J.P. FLATT*
* Department of Biochemistry and Molecular Biology, University of Massachusetts Medical Center, Worcester, MA 01655, U.S.A.
1. Introduction
2. Nitrogen balance and amino acid oxidation
3. Amino acid oxidation during periods of positive or negative energy balance
4. Interactions between energy and protein metabolism
5. Amino acid degradation and gluconeogenesis
6. Summary
References
Proteins, but not
carbohydrates and fats, contain substantial amounts of nitrogen
(N). By measuring nitrogen intake and excretion, it has therefore
been possible to monitor nitrogen (and hence protein) balances
and to assess amino acid (AA) oxidation rates, a task facilitated
by the fact that most of the N liberated by amino acid
degradation in mammals is eliminated in the form of urinary urea.
Thanks to such measurements, it has long been recognized that the
organism strives to maintain approximate nitrogen balance. In the
absence of disease and as long as protein intake is above a
certain minimum, nitrogen balances are indeed quite effectively
maintained by adjustment of AA degradation to protein intake,
regardless of the relative proportions of carbohydrate and fat
provided by the diet to cover the body's energy needs. Protein
metabolism provides an example of a situation where problems of
considerable practical importance can be dealt with so well, that
one often forgets that the mechanisms by which the oxidation of
the various amino acids is coordinated and adjusted to match
protein intake are still poorly understood.
The manner in which
nitrogen balances vary as a function of protein-N intake in
adults is illustrated in the upper panel of Figure 1.
Three typical conditions are represented: (A) When common foods
are consumed, so that dietary protein is accompanied by
carbohydrates and fats to the tune of about 150 kcal/g protein-N
(i.e., 17% of dietary energy as protein). When no food is
consumed, urinary nitrogen excretion is shown here to be about 10
g/d; this amount varies somewhat, depending in particular on the
length of the fasting period. (B) When foods high in
carbohydrate, but containing essentially no protein, are consumed
in amounts sufficient to cover daily energy needs (i.e., - 1500
kcal/d), nitrogen losses are reduced to some 5 g N/d. This
corresponds to the minimal or 'obligatory N loss'. Consumption of
increasing amounts of protein in addition to the protein-free
foods causes the nitrogen balance to become less negative. With
an intake of about 12 g of nitrogen/d, the situation is similar
to that where usual foods are consumed in amounts covering daily
needs, N balance being achieved with intakes of about 12 g of
protein-N/d and about 1800 kcal/d. (C) When only protein is
consumed, nitrogen balances are less negative than during total
food deprivation, but ingestion of 12 g of protein-N per day is
not sufficient to achieve N balance. Nitrogen balance can be
approached even while the energy balance remains markedly
negative, but substantially higher protein intakes are required.
This situation is commonly encountered during a 'protein-sparing
modified fast', where, after one week of adaptation, a dose of
1.5 g protein/kg body weight/d is generally sufficient to
maintain N balance (LINDNER and BLACKBURN, 1976). In subjects
affected by trauma or disease, the three curves are all shifted
downward, as these conditions bring about increased protein
breakdown to amino acids as well as a more rapid amino acid
oxidation (KINNEY and ELWYN, 1983). In protein-depleted
individuals, on the other hand, N balances are generally less
negative and they become more markedly positive during
reconvalescence than in well-fed subjects, so that the three
curves shown in the upper panel of Figure 1 are shifted
upward in such conditions (WATERLOW, 1987). However, even then, N
balances do not rise much above a few grams of protein-N retained
per day, regardless of the fact that protein and other nutrients
may be consumed (or infused parenterally) in amounts greatly
exceeding protein and energy requirements (KINNEY and ELWYN,
1983).
Upper panel: Nitrogen balances in adults as a function of protein nitrogen intake: (A) when consuming mixed foods (-), (B) 1500 kcal/d of non-protein energy plus variable amounts of protein (-), or (C) only protein (---).
Lower panel: Nitrogen excretion rates as a function of protein-N intakes, as implied by the nitrogen balances shown in the upper panel. (Nitrogen balance is shown by.....)
Restatement
of these quite familiar N-balance patterns allows us to examine
what they imply about rates of amino acid degradation (lower
panel of Figure 1), of which one is generally less well
aware. The thin dotted line shows the conditions for which N
excretion matches N intake, i.e. when N balance is equal to zero.
It can be seen that on high protein and energy intakes, nitrogen
excretion increases in direct proportion to further increments in
protein intake. On the other hand, amino acid oxidation rates
vary substantially when food intake is restricted, as they are
then markedly influenced by the availability of other fuels,
i.e., glucose, free fatty acids (FFA) and ketone bodies (FLATT
and BLACKBURN, 1976). This reflects the fact that maintenance of
ATP levels takes priority in all cells, so that any available
substrates will be used to regenerate ATP from ADP and phosphate.
Consuming carbohydrate to maintain glucose availability reduces
the need to obtain energy by amino acid oxidation. Ingestion of
some 100 g of carbohydrate per day reduces N excretion by about
half, a phenomenon well known as the 'protein-sparing effect of
dietary carbohydrate' (GAMBLE, 1946). Unfortunately, influx of
exogenous carbohydrates is much less effective in curtailing N
losses in the face of disease, even in high doses (KINNEY and
ELWYN, 1983). During prolonged starvation, mobilization of
endogenous fat reserves can yield enough substrates to meet almost
all of the body's energy needs, thanks to the production of
ketone bodies by the liver which can be used by the brain instead
of glucose (CAHILL, 1970). However, about two weeks of starvation
elapse before circulating ketone body levels rise enough for the
combined availability of FFA, ß-hydroxy-butyrate and
acetoacetate to allow maximal curtailment of amino acid
oxidation. Having learned to recognize the importance of energy
metabolism on the nitrogen balance, much attention has been
focused on the interactions between metabolic fuels, hormone
levels and the protein economy (CAHILL, 1971; FLATT and
BLACKBURN, 1974). However, the concepts which have evolved and
which shape much of our thinking in this area are based primarily
on observations made under conditions of protein and/or energy
deprivation. They are of little help in explaining the control of
amino acid oxidation under conditions of plenty.
Differences between
hypocaloric and hypercaloric situations are summarized
schematically in Figure 2 which shows the regulatory
interactions in the 'metabolic fuel regulatory system' (FLATT and
BLACKBURN, 1976). As suggested by the 'metabolic funnel',
the main pathways of intermediary metabolism have the effect of
channelling various substrates toward the mitochondrial system
where their ultimate conversion to CO2 by the reaction
of the citric acid cycle is accomplished. The first equation
states the fact that overall substrate oxidation is dictated by
energy expenditure, but also that determination of overall
substrate oxidation (e.g., by indirect calorimetry) provides a
measure of total energy expenditure. In hypocaloric situations
(second equation), the extent to which amino acid oxidation
will rise above the minimum obligatory rate is determined
primarily by the extent to which energy needs can be covered by
oxidation of glucose, FFA and KB. In hypercaloric situations (third
equation), amino acid oxidation increases in direct proportion to
further increments in protein intake, regardless of carbohydrate
and fat substrate availability (Figure 1). Many key
features of metabolic control in hypocaloric situations are thus
obviously not applicable in hypercaloric situations. For
instance, while the availability of carbohydrate and high
circulating insulin levels are believed to be particularly
critical in limiting amino acid oxidation (CAHILL, 1971), amino
acid oxidation in hypercaloric situations will proceed as rapidly
as needed to match protein intake, in spite of high glucose
availability and elevated insulin levels. In fact, the increase
in amino acid oxidation encountered following trauma, and to an
even greater extent during periods of sepsis (KINNEY and ELWYN,
1983) occurs in spite of the hyperglycemia and the high insulin
levels which generally prevail under such conditions.
In trying to make sense of
these seeming inconsistencies, it is useful to remember that most
enzymes catalyzing the transformation of substrates along the
main pathways of intermediary metabolism are present in
considerable excess. Such enzymatic 'reserve capacity'
allows the organism to digest and store occasional, unusually
large nutrient influxes, and to handle the exceptional physical
activity demands encountered from time to time. When such demands
remain altered for some time, induction and repression of protein
synthesis bring about appropriate increases or decreases in
enzyme levels and in the muscle mass, so as to assure the
maintenance of an adequate, but not excessive reserve capacity
for the new set of circumstances. The availability of such excess
enzymatic activities is what enables the organism to strive for a
desirable overall integration of substrate oxidation, without
being impeded by lack of catalytic power at some particular steps
in the main metabolic pathways. As a consequence of such
relatively high enzyme levels, many metabolic intermediates are
maintained at near equilibrium conditions, a situation where
substrate influxes and drainage, as well as KM and competition
for coenzymes can markedly influence flux rates. When metabolic
fuel availability is not a factor, variations in amino acid
concentrations can thus play a major role in bringing about
changes in amino acid degradation rates (HARPER et al.,
1984). This may appear to be a rather primitive mechanism,
potentially associated with a good measure of instability, though
this is presumably not too threatening in this instance because
protection against undue depletion is offered by the constant
influx of amino acids liberated by degradation of endogenous
proteins. Such a mechanism readily accounts for the fact that an
influx of dietary amino acids enhances amino acid oxidation in
spite of the anabolic conditions prevailing post-prandially. One
can, of course, also expect that the rate of amino acid oxidation
will be influenced by changes in the rate of protein breakdown,
as well as by the avidity with which amino acids are used for
protein synthesis (KINNEY and ELWYN, 1983; WATERLOW, 1986).
The various pathways of intermediary metabolism have the effect of a 'metabolic funnel', channelling the various intermediates toward common steps leading to terminal oxidation to CO2 in the mitochondria. The captions in the lower part of the figure show that the overall rate of energy expenditure determines total substrate oxidation; that the availability of glucose, free fatty acids (FFA) and ketone bodies (KB) to meet energy demands is an important factor in determining amino acid oxidation rates when food intake is restricted (hypocaloric situations); and that amino acid oxidation in hypercaloric situations is largely determined by protein intake, a condition where the rate of amino acid oxidation influences carbohydrate plus fat oxidation, rather than the reverse.
Substrate level interactions between amino acid, carbohydrate and fat oxidations are bound to occur, notably because pyruvate, fatty acids, ketone bodies and the µ-keto derivatives of the branched-chain amino acids (BCAA) all need to react with Coenzyme A (CoA) to enter the pathways for their oxidative degradation. When glucose is available, insulin facilitates glucose transport and activates pyruvate dehydrogenase (PDH) (DENTON et al., 1987) so that pyruvate can capture most of the CoA made available by oxidation of acetyl-CoA, thereby restraining the degradation of amino acids needing CoA for their oxidation (Figure 3). During prolonged starvation, once starvation ketosis is fully induced, FFA and KB appear together to be almost as effective in preventing the µ-keto derivatives of the BCAA from reacting with CoA as a plentiful supply of glucose. When food and protein intake are limiting, losses and failure to replenish the BCAA pool appear to assume special importance in determining overall amino acid oxidation and N balance (WALSER, 1983). During disease, the supply of carbohydrate is diminished due to spontaneous or forced reduction in food intake. The release of 'catabolic hormones' causes mobilization of fuels as well as insulin resistance. The increases in insulin secretion needed to control blood glucose levels in such situations is not sufficient to counteract the lipolytic effect of catecholamine on FFA mobilization when stress is high (KINNEY and ELWYN, 1983), but when stress is minor, FFA levels may be lower than during equivalent food deprivation in the absence of disease (NEUFELD et al., 1980). However, insulin levels are sufficient to markedly inhibit ketogenesis even in the face of elevated circulating levels of catabolic hormones (WATTERS and WILMORE, 1986). As demonstrated by various observations, ketone bodies have an effect in reducing amino acid oxidation which goes beyond that accomplished by high FFA availability (SHERWYN et al., 1975; FLATT and QUAIL, 1981) (Figure 4). Given the limited availability of glucose and the effect of insulin resistance on the control of PDH and pyruvate conversion to acetyl-CoA, stress plus food deprivation create a situation for increased amino acid oxidation. Indeed, rapid protein-wasting is induced in the absence of disease, if the mobilization of the fat reserves and ketogenesis are inhibited by administration of antilipolytic agents, though these agents have no effect on N excretion in fed animals (TALKE et al., 1973). It is noteworthy that the increase in amino acid degradation which occurs as a result of lack of other fuel leads to a rise in blood glucose levels, indicating that increased amino acid degradation is not driven by the need to provide glucose (TALKE et al., 1973). The effects of sepsis and trauma in inhibiting ketogenesis thus appear to be important factors in explaining protein-wasting (FLATT and BLACKBURN, 1974, WEDGE et al., 1976), and competition for CoA provides a widely applicable concept in helping to understand how rate-setting interactions between intermediates of carbohydrate, fat and protein metabolism can occur at the substrate level.
High free fatty acid (FFA) and ketone body (KB) availability under fasting conditions, or a plentiful supply of pyruvate in the fed state assure that the µ-keto derivatives of the branched-chain amino acids (BCKA) are not rapidly drawn into irreversible degradation by reaction with Coenzyme A (CoA), unless their concentrations are sufficiently high to compete more intensely for CoA.
When changing from situations of high to low protein intake, or vice-versa, a few days are required before N balance is again achieved, during which small amounts of protein are lost or gained. Since these are a prerequisite for the reestablishement of N balance, small changes in protein content appear to play an important role in adjusting amino acid oxidation rates to amino acid intake This can be envisioned to result from the expansion or contraction of rapidly turning-over protein pools, which would be expected to influence the concentration of intracellular amino acids between meals. Coupled with the evidence that amino acid degradation is greatly influenced by their availability when metabolic fuel availability is not a factor (HARPER et al., 1984), the role of rapidly turning-over proteins (perhaps a part of the so-called 'labile protein pool'), may be seen as exerting a function analogous to that of a 'fly wheel' in making amino acid oxidation commensurate with amino acid intake.
Intraperitoneal injection of CC14 (1 mL/kg body weight) to 24-hour fasted rats (-), which does not reduce FFA mobilization, causes marked lowering of ketone body levels and a concomitant increase in urinary-N excretion relative to fasting controls (o--o). (Adapted from FLATT and QUAIL, 1981).
It is often believed that
gluconeogenesis from amino acids is induced during starvation to
provide glucose and that the need to provide glucose is a driving
factor in the increased gluconeogenesis taking place after trauma
and during disease (KINNEY and ELWYN, 1983). This view may or may
not include the belief that gluconeogenesis from amino acids is a
minor process in the fed state. In fact, conversion of the
glucogenic moieties of the degraded amino acids to glucose occurs
even in the fed state. A detailed quantitative analysis of the
energy exchanges associated with the degradation of amino acids
in man (JUNGAS et al., 1992) demonstrates that
gluconeogenesis and export of glucose is essential, because
complete oxidation of the amino acid mixture by the liver would
provide much more ATP than needed by this organ. Gluconeogenesis
from amino acids must thus be regarded as a normal process
associated with amino acid degradation, occurring at higher rates
under conditions of normal food intake than during fasting.
The
activation of pyruvate dehydrogenase by insulin is one of the
mechanisms allowing insulin to promote carbohydrate oxidation
(DENTON et al., 1987). The first irreversible steps in the
degradation of the branched-chain amino acids are catalyzed by µ-ketoacid oxidizing enzyme complexes which
are similar to PDH and can also be activated by insulin (HARPER et
al., 1984). It seems that glucose availability and the
readiness to oxidize pyruvate which are enhanced by insulin
explain the carbohydrate-sparing effect of exogenous carbohydrate
(Figure 3). However, high insulin levels also open the
pathway for the irreversible degradation of the BCAA, notably in
skeletal muscle. Since the BCAA are indispensable amino acids,
their loss by the irreversible oxidation of their µ-keto derivatives condems the other amino
acids to being degraded as well, since they cannot be used for
protein synthesis without enough BCAA.
The concepts which have
evolved to explain the interactions between protein and energy
metabolism are based primarily on observations made under
conditions of protein and/or energy deprivation. Though they
shape much of our thinking, they are not appropriate to explain
the control of amino acid oxidation when food intake is high. The
situation is analogous to the irrelevance of concepts such as
'biological value' or 'net protein utilization' for conditions
where protein and energy intakes are high.
CAHILL, G.F., Jr.:
Starvation in man. N. Engl. J. Med., 282, 668-675 (1970).
CAHILL, G.F., Jr.: Physiology of insulin in man. Diabetes, 20, 785-799 (1971).
DENTON, R.M., McCORMACK, J.G., MIDGLEY, P.G.W., RUTTER, G.A.: Hormonal regulation of fluxes through pyruvate dehydrogenase and the citric acid cycle in mammalian tissues. Biochem. Soc. Symp., 54, 127-143 (1987).
FLATT, J.P., BLACKBURN, G.L.: The metabolic fuel regulatory system: implications for proteinsparing therapies during caloric deprivation and disease. Am. J. Clin. Nutr., 27, 175-187 (1974).
FLATT, J.P., QUAIL, J.M.: Effects of liver damage on ketone-body production and nitrogen balance in starved rats. Biochem. J., 198, 227-230 (1981).
GAMBLE, J.L.: Physiological information gained from studies on the life-raft ration. Harvey Lectures, 42, 247-273 (1946).
HARPER, A.E., MILLER, R.H., BLOCK, K.P.: Branched-chain amino acid metabolism. Annul Rev. Nutr., 4, 409-454 (1984).
JUNGAS, R.L., HALPERIN, M.L., BROSNAN, J.T.: Quantitative analysis of amino acid oxidation and related glucourogenesis in humans. Physiol. Rev., 72, 419-448 (1992).
KINNEY, J.M., ELWYN, D.H.: Protein metabolism and injury. Annul Rev. Nutr., 3, 433-466 (1983).
LINDNER, P.G., BLACKBURN, G.L.: Multidisciplinary approach to obesity utilizing fasting modified by protein-sparing therapy. Obesity/Bariatric Med., 5, 198-213 (1976).
NEUFELD, H.A., PAU, J.G., KAMINSKI, M.V., GEORGE, D.T., JABRLING, P.B., WANNEMACHER, K.W., Jr., BEISEL, W.R.: A probable endocrine basis for the depression of ketone bodies during infection or inflammatory state in rats. Endocrinology, 107, 596-601 (1980).
SHERWYN, R.S., HANDLER, R.G., FELIG, R.: Effect of ketone infusions on amino acid and nitrogen metabolism in man. J. Clin. Invest., 55, 1382-1389 (1975).
TALKE, H., MAIER, K.P., KERSTEN, M., GEROK, W.: Effect of nicotinaminde on carbohydrate metabolism in the rat liver during starvation. Eur. J. Clin. Invest, 3, 467-474 (1973).
WALSER, M.: Nutrition in renal failure. Annul Rev. Nutr., 3, 125-154 (1983).
WATERLOW, J.C.: Metabolic adaption to low intakes of energy and protein. Annul Rev. Nutr., 6, 495-526 (1986).
WATTERS, J.M., WILMORE, D.W.: Role of catabolic hormones in the hypoketonemia of injury. Br. J. Surg., 73, 108-110 (1986).
WEDGE, J.H., DE CAMPOS, R., KERR, A., SMITH, R., FARRELL, R., ILIE, V., WILLIAMSON, D.H.: Branched-chain amino acids, nitrogen excretion and injury in man. Clin. Sci. Molec. Med., 50, 393-399 (1976).