Download Muscle alanine synthesis and hepatic gluconeogenesis

Biochemical Review
Muscle alanine synthesis and hepatic gluconeogenesis
KEITH SNELL
Division of Biochemistry, Department of Biochemistry,
Universiry of Surrey, Guiliford, Surrey GU2 5XH,U.K.
Gluconeogenesis is the metabolic process involving the synthesis de novo of glucose molecules from other carbon precursors
during situations in which there is an enhanced oxidative utilization of body glucose or in which there is a dietary insufficiency of carbohydrate to supply the body requirements for oxidative glucose metabolism. Glucose synthesis is, for the most
part, confined to the liver, although in prolonged starvation in
man the contribution by the kidney becomes substantial (Owen
et al., 1969). In certain situations the hepatic gluconeogenic
pathway may be operative in the face of an adequate dietary
supply of carbohydrate, when carbon precursors are diverted
towards the repletion of liver glycogen rather than to glucose
synthesis for hepatic release into the circulation. Such situations
include re-feeding after starvation (Hems et al., 1972) and the
analogous situation of weaning following the suckling stage of
neonatal development (K. Snell, unpublished work).
Hepatic gluconeogenesis is regulated within the liver by metabolic and hormonal factors and indirectly by the supply of gluconeogenic precursors from peripheral tissues (see Exton, 1972
for review). Not only are the various precursors utilized at
different rates for glucose formation within the liver, but the
physiological circulating concentrations of many precursors lie
below the saturating levels for hepatic glucose synthesis, as
demonstrated in the perfused liver in vitro (Exton & Park, 1967;
Mallette et al., 1969a) and the whole animal in vivo (Aikawa et
al., 1972). Thus both the rate of precursor supply and the nature
of the precursors will influence the rate of glucose formation by
the liver. The principal endogenous glucogenic precursors are
lactate, pyruvate and amino acids derived from skeletal muscle
and other tissues, and glycerol derived from the hydrolysis of triacylglycerol in adipose tissue. Cahill and his colleagues have
quantified the relative contribution of these various precursors to
glucose formation in man, particularly during starvation when
gluconeogenesis is increased (Cahill, 1970). As starvation is
extended beyond 24 h, the relative contribution of proteinderived amino acids to total hepatic glucose production
increases. A similar increased contribution has been observed in
diabetic subjects where gluconeogenesis is also elevated above
normal (Wahren et al., 1972). Of these amino acids, the contribution by alanine to hepatic glucose output is quantitatively the
most significant. Hepatic extraction of alanine exceeds that of all
other amino acids in the rat (Ishikawa, 1977) and man (Felig et
al., 1969a) in vivo, and, in addition, it has been shown in the
perfused liver in vitro that alanine is the best amino acid substrate for gluconeogenesis in the adult rat (Ross et al., 1967;
Exton & Park, 1967). A specific stimulation of alanine transport into the liver cell has been found in starvation and after glucagon stimulation of gluconeogenesis (Mallette et al., 19696;
Fehlmann et al., 1979).
Sources of circulating alanine
Measurements of arteriovenous differences across specific
organs and work with isolated preparations of various tissues
have shown that alanine is released into the circulation by a
number of tissues, including skeletal muscle, heart, small intestine, kidney, adipose tissue and brain (Aikawa et al., 1973; IshiVol. 8
kawa, 1977; Ruderman & Berger, 1974; Snell & Duff,1977a,b;
Taegtmeyer et al., 1977; Windmueller & Spaeth, 1974, 1975;
Hanson & Parsons, 1977). Of these sources, skeletal muscle is
undoubtedly the most significant contributor to circulating
alanine, because it represents such a large fraction of the total
body mass compared with other tissues and, moreover, contains almost 60% of the total body protein. Muscle protein
therefore serves as an important metabolic reservoir of amino
acid precursors for gluconeogenesis. Measurements of arteriovenous differences of amino acids across forearm and leg
muscles in man (Felig et al., 1970; Felig & Wahren, 197 1) or rat
(Ishikawa, 1977), and measurements of the pattern of amino
acid release from the perfused hind-limb (Ruderman & Berger,
1974) or isolated rat muscle preparations (Odessey et al., 1974;
Garber et al., 1976a) have shown that the relative proportions of
the amino acids released by muscle do not correspond to the
fractional amino acid composition of muscle proteins. The
striking feature is that alanine and glutamine together account
for more than 50% of the total amino acids released by skeletal
muscle, which is considerably more than their relative abundance in muscle proteins. The implication of these findings is
that alanine and glutamine release involves not only muscle proteolysis but also the synthesis de nouo of these amino acids from
other amino acids within the muscle. The importance of alanine
as a precursor for hepatic gluconeogenesis has been discussed
above. In contrast, glutamine is not removed from the circulation by the liver but by the small intestine and kidney. The
mucosal epithelial cells of the small intestine apparently utilize
glutamine in a major way as a respiratory fuel (Hanson &
Parsons, 1977; Windmueller & Spaeth, 1978; Watford et al.,
1979). In addition, glutamine contributes both carbon and
nitrogen for alanine formation and release by the intestine, so
that glutamine is indirectly involved in hepatic gluconeogenesis
insofar as it serves as a precursor of circulating alanine. Extraction of glutamine by the kidney is elevated in starvation and
under acidotic conditions when the amino nitrogen is removed
to produce ammonia for the neutralization of the acid urine (see
Pitts, 1964). The carbon of glutamine may be oxidized or used
for the synthesis of glucose (Hems, 1972); renal gluconeogenesis becomes of increasing significance with prolonged starvation
(Cahill & Owen, 1968). What contribution glutamine makes to
the net release of alanine by the kidney (Aikawa et al., 1973) is
presently not known. Thus both alanine and glutamine released
by skeletal muscle will make significant contributions to body
glucose formation, although it is only alanine that is involved
directly in hepatic gluconeogenesis.
Muscle alanineformation
Because alanine release by muscle is out of proportion to its
relative abundance in muscle proteins, it is presumed that synthesis de nouo is involved in the formation of this amino acid in
muscle. Alanine is formed by the transamination of glutamate
and pyruvate catalysed by alanine aminotransferase (EC
2.6.1.2). For amino acids other than glutamate to contribute
amino groups for alanine synthesis, they must first be transaminated with 2-oxoglutarate to produce glutamate, since direct
transamination with pyruvate in muscle is essentially restricted
to glutamate (Rowsell, 1956; Rowsell & Corbett, 1958). Muscle
2-oxoglutarate transamination only occurs to any appreciable
extent with aspartate and the branched-chain amino acids
(valine, leucine and isoleucine) (Krebs, 1975), so that these
205
BIOCHEMICAL SOCIETY TRANSACTIONS
206
Amino-oxyacetata
1-Cycloserine
r--Valine
I
I
I
/
I
3-Methyl-2-oxobutanoata-
.-,-Succinyl-CoA
!Citric acid cycle
1
Oxaloacetate
--
Alanine
'm
'
T
3-Mercap;opicdinate
Phoophoenolpyruvste
Lactatej l r " r u ~ ~ ~ Acetyl-CoA
i c h l o ~ o ~ c e t a t e
I
Miowlo
Glycogen
A
b
other tissues
Liver
1
1
amino acids and glutamate itself will be the major precursors for
alanine formation. The linked transaminations catalyse the
reactions:
L-Amino acid + 2-oxoglutarate * L-glutamate+ 2-moacid
L-Glutamate + pyruvate s 2-oxoglutarate + L-alanine
Net = L-Amino acid + pyruvate = 2-0x0 acid + L-alanine
Evidence for the involvement of transamination in alanine formation has come from the inhibitory effects of L-cycloserine and
amino-oxyacetate (potent transaminase inhibitors) on alanine
release from muscle in vivo (Blackshear et al., 1975) and in
vitro (Ruderman & Berger, 1974; Garber et al., 1976a,b; Snell
& Duff, 1977a; Snell, 1978) (see Scheme 1).
Alanine aminotransferase catalyses a freely reversible reaction with an equilibrium constant of 0.67 (Krebs, 1953).
:Measurements of the mass-action ratio of the reactants in
freeze-clamped leg muscle from fed rats gave a value of 0.58
(Ruderman et al., 1977), which suggests that the concentrations of reactants in the cell are near to equilibrium values. Thus
changes in the intracellular concentration of either glutamate or
pyruvate will lead to changes in alanine concentration. The
increased formation and release of alanine by the leg muscles of
diabetic rats is apparently the result of an increase in intracetlular glutamate concentration leading to a displacement of the
ahnine amhotransferase reaction from equilibrium and an
increase in the mass action ratio from 0.58 to 0.95 (Ruderman
et al., 1977). Since the branched-chain aminotransferase reaction has a similar equilibrium constant (0.57 with leucine as
reactant; Taylor et al., 1970), it is to be expected that the provision of branched-chain amino acids would lead to an increase
in intracellular glutamate concentration and an increased formation and release of alanine by the linked transaminations
shown above. A number of workers have shown a stimulation of
alanine release by various muscle preparations with added
branched-chain amino acids (Odessey et al., 1974; Ruderman &
Berger, 1974; Garber et al., 19766; Goldstein & Newsholme,
1976; Snell, 1976; Snell & Duff, 1977~).The freely reversible
nature of these transaminations is shown by the fact that the
addition of a branched-chain keto acid, 4-methyl-2-oxopentanoate or 3-methyl-2-oxobutanoate to hemidiaphragm incubations decreased alanine release to 50 and 33% respectively of
the control values obtained without added keto acid (K. %ell&
D. A. Duff, unpublished observations). Similar observations
have been made by Spydevold et al. (1976) in the perfused hindlimb preparation. The addition of the branched-chain keto acids
displaces the branched-chain aminotransferase reaction towards
branched-chain amino acid and 2-oxoglutarate formation at the
expense of glutamate. The decreased glutamate concentration
will result in increased alanine transamination with 2-oxoglutarate, thus reducing the net formation of alanine.
Changes in intracellular pyruvate availability for transamination to alanine can be brought about by stimulating pyruvate
oxidation with dichloroacetate (which activates pyruvate
dehydrogenase). This agent decreases alanine release from
muscle In vivo (Blackshear et al., 1975) and decreases alanine,
pyruvate and lactate release from muscle preparations In ultro
(Snell, 1976; SneU & Duff, 1977a; Goodman et al., 1978) (see
Scheme 1). The hypoglycaemic action of dichloroacetate (Stacpoole & Felts, 1970; Blackshear et al., 1974) may, in part, be
due to a diminished supply of gluconeogenic precursors from
muscle to the liver. On the other hand, increasing the availability of intracellular pyruvate by perfusing hind-limb preparations with lactate (Ruderman & Berger, 1974), by incubating
isolated muscles in uftro with glucose (Odessey et al., 1974;
Snell & Duff, 1977a) or pyruvate (Garber et al., 1976a), or by
infusing pyruvate into post-absorptive man in uivo (Pozefsky &
Tancredi, 1972) all increased the release of alanine. A consideration of the intracellular source of pyruvate for muscle alanine
formation has led a number of workers to emphasize the role of
glycolysis, and this has prompted the formulation of a 'glucose+
a l e e cycle' (Mallette et al., 1969a; Felig et al., 1970; Felig &
Wahren, 1974).
The glucose-alanine cycle
The formulation of a glucose-alanine cycle embodies the concept that glucose taken up by muscle from the bloodstream
1980
207
BIOCHEMICAL REVIEW
serves directly, or indirectly via glycogen, as a glycolytic source
of pyruvate for alanine synthesis. Alanine is released into the circulation, taken up by the liver and converted by gluconeogenesis into glucose, which is then returned to the bloodstream (see
Scheme 1). This scheme involves a constant recycling of glucose carbon between muscle and liver and complements the
carbon recycling via lactate of the Cori cycle. The Cori cycle
ensures that when skeletal muscle derives the energy for contraction by anaerobic glycogenolysis, the muscle lactate so
formed can be reconverted by the liver to glucose, which can
then serve to replenish the muscle glycogen stores (Cori, 193 1).
The physiological role of the glucose-alanine cycle, however,
probably rests in the fact that it serves as a transport system for
amino acid nitrogen from muscle to liver, where the nitrogen is
eliminated via conversion to urea Deamination of amino acids
and ammonia formation occurs during muscular exercise
through the operation of the purine nucleotide cycle (Lowenstein, 1972; Lowenstein 8c Goodman, 1978). Diversion of some
of the amino acid nitrogen towards alanine formation would
serve to prevent excessive generation of potentially toxic
ammonia by the purine nucleotide cycle. The precise roles of
amino acid breakdown and ammonia generation in muscle are
not clear, but it may be that one function of ammonia formation is as an intracellular buffer to hydrogen ion production
during ATP hydrolysis associated with contraction. Alanine formation may also serve as an alternative glycolytic end product
to lactate. Diversion of some of the pyruvate towards alanine
synthesis might prevent an excessive build-up of lactate, which
might disturb the cellular redox state through the lactate
dehydrogenase reaction. Both of these roles of alanine formation suggest that alanine release would be proportional to the
extent of muscle anaerobic glycolysis and the intensity of muscular exercise. Felig 8c Wahren (1971) found with exercising
human subjects that an increased release of alanine from leg
muscles was indeed proportional to the intensity of the work
being performed. However, contradictory findings have been
reported with the contracting perfused hind-limb preparation of
the rat, where lactate (and ammonia) release increased with the
frequency of muscle contractions, but alanine release was
decreased (although its concentration in muscle tissue did
increase) (Goodman 8c Lowenstein, 1977).
Evidence for decreased alanine formation due to a limitation
of the glycolytic provision of pyruvate has come from
observations on a patient with McArdle’s syndrome (Wahren er
al., 1973). This condition is characterized by a lack of muscle
phosphorylase and a consequent impairment of glycogenolysis.
During muscular exercise the rate of glycolysis fails to increase,
and decreases in arterial pyruvate and alanine concentrations
were observed. Furthermore, the net release of alanine from leg
muscle normally observed during exercise is abolished, and
indeed reversed, in patients with McArdle’s syndrome (Wahren
et al., 1973). In this pathological situation it appears that the
rate of glycolysis becomes a limiting factor for alanine release
during muscular exercise.
In most physiological situations it seems unlikely that the rate
of glycolysis is rate-limiting for alanine formation. Stimulation
of glucose uptake and glycolysis by the addition of insulin to
muscle incubations (K. Snell 8c D. A. Duff, unpublished observation; Odessey et al., 1974; Garber et al., 1976a) or to hind-
limb perfusions (Ruderman & Berger, 1974; Goodman &
Lowenstein, 1977) had a negligible or slightly inhibitory effect
on alanine release, despite a stimulation of lactate and pyruvate
release. Similarly, the addition of dibutyryl cyclic AMP to
muscle homogenates (Snell, 1975) or B-adrenergic agonists to
intact muscle preparations (Garber et al., 1976c; Li & Jefferson, 1977; Goodman & Lowenstein, 1977) decreased alanine
release while increasing lactate and pyruvate release. The
decreased alanine release in these experiments with hormonal
agents is probably due to inhibitory effects on protein degradation (Fulks et al., 1975; Li 8c Jefferson, 1977) and hence a
decreased availability of amino acids for glutamate formation
and danine synthesis. A lack of association between glycolysis
and alanine formation is also seen during starvation, where rates
of release of pyruvate and lactate by incubated diaphragm
muscle show a progressive increase as starvation periods
become greater, but alanine release displays a delay of about
24h before increases are noted (Fig. 1; Snell 8c Duff, 1979).
Again, the pattern of alanine release during progressive starvation correlates with protein degradation (Li & Goldberg,
1976). Pardridge & Davidson (1979) have also observed a lack
of correlation between glycolysis and alanine formation in a
myogenic line of rat skeletal-muscle cells in tissue culture. They
conclude that alanine formation is linked to the intracellular concentration of pyruvate (above a certain critical level), but that
this is not necessarily a function of the rate of glucose utilization.
Although recycling of glucose carbon in a glucose-alanine
cycle may be a source of much of the pyruvate used for alanine
synthesis in exercising muscle in the fed rat, the situation during
starvation, when there is an increased body requirement for net
glucose synthesis, is different. The glucose-alanine cycle will not
provide for any net addition of glucose to the body pool. In the
face of continued oxidative glucose metabolism and in the
absence of any dietary carbohydrate, glucose must therefore be
synthesized from endogenous precursors. The major consumer
of glucose during moderate starvation is the brain, although
there is an adaptation of brain metabolism towards ketone-body
utilization in prolonged starvation, which lessens this glucose
requirement somewhat. During starvation, muscle protein is
broken down and the amino acid carbon, which is released
mainly in the form of alanine, can be converted by the liver to
glucose for use by the brain. Indeed, calculations, based on
arteriovenous differences across gluconeogenic tissues and
nitrogen excretion in humans (Owen et al., 1969; Cahill, 1970),
show that in starvation the glucose requirements of the brain can
be accounted for by gluconeogenesis from amino acids derived
from muscle protein (plus a smaller contribution from glycerol)
(see Newsholme, 1976). In the very early stages of starvation
some of the pyruvate available for alanine formation may arise
by glycolysis from the partial breakdown of muscle glycogen
reserves (Sugden et al., 1976; Snell & Duff, 1979). However,
muscle protein breakdown is potentially a much more significant source of carbon and one that persists after glycogen
breakdown has ceased (Table 1; Snell & Duff, 1979). Thus
pathways must exist in muscle for the conversion of amino acid
carbon skeletons to pyruvate for alanine formation, so that the
alanine released may make a net gluconeogenic contribution to
the body glucose pool.
Table 1. Potential yield of blood glucose from rat diaphragm-muscle glycogen and protein during starvation
Data taken from Snell & Duff (1979). The maximum glucose yield assumes a maximum yield of 1g of glucose from 1.7g of protein
(Krebs, 1964).
Starvation period
(h)
0-24
24-48
48-72
Vol. 8
Loss of glycogen
(mg/diaphragm)
0.25
0.00
0.07
Maximum glucose yield
(mol)
1.4
0.0
0.4
Loss of protein
(mg/diaphragm)
8.7
2.2
3.4
Maximum glucose yield
(WOl)
28.5
1.2
11.1
208
Pathways of alanine synthesis from amino acids
Valine, isoleucine, aspartate and glutamate are all theoretically able to supply pyruvate carbon for alanine synthesis, by
contributing net carbon to the citric acid cycle. C, citric acidcycle intermediates can then produce pyruvate or phosphoenolpyruvate by decarboxylation. Leucine, which does not contribute net carbon to the citric acid cycle, is unable to produce
pyruvate in this way. There is considerable evidence that the
pyruvate for muscle alanine synthesis may originate, in part,
from other amino acids. Inhibition of glucose uptake and metabolism by inhibitors of glycolysis, such as NaF (Goldstein &
Newsholme, 1976) or iodoacetate (Garber et al., 1976a), had no
effect on the rate of alanine release, or on the stimulation of alanine release by added amino acids, in isolated muscle incubations. Contradictory findings with these inhibitors have been
reported by Chang & Goldberg (1978~).Garber et al. (19766)
concluded on the basis of carbon-balance measurements that, in
the presence of iodoacetate, the stimulation of alanine formation by the addition of amino acids was accompanied by a
net efAux of total carbon that could only be accounted for by a
contribution of amino acid carbon to alanine formation. Similar
conclusions were reached by Goldstein & Newsholme (1976),
who found that incubation of hemidiaphragms with glutamate or
isoleucine in the absence of glucose increased alanine output and
increased the intracellular concentration of pyruvate.
Evidence for a contribution of branched-chain amino acid
carbon to the concentrations of citric acid-cycle intermediates
has been presented (Spydevold et al., 1976; Lee & Davis, 1979),
and estimates of the rate of flux of C, citric acid-cycle intermediates to C, metabolites show that this is in excess of that
required to account for reported rates of alanine release (Lee &
Davis, 1979). On the other hand, Chang & Goldberg ( 1 9 7 8 ~ )
discount any contribution of amino acidderived pyruvate to
alanine formation on the grounds that the rate of formation of
glycolytically derived pyruvate in diaphragms incubated with
5 mM-glucose exceeds the possible net formation of pyruvate
from proteolytically derived amino acids by some 30-fold. However, the rate of glycolysis in such muscle incubations in uitro is
many times greater than that which occurs in resting muscle in
uiuo or in the perfused hind-limb (Maizels et al., 1977). Grubb
(1976) has shown that in the perfused hind-limb only 2.7% of
the glucose taken up by the preparation appears as alanine, and
she calculates that there is a short-fall of carbon for alanine formation that can only be balanced by an additional contribution
through amino acid metabolism.
It seems that indirect calculations of potential contributions
by glycolysis and amino acid metabolism to muscle pyruvate
formation give little indication of the actual relative contribution by either process. The absolute value for potential pyruvate formation from amino acids (Chang & Goldberg, 1978a) is
sufficient to account for observed rates of alanine formation.
Indeed in a subsequent paper, Chang & Goldberg (19786)
calculate, on the basis of the amino acid frequency in muscle
proteins and the actual measured rates of release by muscle, that
the rate of intracellular metabolism of amino acids produced by
protein breakdown is sufficient to account for the formation of
glutamine de nouo. But, by using the same considerations, one
could equally well account for the formation de nouo of alanine
in a similar way. On the basis of the relative incorporation of
radioactivity from [I4Clvalineor [14Clsuccinateinto alanine and
glutamine, Chang & Goldberg (19786) conclude that glutamine
formation is the major fate of amino acid metabolism by muscle.
However, such quantitative conclusions of net carbon flux are
difficult to make in view of the possibilities of radioactivity
exchange reactions and particularly since the specific radioactivities of the immediate precursors of the alanine and glutamine
were not known. In addition, the assumptions that the specific
radioactivity of lactate released into the medium is equal to that
of intramitochondrial pyruvate, and that the ratio of radioactivity recovered in lactate to that recovered in CO, equals the
BIOCHEMICAL SOCIETY TRANSACTIONS
ratio of the rate of lactate release to that of pyruvate oxidation
are not valid (Ottaway & Mowbray, 1977), and so the quantitative interpretations of the metabolic fates of pyruvate and of
carbon chains entering the citric acid cycle (Chang & Goldberg, 19786) are very uncertain. The findings of incorporation of
radioactivity from [I4Clvaline, [“Clsuccinate (Chang & Goldberg, 19786) or [14Clpropionate (Lee & Davis, 1979) into
alanine by muscle preparations indicate that the necessary enzymic reactions for the conversion of amino acid carbon to alanine
are present in muscle.
The pathway for the production of pyruvate from citric acidcycle intermediates could involve the decarboxylation of malate
by NAD+- or NADP+-dependent malate dehydrogenase (decarboxylating), or the decarboxylation of oxaloacetate to phosphoenolpyruvate by phosphoenolpyruvate carboxykinase followed
by pyruvate kinase action. Both phosphoenolpyruvate carboxykinase (Opie & Newsholme, 1967; Nolte et al., 1972; Newsholme & Williams, 1978; Snell & Duff, 1979) and NAD+- or
NADP+-dependent malate dehydrogenase (Nolte et al., 1972;
Lin & Davis, 1974; Newsholme & Williams, 1978; Snell &
Duff, 1979) activities have been found in skeletal muscle. During
starvation, when muscle alanine release is increased (Karl et al.,
1976; Snell & Duff, 1979), phosphoenolpyruvate carboxykinase activity is also increased (Newsholme & Williams, 1978;
Snell & Duff, 1979), whereas NADP+-dependent malate dehydrogenase activity is decreased (Snell & Duff, 1979) or is
unchanged (Karl et al., 1976; Newsholme & Williams, 1978).
Similarly, dietary supplementation with valine or isoleucine
leads to an increase in alanine release by muscle (Snell, 1979),
with an increase in phosphoenolpyruvate carboxykinase activity but little change in NADP+-malate dehydrogenase activity
(Newsholme & Williams, 1978; Snell, 1979). Also, mercaptopicolinate, an inhibitor of phosphoenolpyruvate carboxykinase,
inhibits the glutamate- or valine-stimulated release of alanine by
diaphragm muscle from starved rats ( b e l l & Duff, 1977~).This
is further evidence for a pathway of pyruvate formation for
alanine synthesis, which involves decarboxylation of citric acidcycle intermediates via phosphoenolpyruvate carboxykinase
(Scheme 1). Alanine release in the presence of leucine, which
cannot contribute net carbon to the citric acid cycle, was
unaffected by mercaptopicolinate. In fed rats, mercaptopicoh a t e was ineffective in blocking glutamate- or valine-stimulated alanine release by diaphragm muscle (Snell & Duff,
1977a) or in blocking incorporation from [14Clpropionateinto
alanine in perfused hind-limb preparations (Lee & Davis, 1979).
The reasons for the difference between fed and starved rats in
this respect are not clear, but it suggests that in the fed rat decarboxylation of citric acid-cycle intermediates occurs by a
route other than via phosphoenolpyruvate carboxykinase. A
similar difference in mercaptopicolinate sensitivity between fed
and starved rats was observed for valine-stimulated alanine
release from adipose tissue ( h e l l & Duff, 19776), and the agent
was ineffective in blocking glutamine-stimulated alanine release
by isolated intestinal preparations from fed rats (Hanson &
Parsons, 1977; Watford et al., 1979). Thus the pathway for
pyruvate production from citric acid-cycle intermediates appears to vary between different tissues, and in the same tissue in
different physiological situations. Nevertheless, in situations
associated with enhanced rates of gluconeogenesis, muscle
appears to synthesize alanine by a pathway involving phosphoenolpyruvate carboxykinase.
Branched-chain amino acid metabolism and alanine formation
Muscle is believed to be the major body site of branchedchain amino acid metabolism (Adibi, 1976), in contrast with
most other amino acids that are predominantly metabolized by
the liver. Thus the release of branched-chain amino acids by
muscle is less than would be expected from their relative frequency in muscle proteins (Ruderman & Berger, 1974; Odessey et al., 1974), and it can be calculated that their intracellular
1980
209
BIOCHEMICAL REVIEW
metabolism is sufficient to account for the observed rate of alanine synthesis de nouo (cf. Chang & Goldberg, 19786). In vitro
the branched-chain amino acids are among the most effective
precursors for alanine formation (Ruderman & Berger, 1974;
Odessey et al., 1974; Garber et al., 19766; Snell & Duff,
1977~).Intracellular protein breakdown is probably the major
source of the amino acids, particularly in situations such as starvation and diabetes, although an additional supply may come
from the liver, which releases branched-chain amino acids
during starvation in rats (Mallette et al., 19698; Bloxham, 1972)
and man (Marliss et af., 1971), in diabetic rats (Miller, 1962)
and man (Felig, 1973), and after protein ingestion in normal and
diabetic man (Wahren et al., 1976). In the last example the
actual source of the amino acids is from intestinal absorption,
but, because of the minor role of the liver in their metabolism,
they escape hepatic uptake and account for more than half of
the initial total amino acid uptake by leg muscles (Wahren et al.,
1976). Protein feeding results in a continuous output of alanine
and glutamine by leg muscles (Wahren et al., 1976), and it is
likely that this is due, in large part, to metabolism of the
branchedchain amino acids taken up by the muscle. It is significant in this respect that dietary supplementation of rats with
branched-chain amino acids results in increased alanine output
by muscles in uitro (Snell, 1979) and in elevations of phosphoenolpyruvate carboxykinase (the enzyme implicated in the
carbon pathway from branched-chain amino acids to alanine)
(Newsholme & Williams, 1978; Snell, 1979), and branchedchain keto acid dehydrogenase (the rate-limiting enzyme of
branched-chain amino acid catabolism) (Shinnick & Harper,
1977). Similarly, starvation, which results in increased alanine
release by muscle (Snell & Duff, 1979), is also accompanied by
increased branched-chain amino acid metabolism (Goldberg &
Odessey, 1972) and elevated phosphoenolpyruvate carboxykmase activity (Newsholme & Williams, 1978; Snell & Duff,
1979). These findings suggest that there is an intimate link
between branched-chain amino acid oxidation and alanine formation in muscle.
As well as acting themselves as precursors for alanine synthesis, the branched-chain amino acids may also control the flow
of other amino acid mecursors derived from protein breakdown
and so indirectly atTect alanine synthesis (see Scheme 1). Thus
the branched-chain amino acids, and leucine in particular, uniquely stimulate protein synthesis and independently inhibit protein degradation in muscle (Odessey et al., 1974; Fulks et al.,
1975; Buse & Reid, 1975). The mechanisms of these effects on
protein turnover are not known, but metabolism of the
branchedchain amino acids may play a crucial regulatory role,
either by determining the intracellular concentrations of the
amino acids themselves or of some regulatory product of amino
acid oxidation, which in turn influence the process of protein
turnover. The association between branched-chain amino acid
metabolism, protein degradation and alanine release is illustrated by the similarity in the patterns of these metabolic processes
during progressive starvation (Fig. 1).
One feature of branched-chain amino acid metabolism and
alanine formation that is not apparent from Scheme 1 is the
possible location of the various reactions in different intracellular compartments. The coupling of the branched-chain aminotransferase and alanine aminotransferase to effect the transfer of
nitrogen from the branchedchain amino acids to pyruvate to
form alanine suggests that it might be appropriate if these two
enzymes shared the same intracellular compartment to facilitate
this transfer. Over 90% of alanine aminotransferase activity is
present in the cytosol fraction of various muscle types (Table 2;
DeRosa & Swick, 1975). Moreover, kinetic considerations suggest that the cytosol isoenzyqe form of alanine aminotransferase is more appropriate for catalysing alanine formation from
pyruvate than the mitochondrial isoenzyme form (DeRosa &
Swick, 1975). Ichihara et al. (1975) have reported that most of
the branchedchain aminotransferase activity (72%) is located in
Vol. 8
lo
r
- 0.4
$=
.c,
E
- 0.3.Zo 0
-
0.2
-
0.1
3:
3 e:
O M
I=---
4z
> E
OL
-0
OL
1
0
2
3
Starvation period (days)
Fig. 1. Effect of progressive starvation on alanine release, valine
oxidation andprotein degradation by rat muscle
Alanine release (0) in the presence of 3 mwvaline (Snell & Duff,
1979) and [ l-l'Clvaline oxidation (0)
in the presence of 0.1mMvaline (Goldberg & Oddessey, 1972) were measured in incubations with hemidiaphragms. Protein degradation was
measured by the rate of tyrosine released (A) by soleus muscle
incubated with 10mM-ghCOSe and 0.5 mM-cycloheximide (Li &
Goldberg, 1976). The data are taken from the references cited.
Table 2. Subcellular distribution of alanine aminotransferase
in rat muscle tissues
Particle and cytosol fractions were prepared by differential
sedimentation from rat muscle homogenates as described by
Snell (1978). Alanine aminotransferase was assayed as previously (Snell, 1978) and activities are given as the meanf
S.E.M. for four observations in each case
Alanine aminotransferase activity (pmollmin per g
of tissue)
Muscle tissue
Diaphragm
Heart
Hind-limb
r
Total
2.75 k0.24
2.10k0.20
4.84k0.35
Particulate
0.19kO.02
0.12k0.01
0.54+0.02
cytosol
3
2.48k0.22
1.91 k0.28
4.78k0.32
the particulate fraction of (unspecified) muscle. In addition, the
subsequent oxidation of the keto acid carbon chain derived from
the branched-chain amino acids takes place intramitochondrially (see Dancis & Levitz, 1972). From these findings it may
be inferred that there must be a transfer of branched-chain
amino acid nitrogen from the mitochondria to the cytosolic location of alanine aminotransferase, and Scheme 2(a) (Snell &
Duff, 1977a) is an attempt to account for this. In this scheme
mitochondrial aspartate efflux serves a dual function as a carrier
of both branched-chain amino acid nitrogen and of oxaloacetate carbon derived from oxidation of the valine carbon skeleton. Transamination of aspartate in the cytosol not only donates
amino nitrogen for glutamate and alanine formation, but also
generates oxaloacetate, which, through the action of phosphoenolpyruvate carboxykinase, can provide phosphoenolpyruvate
and pyruvate for alanine formation. The presence of aspartate
aminotransferase in both mitochondrial and cytosolic compartments (Snell, 1978) is necessary for aspartate to fulfil this function. However, Odessey & Goldberg (1979) have reported that
branched-chain aminotransferase activity is largely (68%)
cytosolic in hind-limb muscles, while confirming that further
metabolism of the carbon skeleton is confined to the mitochondrial compartment. This contradictory finding on the intracellular location of the aminotransferase activity necessitates
a reappraisal of the scheme for the compartmentalization of the
reactions involved in alanine formation, as shown in Scheme
2(b). In this revised scheme, mitochondrial aspartate efflux
is still required to provide cytosolic oxaloacetate in order to
BIOCHEMICAL SOCIETY TRANSACTIONS
210
Cytosol
IbJ
VaIine ~ 2 - O x o g l u ~ r a t e ~ . ~ l a n i n e
3-MetwZ-oxobutyrate
Glutamate
P y r u v a t e L Phosphoenolpyruvate
Aspartate'
2-Oxoglutarate
Cytosol
Mitochondria
4
3 - M e t h v l - 2 - o x o b u t v r a t e ~ Succinyl-CoA,~
-
I
I
Aspartate
2-Oxoglutarate
6 V A . B
0xaloacetate"Glutamate
I
Scheme 2. Intracellular compartmentalization of the proposed pathway of alanine
formationfrom valine in muscle of starved rats
Mitochondrial-cytosolic inter-relationships based on ( a ) mitochondrial localization of
branched-chain aminotransferase, and (b) cytosolic localization of branched-chain aminotransferase are shown. The locus of action of the inhibitor 2-amino-4-methoxy-trans-but3-enoate fAMB) is indicated. The enzymes involved are: 1, alanine aminotransferase (EC
2.6.1.2); 2, cytosolic aspartate aminotransferase (EC 2.6.1.1); 3, phosphoenolpyruvate
carboxykinase (EC 4.1.1.32); 4,pyruvate kinase (EC 2.7.1.40); 5, branched-chain aminotransferase (EC 2.6.1.6); 6, mitochondrial aspartate aminotransferase (EC 2.6.1.1); 7,
enzymes converting 3-methyl-2-oxybutyrate into succinyl-CoA; 8, enzymes of the citric
acid cycle converting succinyl-CoA into oxaloacetate.
produce pyruvate for alanine formation by the pathway involv- formation, this prevents the reoxidation of cytosolic NADH,
ing phosphoenolpyruvate carboxykinase. Evidence for a role of which is normally effected by the reduction of pyruvate to lacaspartate transamination in the formation of alanine has come tate by the lactate dehydrogenase reaction.
from experiments with 2-amino-4-methoxy-trans-but-3-enoic
acid, a relatively specific inhibitor of aspartate aminotransferase, Role of alanine in regulation of interorgan metabolism
which diminished the valine-stimulated release of alanine by diaThe importance of muscle alanine formation in hepatic glucophragm muscle in vitro (Snell, 1978). Other evidence has shown neogenesis during starvation has been emphasized in this disthat there is a mitochondrial efflux of aspartate in the perfused cussion. During starvation there is a net breakdown of body
rat heart associated with an increased concentration of alanine protein to provide amino acid carbon, carried mainly in the form
of alanine, for the net synthesis of glucose by the liver. The
(Safer & Williamson, 1973).
Where the source of pyruvate for alanine formation is from major organ involved in the utilization of circulating glucose in
glycolysis rather than from citric acid-cycle intermediates (as these circumstances is the brain. However, it is evident that a
when leucine serves as a precursor), then the oxaloacetate continued breakdown of proteins, as starvation continues for
carbon from aspartate must be returned from the cytosol to the .prolonged periods, is likely to jeopardize survival. Thus it is not
mitochondria in the form of malate to prevent depletion of mito- surprising that adaptive mechanisms have evolved to prevent the
chondrial oxaloacetate. The cytosolic formation of malate from dissolution of body protein in this situation, and nitrogen excreoxaloacetate also serves to reoxidize cytosolic NADH formed in tion is drastically curtailed. Coincident with this diminution of
the triose phosphate dehydrogenase glycolytic reaction and protein catabolism, glucose utilization and production are
ensures a continued glycolytic production of pyruvate for markedly reduced (Owen et al., 1969), and this is due to a
alanine formation. This circumvents the theoretical dificulty decreased utilization of glucose by the brain, which adapts to the
t h t when pyruvate formed by glycolysis is utilized for alanine utilization of circulating ketone bodies to satisfy its energy
1980
21 1
BIOCHEMICAL REVIEW
Scheme 3. Regulatory role of alanine and ketone bodies in the interorgan metabolism of
carbohydrate,fat and protein
Solid lines indicate biochemical transformations and broken lines indicate regulatory
effects; 0.stimulation; 0,
inhibition; BCAA, branched-chain amino acids. This scheme is
modified from that of Newsholme (1976). In starvation, ketone bodies replace glucose as a
fuel for brain metabolism, restrict lipolysis in adipose tissue, so that fatty acid provision
keeps pace with oxidation in muscle and with ketogenesis, and conserve muscle protein by
inhibiting protein breakdown and alanine formation. Diminished muscle alanine release
restricts hepatic gluconeogenesis, in keeping with the adaptations in brain metabolism. In
protein feeding, alanine serves as a gluconeogenic precursor, stimulates hepatic glucose
production and restricts hepatic ketogenesis, in keeping with the availability of glucose as
a fuel. Overlaying these metabolic controls are hormonal controls by insulin (released by
ketone bodies) and glucagon (released by alanine). Insulin and glucagon release is also
influenced by the blood glucose concentration (not shown).
requirements (Owen et al., 1967; 1969; Ruderman et al., 1974). mechanism by which protein breakdown and alanine release is
Because of this switch in metabolic behaviour by the brain, there diminished during prolonged starvation is not explicable in terms
is a reduced requirement for gluconeogenesis, which in turn of the hormonal changes in this situation (insulin is lowered and
serves to spare body protein breakdown.
glucagon is raised). It seems that ketone bodies may regulate the
The key factor responsible for the reduction in gluconeogene- breakdown of protein and the release of alanine by a direct
sis during prolonged starvation appears to be the decreased action on muscle. Blackburn et al. (1973) and Hoover et al.
availability of alanine from peripheral tissues. Thus circulating (1975) showed that infusions of amino acid into patients
alanine concentrations fall to a much greater extent than other recovering from surgery caused a hyperketonaemia while
amino acids, and this is reflected in a fall in the uptake of alanine reducing nitrogen loss. Infusion of ketone bodies into human
by the splanchnic bed (Felig et al., 1969~).Measurements of subjects resulted in a specific decline in plasma alanine compararteriovenous differences across the forearm of starved human able with that observed in starvation and caused a decrease in
subjects showed that the release of amino acids, but most urinary nitrogen excretion in starved subjects (Sherwin et al.,
dramatically that of alanine, was reduced during prolonged star- 1975). The addition of ketone bodies to incubated diaphragm
vation (Felig et al., 1970). That the availability of alanine, from muscle in vitro resulted in decreased protein breakdown and a
muscle and other peripheral tissues, rather than a change in its decrease in alanine release (Palaiologos & Felig, 1976). The
hepatic metabolism, is responsible for limiting gluconeogenesis mechanism of this effect of ketone bodies is not known, but, in
in this situation is shown by the immediate rise in blood glucose view of their inhibitory action on branched-chain amino acid
concentration, and in the incorporation of radioactivity from oxidation (Buse et al., 1972), one possibility is the accumulation
[“Clalanine into blood glucose that is observed after alanine of branched-chain amino acids in the muscle leading to inhibiinfusion into starved human subjects (Felig et al., 1969b). The tion of protein degradation and/or stimulation of protein
VOl. 8
9
212
synthesis, which would diminish the flow of amino acid
precursors for alanine formation. Inhibition of branchedchain amino acid metabolism by ketone bodies may also
directly affect alanine formation by blocking the flow of
carbon skeletons to pyruvate. Whatever the mechanism, these
findings indicate that the circulating concentrations of ketone
bodies regulate protein turnover in muscle, and, by diminishing
alanine release, reduce the blood concentration of alanine and
thereby decrease gluconeogenesis. Thus ketone bodies have a
dual role in prolonged starvation: they diminish glucose utilization by the brain by serving as an alternative fuel, and they
reduce gluconeogenesis through a decrease in alanine release,
which conserves body protein. The effect of ketone bodies on the
regulation of protein and amino acid metabolism extends the
role of these metabolites in the integration of metabolism in the
body (see Newsholme, 1976). Scheme 3 depicts these inter-relationships in metabolism and includes the effects of ketone bodies
on fat, carbohydrate and protein metabolism.
A further regulatory facet of alanine-ketone-body interactions is the effect of alanine on ketone-body production.
Alanine infusion is known to result in hypoketonaemia in man
(Genuth, 1973;Genuth & Castro, 1974) and rats (Ozand et al.,
1977). This effect is due to an inhibition of ketogenesis in the
liver (Ozand et al., 1978). An increase in circulating alanine will
result in an increase in gluconeogenesis (see above), but there is
also evidence that alanine will stimulate gluconeogenesis from
other precursors (Ozand el al., 1978). In addition, alanine is
known to promote the release of pancreatic glucagon (Muller et
al., 1971; Wise et al., 1972, 1973a,b), which stimulates hepatic
gluconeogenesis directly (Exton, 1972), as well as indirectly by
increasing alanine transport into the liver (Mallette ef al., 19696;
Fehlman el al., 1979). Thus an increase in circulating alanine,
such as occurs after protein feeding, will encourage glucose formation and inhibit ketogenesis, leading to a switch in body fuel
metabolism as the availability of glucose increases. When glucose is plentiful, there is evidence that alanine may have a stimulatory role in diverting gluconeogenic substrates towards glycogen synthesis (Katz et al., 1976).
It appears that the circulating concentrations of ketone bodies
and alanine may complement each other in controlling the interorgan patterns of carbohydrate, fat and protein metabolism
(Scheme 3).
Conclusions
It is clear that alanine plays a major role in glucose homoeostasis. One illustration of this is the many disease states in which
disturbances in glucose homoeostasis are associated with
alterations in the availability of alanine from peripheral tissues
(Felig, 1973; Ruderman, 1975). The carbon of the alanine
released from muscle must be derived from amino acids produced by protein breakdown, if the alanine is to make a net contribution to body glucose through hepatic gluconeogenesis. The
quantitative aspects of the amount of pyruvate for alanine formation that is produced from other amino acids compared with
that originating by glycolysis are not yet clear. The metabolic
pathway involved in alanine synthesis using the carbon of other
amino acids remains to be established, although current evidence favours a role for phosphoenolpyruvate carboxykinase in
this process. There appears to be a correlation between
branched-chain amino acid metabolism and alanine formation in
muscle, but further understanding is required of the precise interrelationship and their metabolic and hormonal regulation. An
understanding of this problem may help to explain the bouts of
hypoglycaemia associated with the genetic impairment of
branchedchain amino acid metabolism in Maple Syrup Urine
disease. The branched-chain amino acids may provide a link
between muscle protein turnover and alanine release, and future
work in this area might provide insights into possible mechanisms that underlie various muscle-wasting conditions, such as
BIOCHEMICAL SOCIETY TRANSACTIONS
diabetes, muscular dystrophy and tumour-related cachexia.
Circulating alanine concentrations influence metabolic processes in other tissues and vary inversely to ketone-body concentrations. The concentration of alanine and ketone bodies
may play an important role in integrating the interorgan metabolism of carbohydrate, fat and protein.
Adibi, S. A. (1976)Metab. Clin. Exp. 25, 1287-1302
Aikawa, T., Matsutaka, H., Takezawa, K. & Ishikawa, E. (1972) Biochim. Biophys. Acta 279,234-244
Aikawa, T., Matsutaka, H., Yamamoto, H., Okuda, T., Ishikawa, E.,
Kawano, T. & Matsumura, E. (1973) J. Biochem. (Tokyo) 74,
1003-1017
Blackburn, G. L., Flatt, J. P., Clowes, G. H. A., O’Donnell, T. F. &
Hensle, T. E. (1973) Ann. Surg. 125,588-594
Blackshear, P. J., Holloway, P. A. H. & Alberti, K. G. M. M. (1974)
Biochem. J. 142,279-286
Blackshear, P. J., Holloway, P. A. H. & Alberti, K. G. M. M. (1975)
Biochem. J. 150,379-387
Bloxham, D. L. (1972) Br. J. Nutr. 27,233-247
Buse, M. G. & Reid, S . S . (1975)J. Clin. Invest. 56, 1250-1261
Buse, M. G., Biggers, J. F., Friderici, K. H. & Buse, J. F. (1972)J. B i d .
Chem. 247,8085-8096
Cahill, G. F. (1970) N. Engl. J. Med. 282,668-675
Cahill, G. F. & Owen, 0. E. (1968) in Carbohydrate Metabolism and
its Disorders (Dickens, F., Randle, P. J. & Whelan, W. J., eds.), vol.
1, pp. 497-552, Academic Press, London
Chang, T. W. & Goldberg, A. L. ( 1 9 7 8 ~ J.
) Biol. Chem. 253, 36773684
Chang, T. W. & Goldberg, A. L. (19786) J. B i d . Chem. 253, 36853695
Cori, G. F. (193 1) Physiol. Rev. 11, 143-275
Dancis, J. & Levitz, M. (1972) in The Metabolic Basis of Inherited
Disease (Stanbury, J. B., Wyngarden, J. B. & Fredrickson, D. S.,
eds.), pp. 4 2 6 4 3 9 , McGraw-Hill, New York
DeRosa, G. & Swick, R. W. (1975)J. Biol. Chem. 250,7961-7967
Exton, J. H. (1972) Metab. Clin. Exp. 21,945-990
Exton, J. H. & Park, C. R. (1967) J. B i d . Chem. 242,2622-2636
Fehlmann, M., Le Cam, A., Kitabgi, P., Rey, J.-F. & Freychet, P.
(1979) J. Biol. Chem. 254,401-407
Felig, P. (1973) Metab. Clin. Exp. 22, 178-207
Felig, P. & Wahren, J. (1971) J. Clin. Inuest. 50,2703-2714
Felig, P. & Wahren, J. (1974) Fed. Proc. Fed. Am. SOC.Exp. Biol. 33,
1092- 1097
Felig, P., Owen, 0. E., Wahren, J. Br Cahill, G. F. ( 1 9 6 9 ~ J.
) Clin.
Inuest. 48,584-594
Felig, P., Marliss, E., Owen, 0. E. & Cahill, G. F. (1969b) Adu.
Enzyme Regul. 7 , 4 1 4 6
Felig, P., Pozefsky, T., Marliss, E. & Cahill, G. F. (1970) Science 167,
1003-1004
Fulks, R. M., Li, J. B. & Goldberg, A. L. (1975) J. B i d . Chem. 250,
290-298
) B i d . Chem. 251,
Garber, A. J., Karl, I. E. & Kipnis, D. M. ( 1 9 7 6 ~J.
826-836
Garber, A. J., Karl, I. E. & Kipnis, D. M. (19766) J. B i d . Chem. 251,
836-84 3
Garber, A. J., Karl, I. E. & Kipnis, D. M. (1976~)J . Biol. Chem. 251,
85 1-857
Genuth, S . M. (1973) Metab. Clin. Exp. 22, 927-937
Genuth, S. M. & Castro, J. (1974)Metab. Clin. Exp. 23,375-386
Goldberg, A. L. & Odessey, R. (1972) Am.J. Physiol. 233, 1384-1391
Goldstein, L. & Newsholme, E. A. (1976) Biochem. J . 154,555-558
Goodman, M. N. & Lowenstein, J. M. (1977) J. B i d . Chem. 252,
5054-5060
Goodman, M. N., Ruderman, N. B. & Aoki, T. T. (1978) Diabetes 27,
1065-1074
Grubb, B. (1976) Am. J. Physiol. 230, 1379-1384
Hanson, P. J. & Parsons, D. S . (1977) Biochem. J . 166,509-5 19
Hems, D. A. (1972) Biochem. J. 130,671-680
Hems, D. A., Whitton, P. D. & Taylor, E. A. (1972) Biochem. J. 129,
529-538
Hoover, H. C., Grant, J. P., Gorschbath, C. & Ketcham, A. S. (1975)
N. Engl. J. Med. 293, 172-175
Ichihara, A., Yamasaki, Y., Masuji, H. & Sato, J. (1975) in Isoenzymes (Markert, C. L., ed.), vol. 3, pp. 875-889, Academic Press,
New York
Ishikawa, E. (1977) Adu. Enzyme Regul. 14, 117-136
1980
BIOCHEMICAL REVIEW
Karl, I. E.,Garber, A. J. & Kipnis, D. M. (1976)J. Biol. Chem. 251,
844-850
Katz, J., Golden, S. & Wals, P. A. (1976)Proc. Natl. Acad. Sci. U.S.A.
73,3433-3437
Krebs, H. A. (1953)Biochem. J. 54,82-86
Krebs, H. A. (1964)in Mammalian Protein Metabolism (Munro, H. N.
& Allison, J. B., eds.), vol. 1, pp. 125-176, Academic Press, New
York
Krebs, H. A. (1975)Adv. Enzyme Regul. 13,449-472
Lee, S.-H. & Davis, E.J. (1979)J.Biol. Chem. 254,420-430
Li, J. B. & Goldberg, A. L. (1976)Am. J. Physiol. 231,441-448
Li, J. B. & Jefferson, L. S. (1977)Am. J. Physiol. 232, E243-E249
Lm,R. C. & Davis, E. J. (1974)J.Biol. Chem. 249,3867-3875
Lowenstein, J. M. (1972)Physiol. Rev. 52,382-414
Lowenstein, J. M. & Goodman, M. N. (1978)Fed. Proc. Fed. Am. SOC.
Exp. Biol. 37,2308-2312
Maizells, E. Z.,Ruderman, N. B., Goodman, M. N. & Lan, D. (1977)
Biochem. J. 162,557-568
Marliss, E. B., Aoki, T. T., Pozefsky, T., Most, A. & Cahill, G. F.
(1971)J.Clin. Invest. SO, 814-817
Mallette, L. E., Exton, J. H. & Park, C. R. (1969a)J.Biol. Chem. 244,
5713-5 723
Mallette, L. E., Exton, J. H. & Park, C. R. (19696)J. Biol. Chem. 244,
5724-5 728
Miller, L. L. (1962)in Amino Acid Pools (Holten, J. T., ed.), pp. 708721,Elsevier, Amsterdam
Muller, W. A., Faloona, G. R. & Unger, R. H. (1971)J. Clin. Invest.
50,2215-2218
Newsholme, E. A. (1976)Clin. Endocrinol. Metab. 5,543-578
Newsholme, E. A. & Williams, T. (1978)Biochem. J. 176,623-626
Nolte, J., Brdiczka, D. & Pette, D. (1972)Biochim. Biophys. Acta 284,
497-507
Odessey, R. & Goldberg, A. L. (1979)Biochem. J. 178,475-489
Odessey, R., Khairallah, E. & Goldberg, A. L. (1974)J. Biol. Chem.
249,7623-7629
Opie, L. H. & Newsholme, E. A. (1967)Biochem. J. 103,391-399
Ottaway, J. H. & Mowbray, J. (1977)Curr. Top. Cell. Regul. 12, 108208
Owen, 0.E., Morgan, A. P., Kemp, H. G., Sullivan, J. N., Herrera,
M. G. & Cahill, G. F. (1967)J.Clin.Znvest. 46,1589-1595
Owen, 0.E., Felig, P., Morgan, A. P., Wahren, J. & Cahill, G. F.
(1969)J. Clin. Invest. 48,474-583
Ozand, P. T., Reed, W. D., Givard, J., Hawkins, R. L., Collins, R. M.,
Tildon, J. T. & Cornblath, M. (1977)Biochem. J. 164,557-564
Ozand, P. T., Reed, W. D., Hawkins, R. L., Stevenson, J. H., Tildon,
J. T. & Cornblath, M. (1978)Biochem. J. 170,583-591
Palaiologos, G. & Felig, P. (1976)Biochem. J. 154,709-716
Pardridge, W. M. & Davidson, M. B. (1979)Biochim. Biophys. Acta
585,34-42
Pozefsky, T. & Tancredi, R. G. (1972)J.Clin. Invest. 51,2359-2369
Pitts, R. F. (1964)Am. J. Med. 36,720-742
Vol. 8
213
Ross, B. D., Hems, R.. & Krebs, H. A. (1967)Biochem. J. 102, 94295 1
Rowsell, E.V. (1956)Biochem. J. 64,246-252
Rowsell. E. V. & Corbett, K. (1958)Biochem. J. 70,7~
Ruderman, N. B. (1975)Annu. Rev. Med. 26,245-258
Ruderman, N. B. & Berger, M. (1974)J. Biol. Chem. 249,5500-5506
Ruderman, N. B., Ross, P. S., Berger, M. & Goodman, M. N. (1974)
Biochem. J. 138, 1-10
Ruderman, N. B., Schmahl, F. W. & Goodman, M. N. (1977)Am. J.
Physiol. 233,E109-Ell4
Safer, B. & Williamson, J. R. (1973)J.Biol. Chem. 248,2570-2579
Sherwin, R. S., Hendler, R. G. & Felig, P. (1975)J. Clin. Invest. 55,
1382-1 390
Shinnick, F. L. & Harper, A. E. (1977)J.Nutr. 107,887-895
Snell, K. (1976)Biochem. SOC.Trans. 4,287-288
Snell, K. (1975)in Normal and Pathological Development of Energy
Metabolism (Hommes, F. A. & Van den Berg, C. J., eds.), pp. 7795,Academic Press, New York
Snell, K. (1978)Biochem. SOC.Trans. 6,157-159
Snell, K. (1979)Trends Biochem. Sci. 4, 124-128
Snell, K. & Duff, D. A. (19770)Biochem. J. 162,339-403
Snell, K. & Duff, D. A. (19776)Biochem. Biophys. Res. Commun. 77,
925-931
Snell, K. & Duff, D. A. (1979)Znt. J. Biochem. 10,423-426
Spydevold, O., Davis, E. J. & Bremer, J. (1976)Eur. J. Biochem. 71,
155- 165
Stacpoole, P. W. & Felts, J. M. (1970)Metab. Clin. Exp. 19,7 1-78
Sugden, M. C., Sharples, S. C. & Randle, P. J. (1976)Biochem. J. 160,
817-8 19
Taegtmeyer, H., Peterson, M. B., Ragavan, V. V., Ferguson, A. G. &
Lesch, M. (1977)J.Biol. Chem. 252,5010-5018
Taylor, R. T., Shakespeare, V. & Jenkins, W. T. (1970)J. Biol. Chem.
245,4880-4885
Wahren, J., Felig, P., Cerasi, E. & Luft, R. (1972)J. Clin. Invest. 51,
1870-1878
Wahren, J., Felig, P., Havel, R., Jorfeldt, L., Pernow, B. & Saltin, B.
(1973)N. Engl. J. Med. 288, 774-777
Wahren, J., Felig, P. & Hagenfeldt, L. (1976)J. Clin. Invest. 57,987999
Watford, M., Lund, P. & Krebs, H. A. (1979)Biochem. J. 178, 589596
Windmueller, H. G. & Spaeth, A. E. (1974)J. Biol. Chem. 249,50705079
Windmueller, H. G. & Spaeth, A. E. (1975)Arch. Biochem. Biophys.
171,662-672
Windmueller, H. G. & Spaeth, A. E. (1978)J.Biol. Chem. 253,69-76
Wise, J. K., Hendler, R. & Felig, P. (1972)Science 178,5 13-5 14
Wise, J. K., Hendler, R. & Felig, P. (1973a)N. Engl. J. Med. 288,487490
Wise, J. K.,Lyall, S. S., Hendler, R. & Felig, P. (1973b)J.Clin. Endocrinol. Metab. 37,34 1-344