its relevance to assessing responses to nutrients. Proc. Nutr. Sot. 45: 185-193, 1986.

148. MULLER, M., W. SIEMS, F. BUTTGEREIT, R. DUMDEY, AND S. M. RAPOPORT. Quantification of ATP-producing and consuming processes in Ehrlich ascites tumour cells. Eur. J. Biochem. 161: 701- 705, 1986.

149. MURPHY, M. P. Slip and leak in mitochondrial oxidative phosphorylation. Biochim. Biophys. Acta 977: 123-141, 1989.

150. MURPHY, M. P., G. C. BROWN, AND M. D. BRAND. Thermodynamic limits to the stoichiometry of H’ pumping by mitochondrial cytochrome oxidase. Biochem. J. 252: 473-479, 1985.

151. NAGY, K. A. Field metabolic rate and food requirement scaling in mammals and birds. Ecol. Monogr. 57: 112- 128, 1987.

152. NEWSHOLME, E. A. Sounding board: a possible basis for the control of body weight. N. Engl. J. Med. 302: 400-405, 1980.

153. NEWSHOLME, E. A., AND A. H. UNDERWOOD. The control of glycolysis and gluconeogenesis in the kidney cortex. Biochem. J. 99: 24-26, 1966.

154. NICHOLLS, D. G. The influence of respiration and ATP hydrolysis on the proton-electrochemical gradient across the inner membrane of rat-liver mitochondria as determined by ion distribution. Eur. J. Biochem. 50: 305-315, 1974.

155. NICHOLLS, D. G., AND R. M. LOCKE. Thermogenic mechanisms in brown fat. Physiol. Rev. 64: l-64, 1984.

156. NOBES, C. D., G. C. BROWN, P. N. OLIVE, AND M. D. BRAND. Non- ohmic conductance of mitochondrial inner membrane of hepatocytes. J. Biol. Chem. 265: 12903-12909, 1990.

157. OATES, J., D. PAPAHDJOPOULOS, AND A. LOYTER. Application of induced fusion to problems in biology. Trends Pharmacol. Sci. 2: 222-224, 1982.

158. PARK, E. A., AND H. E. MORGAN. Energy dependence of RNA degradation in rabbit reticulocytes. Am. J. Physiol. 247 (Cell Physiol. 16): C390-C395, 1984.

159. PAUL, R. J. Smooth muscle energetics. Annu. Rev. Physiol. 51: 331-349, 1989.

160. PEARSE, B. M. F., AND M. S. BRETSCHER. Membrane recycling by coated vesicles. Annu. Rev. Biochem. 50: 85- 101, 1981.

161. PFALLER, W., AND M. RITTINGER. Quantitative morphology of the rat kidney. Int. J. Biochem. 12: 17-22, 1980.

162. PODOLSKY, R. J., AND M. F. MORALES. The enthalpy change of adenosine triphosphate hydrolysis. J. Biol. Chem. 218: 945-956, 1956.

163. POE, M., H. GUTFREUND, AND R. W. ESTABROOK. Kinetic studies of temperature changes and oxygen uptake in a differential calorim- eter in the heart. Arch. Biochem. Biophys. 122: 204-211, 1967.

164. PORTER, R. K., AND M. D. BRAND. Body mass dependence of H’ leak in mitochondria and its relevance to metabolic rate. Nature Lond. 362: 628-630, 1993.

165. PORTER, R. K., AND M. D. BRAND. Cellular oxygen consumption depends on body mass. Am. J. Physiol. 269 (Regulatory Integrative Comp. Physiol. 38): R226-R228, 1995.

166. PORTER, R. K., AND M. D. BRAND. Mitochondrial proton conductance and H’/O ratio are independent of electron transport rate in isolated hepatocytes. Biochem. J. 310: 379-382, 1995.

167. PORTER, R. K., AND M. D. BRAND. Causes of the differences in respiration rate of hepatocytes from mammals of different body mass. Am. J. Physiol. 269 (Regulatory Integrative Comp. Physiol. 38): R1213-R1224, 1995.

168. PRUSINER, S., AND M. POE. Thermodynamic considerations of mammalian thermogenesis. Nature 220: 235-237, 1968.

169. RABKIN, M., AND J. J. BLUM. Quantitative analysis of intermediary metabolism in hepatocytes incubated in the presence and absence of glucagon with a substrate mixture containing glucose, ribose, fructose, alanine and acetate. Biochem. J. 225: 761-786, 1985.

170. RADZIUK, J. Developments in the tracer measurement of gluconeogenesis in vivo: an overview. Federation Proc. 41: 88-90, 1982.

171. REEDS, P. J., M. F. FULLER, AND B. A. NICHOLSON. Metabolic basis of energy expenditure with particular reference to protein. In: Recent Advances in Substrate and Energy Metabolism, edited by J. S. Garrow and W. Halliday. London: Libbey, 1985, p. 46-57.

172. REEDS, P. J., AND C. I. HARRIS. Protein turnover in animals: man in his context. In: Nitrogen Metabolism in Man, edited by J. C. Waterlow and J. M. L. Stephen. Essex, UK Appl. Sci., 1981, p. 391- 408.

173. RENNIE, M. J., AND J. 0. HOLLOSZY. Inhibition of glucose uptake and glycogenolysis by availability of oleate in well oxygenated perfused skeletal muscle. Biochem. J. 168: 161-170, 1977.

174. REYNAFARJE, B., A. ALEXANDRE, P. DAVIES, AND A. L. LEH- NINGER. Proton translocation stoichiometry of cytochrome oxidase. Proc. Natl. Acad. Sci. USA 79: 7218-7222, 1982.

175. ROHRACHER, H. Permanente rhythmische Mikrobewegungen des Warmbluter-Organismus (“Mikrovibration”). Naturwiss. 49: 145- 147, 1962.

176. ROLFE, D. F. S., AND M. D. BRAND. The contribution of mitochondrial proton leak to the rate of respiration in rat skeletal muscle and to SMR. Am. J. Physiol. 271 (Cell Physiol. 40): C1380-C1389, 1996.

177. ROLFE, D. F. S., AND M. D. BRAND. Proton leak and control of oxidative phosphorylation in perfused, resting rat skeletal muscle. Biochim. Biophys. Acta 1276: 45-50, 1996.

178. ROLFE, D. F. S., A. J. HULBERT, AND M. D. BRAND. Characteris- tics of mitochondrial proton leak and control of oxidative phosphorylation in the major oxygen consuming tissues of the rat. Bio- chim. Biophys. Acta 1188: 405-416, 1994.

179. ROTHWELL, N. J., AND M. J. STOCK. A role for brown adipose tissue in diet-induced thermogenesis. Nature 281: 31-35, 1979.

180. ROTHWELL, N. J., AND M. J. STOCK. Diet-induced thermogenesis. In: Mammalian Therrnogenesis, edited by L. Girardier and M. J. Stock. London: Chapman & Hall, 1983, p. 208-233.

181. RUDERMAN, N. B., C. R. S. HOUGHTON, AND R. HEMS. Evaluation

of the isolated perfused rat hindquarter for the study of muscle metabolism. Biochem. J. 124: 639-651, 1971.

182. SCHMIDT, H., W. SIEMS, M. MULLER, R. DUMDEY, M. JAKSTADT,


183.

184.

185.

186.

187.

188.

189.

190.

191.

192.

193

194

195.

196.

197.

198.

199.

200. 201.

202.

AND S. M. RAPOPORT. Balancing of mitochondrial and glycolytic ATP production and of the ATP-consuming processes of Ehrlich mouse ascites tumour cells in a high phosphate medium. Biochem. Int. 19: 985-992, 1989. SCHMIDT-NIELSEN, K. Scaling: Why Is Animal Size So Im- portant? Cambridge, UK: Cambridge Univ. Press, 1984. SCHMIDT-NIELSEN, K. Animal Physiology: Adaptation and Envi- ronment. Cambridge, UK: Cambridge Univ. Press, 1990, p. 193. SCHOCH, G., G. SANDER, E. FUCHS, 0. JOHREN, G. HELLER- SCHOCH, AND H. TOPP. The whole body degradation rates of cytoplasmic tRNA, rRNA and mRNA are correlated with basal metabolic rate in mammals of various sizes. Biol. Chern. Hoppe-Seyler 372: 749, 1991. SCHOCH, G., H. TOPP, A. HELD, G. HELLER-SCHOCH, A. BAL- LAUFF, F. MANZ, AND G. SANDER. Interrelation between whole- body turnover rates of RNA and protein. Eur. J. Clin. Nutr. 44: 647-658, 1990. SCHOLZ, R., AND T. BUCHER. Hemoglobin-free perfusion of rat liver. In: Control of Energy Metabolism, edited by B. Chance. New York: Academic, 1965, p. 393-414. SCHRODINGER, E. What Is Life? Cambridge, UK: Cambridge Univ. Press, 1944. SCHUTZ, Y. The basis of direct and indirect calorimetry and their potentials. Diabetes Metab. Rev. 11: 383-408, 1995. SELLERS, E. A., J. W. SCOTT, AND N. THOMAS. Electrical activity of skeletal muscle of normal and acclimatized rats on exposure to cold. Am. J. Physiol. 177: 372-376, 1954. SIEMS, W., W. DUBIEL, R. DUMDEY, M. MULLER, AND S. M. RAPO- PORT. Accounting for the ATP-consuming processes in rabbit reticulocytes. Eur. J. Biochem. 139: lOl- 107, 1984. SIEMS, W., M. MULLER, R. DUMDEY, H.-G. HOLZHUTTER, J. RATHMANN, AND S. M. RAPOPORT. Quantification of pathways of glucose utilisation and balance of energy metabolism of rabbit reticulocytes. Eur. J. Biochem. 124: 567-576, 1982. SIESJO, B. K. Brain Energy Metabolism. Chichester, UK: Wiley, 1978. SILVA, J. E. Hormonal control of thermogenesis and energy dissipation. Trends Endocrinol. Metab. 4: 25-32, 1993.

SKULACHEV, V. P. Nonphosphorylating respiration as a mechanism minimising the formation of reactive oxygen species in the cell. Mol. Biol. 29: 709-715, 1995. SMITH, R. E. Quantitative relations between liver mitochondrial metabolism and total body weight in mammals. Ann. NYAcad. Sci. 62: 403-422, 1956. SOLTOFF, S. P. ATP and the regulation of renal cell function. Annu. Rev. Physiol. 48: 9-31, 1986.

SORGATO, M. C., F. GALIAZZO, L. PANATO, AND S. J. FERGUSON. Estimation of H+-translocation stoichiometry of mitochondrial ATPase. Biochim. Biophys. Acta 682: 184-188, 1982. SPRIET, L. L., C. G. MATSOS, S. J. PETERS, G. J. F. HEIGEN- HAUSER, AND N. L. JONES. Muscle metabolism and performance in perfused rat hindquarter during heavy exercise. Am. J. Physiol. 248 (Cell Physiol. 17): Clog-C118, 1985. SUGA, H. Ventricular energetics. Physiol. Rev. 70: 247-277, 1990.

SWAMINATHAN, R., E. L. P. CHAN, L. Y. SIN, S. K. F. NG, AND

A. Y. S. CHAN. The effects of ouabain on metabolic rate in guinea pigs: estimation of energy costs of sodium pump activity. Br. J. Nutr. 61: 467-473, 1989. TANFORD, C. Equilibrium state of ATP driven ion pumps in relation to physiological ion concentration gradients. J. Gen. Physiol.

77: 223-229, 1981.

203.

204.

205.

206.

207.

208.

209.

211.

212.

213

215.

216.

217.

218.

219.

220.

221.

TATA, J. R. Basal metabolic rate and thyroid hormones. In: Ad- vances in Metabolic Disorders, edited by R. Levine and R. Luft. New York: Academic, 1964, vol. 1, p. 153-189. TAYLOR, C. R. Structural and functional limits to oxidative metabolism: insights from scaling. Annu. Rev. Physiol. 49: 135-146, 1987. TURRENS, J. F., A. ALEXANDRE, AND A. L. LEHNINGER. Ubisemi- quinone is the electron donor for superoxide formation by complex III of heart mitochondria. Arch. Biochem. Biophys. 237: 408-414, 1985. UGURBIL, K., P. B. KINGSLEY-HICKMAN, E. Y. SAKO, S. ZIMMER, AND P. MOHANAKRISHNAN. 31P NMR studies of the kinetics and regulation of oxidative phosphorylation in the intact myocardium. Ann. NYAcad. Sci. 508: 265-286, 1987. VERCESI, A., B. REYNAFARJE, AND A. L. LEHNINGER. Stoichiom- etry of H+ ejection and Ca2’ uptake coupled to electron transport in rat heart mitochondria. J. Biol. Chem. 253: 6379-6395, 1978. VERHOEVEN, A. J. M., G. GORTER, M. E. MOMMERSTEEG, AND

J. W. M. AKKERMAN. The energetics of early platelet responses: energy-consumption during shape change and aggregation with special reference to protein-phosphorylation and the polyphospho- inositide cycle. Biochem. J. 228: 451-462, 1985. VON STOCKAR, U., L. GUSTAFSSON, C. LARSSON, I. MARISON, P. TISSOT, AND E. GNAIGER. Thermodynamic considerations in constructing energy balances for cellular growth. Biochim. Bio- phys. Acta 1183: 221-240, 1993. WATERLOW, J. C. Protein turnover with special reference to man. Q. J. Exp. Physiol. 69: 409-438, 1984. WATERLOW, J. C., P. J. GARLICK, AND D. J. MILLWARD. Protein Turnover in Mammalian Tissues and the Whole Body. New York: North-Holland, 1978. WATERLOW, J. C., AND D. J. MILLWARD. Energy cost of turnover of protein and other cellular constituents. In: Proceedings of the Tenth Conference of the European Society for Comparative Physi- ology and Biochemistry Innsbruck 1989. Stuttgart: Georg Thieme Verlag, 1989, p. 277-282. WEBSTER, A. J. F. Prediction of energy requirements for growth in beef cattle. World Rev. Nutr. Dietetics 30: 189-226, 1978. WHITTAM, R. Active cation transport as a pacemaker of respiration. Nature 191: 603-604, 1961. WHITTAM, R. The dependence of the respiration of brain cortex on active cation transport. Biochem. J. 82: 205-212, 1962. WIESER, W. Energy allocation by addition and by compensation: an old principle revisited. In: Energy Transformations in Cells and Organisms, edited by W. Wieser and E. Gnaiger. Stuttgart: Georg Thieme Verlag, 1989, p. 98-105. WISNIEWSKI, E., W. S. KUNZ, AND F. N. GELLERICH. Phosphate affects the distribution of flux control among the enzymes of oxidative phosphorylation in rat skeletal muscle mitochondria. J. Biol. Chem. 268: 9343-9346, 1993. WOELDERS, H., W. J. VAN DER ZANDE, A.-M. COLEN, R. J. A. WANDERS, AND K. VAN DAM. The phosphate potential maintained by mitochondria in state 4 is proportional to the protonmotive force. FEBS Lett. 179: 278-282, 1985. WOLEDGE, R. C., N. A. CURTIN, AND E. HOMSHER. Energetic As- pects of Muscle Contraction. London: Academic, 1985, p. 267. WONG-RILEY, M. T. T. Cytochrome oxidase: an endogenous metabolic marker for neuronal activity. Trends Neurosci. 12: 94-101,

1989. ZOLKIEWSKA, A., B. ZABLOCKA, J. DUSZYNSKI, AND L. WOJTC- ZAK. Resting state respiration of mitochondria: reappraisal of the role of passive ion fluxes. Arch. Biochem. Biophys. 275: 580-590, 1989.

Documents

Cellular Energy Utilization and Molecular Origin of Standard … · 2017-10-17 · PHYSIOLOGICAL REVIEWS Vol. 77, No. 3, July 1997 Printed in U.S.A. Cellular Energy Utilization and