Download Extreme hyperinsulinemia unmasks insulin`s effect to stimulate

Extreme hyperinsulinemia unmasks insulin’s effect to
stimulate protein synthesis in the human forearm
TERESA A. HILLIER, DAVID A. FRYBURG, LINDA A. JAHN, AND EUGENE J. BARRETT
Division of Endocrinology and Metabolism, Department of Internal Medicine, and General Clinical
Research Center, University of Virginia Health Sciences Center, Charlottesville, Virginia 22908
phenylalanine; insulin; proteolysis; amino acids; insulin-like
growth factor I; blood flow
OVER SEVERAL DECADES,
in vitro studies convincingly
demonstrated that insulin stimulates protein synthesis
in isolated skeletal and cardiac muscle (32, 35, 44).
Insulin increases translational efficiency via increased
polysome formation (35), an effect mediated in part via
regulation of the phosphorylation of one or more initiation factors (34, 35, 37). Likewise, in isolated muscle
preparations, insulin inhibits the degradation of muscle
proteins, although the cellular mechanisms responsible
remain undefined (24, 33). On the basis of these
observations, insulin’s combined effects to stimulate
protein synthesis and retard proteolysis were thought
responsible for the protein anabolic action of insulin
seen in vivo (51).
Over the past decade, a conundrum arose as investigators began to examine the action of physiological
concentrations of insulin on whole body and muscle
protein metabolism in vivo. Using tracer methods,
several laboratories reported that physiological doses
of insulin inhibit whole body protein degradation (22,
53). Likewise, combining tracer and regional catheterization methods, most (27, 30, 38, 40, 41) but not all (9)
laboratories have demonstrated that insulin specifically retards muscle proteolysis in humans, findings in
accord with insulin’s effect on proteolysis seen in vitro.
Strikingly, in these studies, insulin did not stimulate
protein synthesis either in the whole body or with one
exception (9) in skeletal muscle. Thus insulin’s protein
anabolic action in vivo appeared to only partially
parallel that which was seen in isolated muscle preparations.
Subsequent studies have probed the origin of this
discrepancy. As insulin in vivo lowers the plasma (and
cellular) concentration of many amino acids, replacement of amino acids during hyperinsulinemia might be
required for expression of insulin action on protein
synthesis. However, neither replacing basal amino acid
concentrations (leucine-clamp method; see Ref. 16) nor
regional insulin infusions (which do not affect plasma
amino acids concentrations; see Ref. 38) unmasked any
effect of insulin on protein synthesis. Combined hyperinsulinemia and hyperaminoacidemia did increase
muscle amino acid uptake and protein synthesis (7, 8,
46). However, as hyperaminoacidemia alone increases
amino acid uptake and protein synthesis in both whole
body and limb balance studies (7, 10, 21, 39, 43 57), any
independent effect of insulin in these studies could not
be ascertained, since amino acid concentrations were
not strictly comparable (7, 46). A subsequent study
controlled for amino acid concentrations by raising
systemic amino acid concentrations twofold and selectively increasing the insulin concentration in one forearm. This did not enhance muscle protein synthesis
more than amino acids alone in the contralateral arm
(21).
Two other potential major differences between in
vitro and in vivo studies are age and insulin concentrations. Several recent studies have demonstrated that
physiological doses of insulin in vivo stimulate skeletal
muscle protein synthesis in young (25) but not mature
or aged rats (3, 11). Likewise, in vitro studies demonstrating an effect of physiological doses of insulin on
muscle protein synthesis have used muscle from young
rats (50). A recent study has indicated that this effect is
blunted or lost in muscle from older animals (17). As
0193-1849/98 $5.00 Copyright r 1998 the American Physiological Society
E1067
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
Hillier, Teresa A., David A. Fryburg, Linda A. Jahn,
and Eugene J. Barrett. Extreme hyperinsulinemia unmasks insulin’s effect to stimulate protein synthesis in the
human forearm. Am. J. Physiol. 274 (Endocrinol. Metab. 37):
E1067–E1074, 1998.—Insulin clearly stimulates skeletal
muscle protein synthesis in vitro. Surprisingly, this effect has
been difficult to reproduce in vivo. As in vitro studies have
typically used much higher insulin concentrations than in
vivo studies, we examined whether these concentration differences could explain the discrepancy between in vitro and in
vivo observations. In 14 healthy volunteers, we raised forearm insulin concentrations 1,000-fold above basal levels
while maintaining euglycemia for 4 h. Amino acids (AA) were
given to either maintain basal arterial (n 5 4) or venous
plasma (n 5 6) AA or increment arterial plasma AA by 100%
(n 5 4) in the forearm. We measured forearm muscle glucose,
lactate, oxygen, phenylalanine balance, and [3H]phenylalanine kinetics at baseline and at 4 h of insulin infusion.
Extreme hyperinsulinemia strongly reversed postabsorptive
muscle’s phenylalanine balance from a net release to an
uptake (P , 0.001). This marked anabolic effect resulted from
a dramatic stimulation of protein synthesis (P , 0.01) and a
modest decline in protein degradation. Furthermore, this
effect was seen even when basal arterial or venous aminoacidemia was maintained. With marked hyperinsulinemia, protein synthesis increased further when plasma AA concentrations were also increased (P , 0.05). Forearm blood flow rose
at least twofold with the combined insulin and AA infusion
(P , 0.01), and this was consistent in all groups. These
results demonstrate an effect of high concentrations of insulin
to markedly stimulate muscle protein synthesis in vivo in
adults, even when AA concentrations are not increased. This
is similar to prior in vitro reports but distinct from physiological hyperinsulinemia in vivo where stimulation of protein
synthesis does not occur. Therefore, the current findings
suggest that the differences in insulin concentrations used in
prior studies may largely explain the previously reported
discrepancy between insulin action on protein synthesis in
adult muscle in vivo vs. in vitro.
E1068
HYPERINSULINEMIA STIMULATES PROTEIN SYNTHESIS IN HUMANS
Fig. 1. Experimental protocol. Subjects were fasted overnight. Each
subject received a primed (,33 mCi), continuous (0.43 mCi/min)
infusion of L-[ring-2,6-3H]phenylalanine. After a 90-min tracer equilibrium period, quadruplicate paired arterial and deep venous samples
were taken over 30 min. After baseline samples were obtained, a
balanced amino acid mixture was continuously infused at a rate of
either 0.004 (n 5 4), 0.007 (n 5 6), or 0.015 (n 5 4) ml · min21 · kg21 for
4 h. Insulin (5 mU · min21 · kg body wt21 ) was continuously infused
intra-arterially. Twenty percent dextrose was infused into the contralateral arm at a variable rate to maintain euglycemia. Quadruplicate
paired blood samples were again taken from each of the sampling
catheters between 210 and 240 min of insulin infusion.
METHODS
Subjects. Fourteen healthy (9 females, 5 males), normalweight (68 6 3 kg), young adult (24 6 2 yr) volunteers were
admitted to the University of Virginia General Clinical
Research Center the evening before study. No subject was
taking any medication, and all female participants had a
negative serum pregnancy test 1–2 days before study. The
study protocol was approved by the University of Virginia
Human Investigation Committee, and each subject gave
written consent.
Experimental protocol. Figure 1 schematically depicts the
experimental protocol. After an overnight 12-h fast, a brachial artery catheter and an ipsilateral, retrograde, median
cubital (deep) vein catheter were placed percutaneously. In
the contralateral arm, a second, antegrade, median cubital
vein catheter was placed for infusion of glucose, amino acids,
and tracer. Each subject received a primed (,33 mCi),
continuous (0.43 mCi/min) infusion of L-[ring-2,6-3H]phenylalanine. After a 90-min tracer equilibration period, quadruplicate paired arterial and deep venous samples were taken over
30 min for measurement of phenylalanine concentration and
specific activity and glucose, lactate, and insulin concentrations. Forearm blood flow was measured after each set of arterial and venous samples by capacitance plethysmography.
After obtaining baseline samples, a balanced amino acid
mixture (10 % Travesol; Clintec Nutrition, Deerfield, IL) was
continuously infused at a rate of either 0.004 (n 5 4: 4
females, 0 males), 0.007 (n 5 6: 4 females, 2 males), or 0.015
(n 5 4: 2 females, 2 males) ml · min21 · kg21 for 4 h. The former
two doses were selected in an effort to maintain phenylalanine concentrations near basal in arterial and venous plasma,
respectively. The higher dose was selected to raise phenylalanine approximately twofold, a level comparable to that studied by us previously during physiological hyperinsulinemia.
In the arm containing the arterial catheter, insulin (5 mU ·
min21 · kg body wt21 ) was continuously infused intra-arterially. The insulin infusion was interrupted for ,30 s every 5
min, and 1 ml of blood was withdrawn for measurement of
plasma glucose. Twenty percent dextrose was infused into the
contralateral arm at a variable rate to maintain the plasma
glucose at fasting concentrations (12). Quadruplicate paired
blood samples were again taken from each of the sampling
catheters between 210 and 240 min of insulin infusion.
Forearm blood flow was measured after each pair of blood
samples.
Analytic methods. Plasma glucose was measured using the
glucose oxidase method. Lactate concentrations were measured by a combined lactate analyzer (Yellow Springs Instruments, Yellow Springs, OH). Blood oxygen content was measured spectrophotometrically using an OSM2 hemoximeter
(Radiometer, Copenhagen, Denmark). Plasma insulin was
measured by radioimmunoassay. Phenylalanine concentration and specific activity were measured as previously described (6).
Calculations of forearm phenylalanine kinetics. The net
forearm balance for amino acids was calculated from the Fick
principle
Net balance 5 ([A] 2 [V]) 3 F
(1)
where [A] and [V] are arterial and venous substrate concentrations, respectively, and F is forearm blood flow in milliliters
per minute per 100 ml forearm volume. Measurement of the
absolute rates of synthesis and breakdown of muscle protein
requires knowing the phenylalanine specific activity in the
phenylalanyl-tRNA pool being used for protein synthesis.
This is not experimentally accessible in the forearm. The
rates of protein synthesis and degradation can be estimated
from the kinetics of exchange of labeled phenylalanine across
the forearm as previously described (6, 27). Equations have
been presented for estimating forearm protein synthesis and
degradation that were based on use of either arterial or deep
venous specific activity of [3H]phenylalanine to approximate
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
virtually all human studies have been performed in
adults, the age of the subjects may contribute to the
generally reported lack of effect of physiological insulin
to stimulate protein synthesis.
A second potential major difference between in vitro
and in vivo studies is the insulin concentrations used.
Most in vitro studies demonstrating stimulation of
protein synthesis in adult animals have used insulin
concentrations of 2 mU/ml or above (32–34, 42). In
contrast, much lower concentrations (e.g., 10–200 µU/
ml) have generally been used in vivo (27, 30, 38, 40, 41).
To further resolve this conundrum, we examined
whether insulin at concentrations previously shown to
stimulate protein synthesis in adult muscle in vitro
exerted similar effects in vivo.
In the current study, we infused insulin (5 mU ·
min21 · kg21 ) into the brachial artery to raise local
forearm insulin concentrations by ,1,000-fold. This
insulin infusion lowers circulating amino acids (23), an
effect not seen in vitro in perfused tissues. We prevented hypoaminoacidemia by simultaneous systemic
infusion of a balanced amino acid mixture at one of
three rates selected to either replace or augment plasma
amino acids and then examined the separate effect of
amino acid supply on the response to marked hyperinsulinemia.
E1069
HYPERINSULINEMIA STIMULATES PROTEIN SYNTHESIS IN HUMANS
phenylalanyl-tRNA specific activity. Because recent data
suggest that the specific activity of phenylalanine in venous
plasma more closely approximates that of the tRNA pool (56,
59), we used venous specific activity to estimate rates of
protein synthesis and degradation. Qualitatively comparable
results were obtained using the arterial phenylalanine specific activity as the precursor pool (see below). With the use of
the venous phenylalanine specific activity to reflect the
precursor pool for protein synthesis, synthesis (S) is given by
S 5 ([DPMart 2 DPMvein] 3 F)/SAvein
(2)
where DPMart and DPMvein are disintegrations per minute in
arterial and venous concentrations, respectively, and SAvein is
the specific activity in vein. Muscle protein breakdown (B) is
calculated as
B 5 S 2 net balance
(3)
RESULTS
Forearm glucose, lactate, phenylalanine, and insulin
concentrations. In the three study groups, the arterial
concentration of glucose averaged 4.8 6 0.2 mM in the
basal period and remained at or slightly above basal
levels throughout the 4-h experimental period (mean 5
5.3 6 0.2 mM over last 30 min). The arterial lactate
concentration (measured in subjects given 0.007 ml of
amino acid · min21 · kg21 ) rose modestly during the 4-h
insulin infusion (0.6 6 0.1 mM basal vs. 1.0 6 0.2 mM
with insulin infusion). Infusion of amino acids at 0.004
ml amino acid · min21 · kg21 maintained arterial phenylalanine at basal values; however, the venous concentration declined ,20% (Table 1). The 0.007 ml amino acid ·
min21 · kg21 infusion raised the arterial concentrations
,35% above basal level, while venous concentrations
were not different from basal (Table 1). The highest
Table 1. Concentration of insulin in the forearm vein
and of phenylalanine in the forearm artery and vein
basally (230 to 0 min) and at the end of the insulin
infusion (210–240 min)
Basal
Insulin, µU/ml
0.004 ml · min21 · kg AA21
0.007 ml · min21 · kg AA21
0.015 ml · min21 · kg AA21
Arterial phenylalanine, mM
0.004 ml · min21 · kg AA21
0.007 ml · min21 · kg AA21
0.015 ml · min21 · kg AA21
Venous phenylalanine, mM
0.004 ml · min21 · kg AA21
0.007 ml · min21 · kg AA21
0.015 ml · min21 · kg AA21
Insulin
962
762
961
8,600 6 1,700†
7,100 6 1,600†
11,300 6 5,500†
36 6 1
47 6 3
36 6 4
35 6 1
64 6 4†
80 6 7†
41 6 2
52 6 3
40 6 5
33 6 1†
58 6 5
73 6 9†
Values are means 6 SE. AA, amino acid. † P , 0.01 for basal vs.
insulin infusion period.
Basal
ml · min21 · 100
ml21
Blood flow,
0.004 ml · min21 · kg AA21
0.007 ml · min21 · kg AA21
0.015 ml · min21 · kg AA21
Glucose balance, µmol · min21 · 100 ml21
0.004 ml · min21 · kg AA21
0.007 ml · min21 · kg AA21
0.015 ml · min21 · kg AA21
Lactate balance, µmol · min21 · 100 ml21
0.004 ml · min21 · kg AA21
0.007 ml · min21 · kg AA21
0.015 ml · min21 · kg AA21
Oxygen balance, µmol · min21 · 100 ml21
0.004 ml · min21 · kg AA21
0.007 ml · min21 · kg AA21
0.015 ml · min21 · kg AA21
Insulin
3.2 6 1.3
3.0 6 0.3
4.7 6 0.8
8.2 6 1.3
7.4 6 1.7*
9.4 6 3.0
0.3 6 0.1
1.0 6 0.3
0.9 6 0.3
7.0 6 2.0*
7.7 6 1.8†
12.7 6 3.8*
ND
20.3 6 0.1
ND
ND
21.7 6 0.5*
ND
7.6 6 1.4
7.1 6 0.5
ND
12.1 6 1.8*
11.6 6 0.9†
ND
Values are means 6 SE. ND, not determined. * P , 0.05 and † P ,
0.01 for basal vs. insulin infusion period.
amino acid dose (0.015 ml · min21 · kg21 ) significantly
raised both arterial and venous phenylalanine (Table
1). During the basal period, the venous plasma insulin
concentration averaged between 7 and 9 µU/ml (Table
1). During the experimental period, forearm concentrations rose nearly 1,000-fold (average 7,100–11,300 µU/
ml) and remained elevated for 4 h (Table 1). Euglycemia was maintained throughout and during the last 60
min of the insulin clamp. Glucose was infused at an
average rate of 10.7 6 0.9 mg · min21 · kg21. There was
no significant difference in the rate of glucose infusion
in the three groups (9.6 6 0.6, 12.2 6 2.0, and 9.6 6 1.4
mg · min21 · kg21 for the 0.004, 0.007, and 0.015 ml ·
min21 · kg21 amino acid groups, respectively).
Forearm glucose, lactate, and oxygen balances and
blood flows. Forearm glucose uptake significantly increased in each study group during the experimental
period (Table 2). Forearm lactate release increased in
each of the subjects in whom it was measured. Blood
flow in the insulin-infused arm increased markedly in
the three study groups (average 230%, Table 2). In
response to the high-dose insulin, forearm oxygen
uptake increased by 158 and 162% in subjects given
0.004 and 0.007 ml · min21 · kg21 amino acid, respectively (Table 2).
Forearm phenylalanine kinetics. There was a net
release of phenylalanine by the forearm in each of the
three groups at baseline (Fig. 2). Phenylalanine’s only
metabolic fate in muscle is its incorporation into and
release from muscle protein, as it is not concentrated by
muscle (i.e., intracellular and extracellular concentrations are nearly equal; see Ref. 8). Therefore, the net
postabsorptive release of phenylalanine indicates a net
loss of muscle protein.
Phenylalanine balance across the arm converted
from a net release to a net uptake in each group during
the experimental period (Fig. 2). Protein synthesis rose
significantly in each of the three study groups (Fig. 2).
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
Data presentation and statistical analysis. All data are
presented as means 6 SE. Data on hormone and substrate
concentrations, forearm substrate and oxygen balances, and
forearm amino acid kinetics are presented for each of the
sample periods, i.e., basal and 4 h. Stochastic comparisons
were made using analysis of variance with repeated measures and post hoc comparisons with Duncan’s test (True
Epistat; Epistat Services, Richardson, TX).
Table 2. Forearm blood flow and the balances
for glucose, lactate, and oxygen across forearm muscle
in the basal period (230 to 0 min) and over the last 30
min of insulin infusion (210–240 min)
E1070
HYPERINSULINEMIA STIMULATES PROTEIN SYNTHESIS IN HUMANS
pearance (like S, an index for protein synthesis) increased with insulin at each amino acid infusion dose
(P , 0.02, 0.01, and 0.01 for the 0.004, 0.007, and 0.015
ml·min21 ·kg21 amino acid infusions, respectively). Thus
extreme concentrations of insulin acted comparably
irrespective of the precursor pool model used for tracing
protein synthesis.
DISCUSSION
Fig. 2. Protein kinetics. Protein balance, degradation, and synthesis
rates are shown (measured in nmol phenylalanine · min21 · 100 ml21 ).
Values represent means 6 SE for the basal (open bars) and last 30
min of the insulin infusion (filled bars) periods with the 3 different
rates of amino acid infusion (in ml · min21 · kg21) (* P , 0.05 and ** P ,
0.01 for basal vs. infusion period).
In the entire group of 14 subjects, there was a positive
linear correlation (P , 0.05) between both the arterial
(Fig. 3) and venous phenylalanine concentrations and
the rate of protein synthesis (r 5 0.63 and 0.58,
respectively). Phenylalanine release from muscle protein degradation was similar during the basal period in
each group. Degradation appeared to decline in each
group during the insulin infusion, but the change was
only statistically significant when the data from all
subjects were pooled (54 6 8 vs. 45 6 5 nmol · min21 · 100
ml21, P , 0.05).
The kinetics of forearm protein turnover were also
analyzed using arterial phenylalanine as the index pool
for tracing protein turnover (see METHODS, data not
shown). Use of the arterial phenylalanine specific activity as the precursor pool for estimating muscle protein
synthesis and degradation yielded results comparable
to those presented above based on use of the venous
precursor pool. Specifically, phenylalanine rate of disap-
Fig. 3. Protein synthesis. Regression line (solid line) for the current
study demonstrates that protein synthesis rates are positively correlated with increasing phenylalanine (PHE) concentrations (r 5 0.63,
P , 0.02). Mean 6 SE values are depicted for both variables in each of
the 3 amino acid infusion groups along the regression line. For
comparison, we have graphed (dashed line) protein synthesis rates
with physiological insulin (,80 µU/ml) during euaminoacidemia (38)
and hyperaminoacidemia (21).
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
In this study, the very high insulin concentrations
previously shown to stimulate adult skeletal muscle
protein synthesis in vitro provoked a similar action in
the adult human forearm in vivo. Protein synthesis was
strongly stimulated when amino acids were infused at
any of the three doses tested. At the lowest dose
infused, arterial amino acid concentrations were not
changed, but venous amino acid concentrations fell.
The latter mimics the cellular concentrations (14),
which might also be expected to fall. Despite this,
protein synthesis rates doubled (Table 2 and Fig. 2).
This contrasts to findings with physiological hyperinsulinemia and euaminoacidemia, which leaves forearm
protein synthesis unchanged or even decreased (38,
41). High concentrations of insulin (.1,000 µU/ml)
were also achieved in the study by Denne and colleagues (15). They observed that leg muscle proteolysis
declined but saw no increase in protein synthesis.
However, plasma amino acid concentrations were not
maintained in that study, and the decline in amino acid
concentrations with hyperinsulinemia may have obscured an effect.
HYPERINSULINEMIA STIMULATES PROTEIN SYNTHESIS IN HUMANS
forearm during systemic amino acid infusion did not
further stimulate muscle protein synthesis in the insulin-infused arm (21). The amino acid infusion rate
(0.015 ml · min21 · kg21 ) in that study was the same as
the highest dose used in the present study, and the
arterial phenylalanine concentrations were higher than
in the present study (113 6 4 vs. 80 6 7 µM; see Fig. 3).
Therefore, the stimulation of protein synthesis to 90–
150% above basal rates with marked hyperinsulinemia
in the present study both under basal and hyperaminoacidemic conditions suggests that marked hyperinsulinemia has an effect separate from that of amino acids.
Protein degradation declined modestly in each of the
three study groups in the current study (Fig. 2). This
change (,20% decline overall) was statistically significant when observations for all three groups were
pooled. In the previous human forearm studies, insulin
was found to retard proteolysis by 25–40%. As proteolysis is estimated from the dilution of phenylalanine
specific activity across the muscle bed, the high blood
flows (vide infra) seen during the very high insulin
infusion used here likely added variability to this
measurement. That protein synthesis, the primary
outcome variable, and net phenylalanine balance were
significantly stimulated in each of the three study
groups attests to a quantitatively greater (,100%)
effect on synthesis with the use of high insulin concentrations.
Although the relative change in protein kinetics was
similar in all three groups in the current study, the
absolute rates of basal protein synthesis and degradation were lower in the 0.004 ml · min21 · kg21 group.
Gender differences could partially explain this, as all
four subjects were female in the 0.004 group, whereas
the 0.007 and 0.015 ml · min21 · kg21 groups had a mix of
male and female subjects.1 However, each subject was
compared with his/her own baseline value, and a
consistent and significant effect on protein synthesis
was still observed in all three groups. If anything, our
results may underestimate the effect of marked hyperinsulinemia to stimulate protein synthesis with euaminoacidemic conditions because of the large proportion
of females in two of the groups.
It has been suggested that a stimulation of protein
synthesis in vivo has been difficult to demonstrate in
humans because basal insulin has already maximally
stimulated protein synthesis. Three observations suggest that this may not be the case. First, in the current
study, very high insulin concentrations do further
stimulate protein synthesis. Second, as infusion of
growth hormone (20) or of insulin-like growth factor I
(IGF-I; see Refs. 19 and 21) acutely increases forearm
muscle protein synthesis, synthesis rates are clearly
not at a maximum. Third, in studies of insulin-deficient
diabetic humans, acute replacement of insulin did not
increase whole body or muscle protein synthesis (44,
1 In 76 healthy postabsorptive adult subjects (40 males, 36 females), protein synthesis rates averaged 24 6 2 nmol phenylalanine ·
min21 · 100 ml21 in females and 36 6 3 in males (P , 0.005; see Ref.
31a).
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
In the present study, steady-state extreme hyperinsulinemia shifted forearm muscle phenylalanine balance
from negative to strongly positive. This shift is more
marked than reported with physiological hyperinsulinemia in human forearm or leg muscle, where insulin’s
inhibition of protein degradation accounted for the
entire effect (21, 30, 38, 40, 41). In the forearm, raising
insulin to four different levels throughout the physiological range (20–125 µU/ml) during euaminoacidemia
shifted phenylalanine balance to modestly positive
values (,10 nmol · min21 · 100 ml21 ) in all groups, suggesting a plateau effect of physiological hyperinsulinemia to retard proteolysis (38). The greater shift in the
current study (even in groups 1 or 2 when plasma
amino acid concentrations remained at basal) was
primarily due to the marked stimulation of protein
synthesis that was not previously seen at physiological
insulin concentrations.
In only one human study was muscle protein synthesis reported to increase in response to physiological
concentrations of insulin (9). In that study, protein
synthesis was measured in leg muscle of six subjects,
using both an arterial-venous difference method (as
employed here) as well as a tissue biopsy technique.
Good internal agreement was obtained between results
seen with the arterial-venous difference method and
the biopsy technique, both showing an enhancement of
muscle protein synthesis by physiological hyperinsulinemia. Contrary to previous (6, 15, 30, 38, 41) or
subsequent reports (21), the authors found no evidence
for an effect of insulin to restrain proteolysis. No
previous human limb balance study (including more
than 60 subjects) has seen a stimulation of muscle
protein synthesis by physiological hyperinsulinemia
using the arterial-venous difference method (21, 27, 38,
41). Most have seen an effect of insulin on whole body
and muscle proteolysis. In addition, other studies in
both humans (40) and rodents (59) using muscle biopsies and estimating muscle protein synthesis using a
variety of methods (including determining the labeling
of the aminoacyl-tRNA) have also failed to show an
effect of physiological hyperinsulinemia. Thus the results of Biolo and colleagues (9), although internally
consistent, are divergent with all other in vivo studies,
and we can not currently reconcile their unique findings.
The increases in protein synthesis seen with the
higher rates of amino acid infusion (Figs. 2 and 3)
indicate that raising plasma amino acid concentrations
alone augments protein synthesis. This is in agreement
with several previous human studies showing increased rates of whole body (10, 28) or skeletal muscle
(7, 8, 46) protein synthesis in response to raised amino
acid concentrations with or without added insulin. As
these studies augmented amino acid concentrations
simultaneously with physiological hyperinsulinemia, it
was not possible to assess whether insulin had an
independent effect from that of amino acids. We subsequently addressed this experimentally using a double
forearm cannulation method. We observed that superimposing local, physiological hyperinsulinemia in one
E1071
E1072
HYPERINSULINEMIA STIMULATES PROTEIN SYNTHESIS IN HUMANS
required. An exception is in young rats in which in vivo
studies have demonstrated a stimulatory action of
physiological insulin concentrations on muscle protein
synthesis (25, 26). However, these effects are not seen
in older animals (3, 11, 39) or, with one exception (9), in
adult humans. Thus the muscle’s sensitivity to insulin’s
stimulatory action on protein synthesis may decline
with aging.
In the current study, the effects of extreme hyperinsulinemia to stimulate protein synthesis, enhance oxygen
consumption, markedly increase blood flow, and induce
a strongly positive phenylalanine balance are each
similar to the actions of locally infused IGF-I in human
forearm (19, 21) and differ from the effects seen with
physiological hyperinsulinemia concentrations (27, 30,
38, 41). As the insulin concentrations achieved in the
current study were 4–7 3 1028 M (100-fold above the
physiological range), stimulation of the IGF-I receptor
by insulin is entirely plausible [dissociation constant
(Kd ) of insulin for the IGF-I receptor ,1028 compared
with Kd 10210 for IGF-I (58)]. Furthermore, as the
mitogenic effect of insulin (1026 ) in vitro is mediated via
the IGF-I receptor rather than the insulin receptor (55),
an analogous effect with protein synthesis is possible. If
this is the case, our results introduce a necessary
caution to the interpretation of the many in vitro
studies documenting an effect of high concentrations of
insulin to increase bulk protein synthesis in adult
animals (see Refs. 31 and 35 for reviews). With few
exceptions (18, 50, 51), the insulin concentrations used
in these studies (generally $2 mU/ml) are in a range in
which effects of insulin mediated by the IGF-I receptor
or hybrid IGF-I/insulin receptors (48, 49) could complicate data interpretation. As the current study’s aim
was only to determine if extreme hyperinsulinemia
could reproduce in vivo the stimulation of protein
synthesis seen with similar concentrations in vitro,
further research is needed to elucidate if protein synthesis is indeed being stimulated via IGF-I signaling
pathways.
In summary, these results indicate that insulin at
high concentrations strongly stimulates muscle protein
synthesis in the human forearm. This effect is quite
consistent with the action of insulin described in a
number of in vitro studies using similar concentrations
of insulin but distinct from what is observed with
physiological hyperinsulinemia. Therefore, much of the
discrepancy previously reported between insulin action
on protein synthesis in vivo vs. in vitro may result
directly from differences in insulin concentrations used.
In addition to stimulating protein synthesis, high-dose
insulin resembles the action of IGF-I observed previously (19, 21). As IGF-I receptors can be stimulated by
high concentrations of insulin, the present results
together with findings from in vitro studies raise the
possibility that some or all of insulin’s action to stimulate protein synthesis may be mediated by pathways
other than the insulin receptor.
This study was supported by National Institutes of Health Grants
AR-01881, DK-38578, and RR-00847 to the University of Virginia
General Clinical Research Center.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
47). Collectively, these data suggest that physiological
increments in insulin do not augment bulk protein
synthesis in humans.
In insulin-withdrawn streptozotocin diabetic rats,
replacement of insulin rapidly restores protein synthetic rates in both heart and skeletal muscle (1, 2).
Before treatment, these animals are severely catabolic
(insulin withdrawn for up to 5 days), and it is quite
possible that restoring basal insulin does affect protein
synthesis in this setting. This may be a direct effect of
insulin or an indirect effect achieved by restoring
sensitivity to other growth factors such as IGF-I (31).
Marked hyperinsulinemia combined with amino acid
infusion stimulated blood flow more than twofold (100–
160%) in each of the three groups in the current study.
Both insulin (4) and amino acids likely contributed to
this, as generalized hyperaminoacidemia (21) or infusion of arginine (29) increases forearm blood flow.
Importantly, physiological hyperinsulinemia in the presence of hyperaminoacidemia does not stimulate blood
flow more than hyperaminoacidemia alone (21).
The reported stimulation of limb blood flow with
physiological hyperinsulinemia ranges from 10 to 125%
(5, 13, 30, 36, 38, 45, 52, 54). This variability involves
many factors such as insulin infusion (local vs. systemic and single dose vs. sequential increments in
insulin infusion dose), methodology (capacitance plethysmography vs. thermodilution used as well as variability between laboratories in employing a particular
method), locale (forearm vs. leg), and muscle content of
the limb studied (4, 54). Most studies reporting a
stimulation of blood flow above 50% with physiological
hyperinsulinemia have used a sequential increase of
insulin infusion in the same subject (5, 36, 52).
As our previous studies have extensively used the
same procedure of capacitance plethysmography in the
forearm, it seems most logical to contrast our current
blood flow results with our past results. We previously
observed that insulin at five different concentrations
within the physiological range did not stimulate blood
flow .25%, and this was only in the high physiological
range (27). Additionally, we studied the effects of hyperaminoacidemia (0.015 ml · kg21 · min21 balanced
amino acid infusion) with or without insulin on forearm
flow. Physiological hyperinsulinemia did not stimulate
blood flow beyond hyperaminoacidemia alone. Of interest, in that study and several others, we noted that
local IGF-I strongly stimulated flow when amino acids
were elevated or remained at basal levels (21). Thus the
two- to threefold stimulation of blood flow in the current study with extreme hyperinsulinemia even when
amino acids remained at basal levels is unlike physiological hyperinsulinemia and similar to flow changes
provoked by local forearm IGF-I infusion (19, 21).
The stimulation of protein synthesis in the current
study is consistent with previous in vitro studies of
isolated muscle incubated with insulin at high concentrations. As comparable effects of insulin have generally not been seen in vivo with physiological insulin
concentrations (27, 30, 38, 41, 40, 59), our results
suggest that very high insulin concentrations may be
HYPERINSULINEMIA STIMULATES PROTEIN SYNTHESIS IN HUMANS
This work was presented in abstract form as part of the 56th
Annual Meeting and Scientific Sessions of the American Diabetes
Association, San Francisco, CA, on June 8–11, 1996.
Address for reprint requests: T. Hillier, Univ. of Virginia Health
Science Center, Box 5116, MR4, Charlottesville, VA 22908.
Received 30 September 1997; accepted in final form 19 February
1998.
REFERENCES
advancing age. J. Gerontol. B Psychol. Sci. Soc. Sci. 51: B323–
B330, 1996.
18. Frayn, K. N., and P. F. Maycock. Regulation of protein
metabolism by a physiological concentration of insulin in mouse
soleus and extensor digitorum longus muscles. Biochem. J. 184:
323–330, 1979.
19. Fryburg, D. A. Insulin-like growth factor I exerts growth
hormone- and insulin-like actions on human muscle protein
metabolism. Am. J. Physiol. 267 (Endocrinol. Metab. 30): E331–
E336, 1994.
20. Fryburg, D. A., R. A. Gelfand, and E. J. Barrett. Growth
hormone acutely stimulates muscle protein synthesis in normal
humans. Am. J. Physiol. 260 (Endocrinol. Metab. 23): E499–
E504, 1991.
21. Fryburg, D. A., L. A. Jahn, S. A. Hill, D. M. Oliveras, and
E. J. Barrett. Insulin and insulin-like growth factor-I enhance
human skeletal muscle protein anabolism during hyperaminoacidemia by different mechanisms. J. Clin. Invest. 96 1722–1729,
1995.
22. Fukagawa, N. K., K. L. Minaker, J. W. Rowe, M. N. Googman, D. W. Matthews, D. M. Bier, and V. R. Young. Insulinmediated reduction of whole body protein breakdown: doseresponse effects on leucine metabolism in postabsorptive man. J.
Clin. Invest. 76: 2306–2311, 1985.
23. Fukagawa, N. K., K. L. Minaker, V. R. Young, and J. W.
Rowe. Insulin dose-dependent reductions in plasma amino acids
in man. Am. J. Physiol. 250 (Endocrinol. Metab. 13): E13–E17,
1986.
24. Fulks, R. M., J. B. Li, and A. L. Goldberg. Effects of insulin,
glucose, amino acids on protein turnover in rat diaphragm. J.
Biol. Chem. 250: 290–298, 1975.
25. Garlick, P. J., and I. Grant. Amino acid infusion increases the
sensitivity of muscle protein synthesis in vivo to insulin. Biochem. J. 254: 579–584, 1988.
26. Garlick, P. J., V. R. Preedy, and P. J. Reeds. Regulation of
protein turnover in vivo by insulin and amino acids. Intracellular
Protein Catabolism, edited by E. A. Khairallah, J. S. Bond, and
J. W. C. Bird. New York: Liss, 1985, p. 555–564.
27. Gelfand, R. A., and E. J. Barrett. Effect of physiologic hyperinsulinemia on skeletal muscle protein synthesis and breakdown
in man. J. Clin. Invest. 80: 1–6, 1987.
28. Gelfand, R. A., M. G. Glickman, R. Jacob, R. S. Sherwin,
and R. A. DeFronzo. Removal of infused amino acids by
splanchnic leg tissues in man. Am. J. Physiol. 250 (Endocrinol.
Metab. 13): E407–E413, 1986.
29. Giugliano, D., R. Marfella, G. Verrazzo, R. Acampora, L.
Coppola, D. Cozzolino, and F. D’Onofrio. The vascular effects
of L-arginine in humans: the role of endogenous insulin. J. Clin.
Invest. 99: 433–438, 1997.
30. Heslin, M. J., E. Newman, R. F. Wolf, P. W. T. Pisters, and
M. F. Brennan. Effect of hyperinsulinemia on whole body and
skeletal muscle leucine carbon kinetics in humans. Am. J.
Physiol. 262 (Endocrinol. Metab. 25): E911–E918, 1992.
31. Jacob, R., X. Hu, D. Niederstock, S. Hasan, P. H. McNulty,
R. S. Sherwin, and L. H. Young. IGF-I stimulation of muscle
protein synthesis in the awake rat: permissive role of insulin and
amino acids. Am. J. Physiol. 270 (Endocrinol. Metab. 33): E60–
E66, 1996.
31a.Jahn, L. A., E. J. Barrett, T. Spraggins, M. Genco, L. Im,
and D. A. Fryburg. Effect of gender on forearm muscle metabolism (Abstract). Med. Sci. Sports Exercise, Suppl. 5: 93, 1997.
32. Jefferson, L. S., J. O. Koehler, and H. E. Morgan. Effect of
insulin on protein synthesis in skeletal muscle of an isolated
perfused preparation of rat hemicorpus. Proc. Natl. Acad. Sci.
USA 69: 816–820, 1972.
33. Jefferson, L. S., J. B. Li, and S. R. Rannels. Regulation by
insulin of amino acid release and protein turnover in the
perfused rat hemicorpus. J. Biol. Chem. 252: 1476–1483, 1977.
34. Karinch, A. M., S. R. Kimball, T. C. Vary, and L. S. Jefferson.
Regulation of eukaryotic initiation factor-2B activity in muscle of
diabetic rats. Am. J. Physiol. 264 (Endocrinol. Metab. 27):
E101–E108, 1993.
35. Kimball, S. R., T. C. Vary, and L. S. Jefferson. Regulation of
protein synthesis by insulin. Annu. Rev. Physiol. 56: 321–348,
1994.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
1. Ashford, A. J., and V. M. Pain. Effect of diabetes on the rates of
synthesis and degradation of ribosomes in rat muscle and liver in
vivo. J. Biol. Chem. 261: 4059–4065, 1986.
2. Ashford, A. J., and V. M. Pain. Insulin stimulation of growth in
diabetic rats: synthesis and degradation of ribosomes and total
tissue protein in skeletal muscle and heart. J. Biol. Chem. 261:
4066–4070, 1986.
3. Baillie, A. G. S., and P. J. Garlick. Attenuated responses of
muscle protein synthesis to fasting and insulin in adult female
rats. Am. J. Physiol. 262 (Endocrinol. Metab. 25): E1–E5, 1992.
4. Baron, A. D. Hemodynamic actions of insulin. Am. J. Physiol.
267 (Endocrinol. Metab. 30): E187–E202, 1994.
5. Baron, A. D., H. O. Steinberg, H. Chaker, R. Leaming, A.
Johnson, and G. Brechtel. Insulin-mediated vasodilation contributes to both insulin sensitivity and responsiveness in lean
humans. J. Clin. Invest. 96: 786–792, 1995.
6. Barrett, E. J., J. H. Revkin, L. H. Young, B. L. Zaret, R.
Jacob, and R. A. Gelfand. An isotopic method for in vivo
measurement of muscle protein synthesis and degradation.
Biochem. J. 245: 223–228, 1987.
7. Bennet, W. M., A. A. Connacher, C. M. Scrimgeour, R. T.
Jung, and M. J. Rennie. Euglycemic hyperinsulinemia augments amino acid uptake by leg tissues during hyperaminoacidemia. Am. J. Physiol. 259 (Endocrinol. Metab. 22): E185–E194,
1990.
8. Bennet, W. M., A. A. Connacher, C. M. Scrimgeour, and M. J.
Rennie. The effect of amino-acid infusion on leg protein turnover
assessed by L-[15N]phenylalanine and L-[1-13C]leucine exchange.
Eur. J. Clin. Invest. 20: 37–46, 1990.
9. Biolo, G., R. Y. D. Fleming, and R. R. Wolfe. Physiologic
hyperinsulinemia stimulates protein synthesis and enhances
transport of selected amino acids in human skeletal muscle. J.
Clin. Invest. 95: 811–819, 1995.
10. Castellino, P., L. Luzi, D. C. Simonson, M. Haymond, and
R. A. DeFronzo. Effect of insulin and plasma amino acid
concentrations on leucine metabolism in man. J. Clin. Invest. 80:
1784–1793, 1987.
11. Dardevet, D., C. Sornet, D. Attaix, V. E. Baracos, and J.
Grizard. Insulin-like growth factor-I and insulin resistance in
skeletal muscle of adult and old rats. Endocrinology 134: 1475–
1484, 1994.
12. DeFronzo, R. A., J. D. Tobin, and R. Andres. Glucose clamp
technique, a method for quantifying insulin secretion and resistance. Am. J. Physiol. 237 (Endocrinol. Metab. Gastrointest.
Physiol. 6): E214–E223, 1979.
13. De Haan, C. H. A., F. M. H. van Dielen, A. J. H. M. Houben,
P. W. de Leeuw, F. C. Huvers, J. G. R. De Mey, B. H. R.
Wolffenbuttel, and N. C. Schaper. Peripheral blood flow and
noradrenaline responsiveness: the effect of physiologic hyperinsulinemia. Cardiovasc. Res. 34: 192–198, 1997.
14. Del Prato, S., R. DeFronzo, P. Castellino, J. Wahren, and A.
Alvestrand. Regulation of amino acid metabolism by epinephrine. Am. J. Physiol. 258 (Endocrinol. Metab. 21): E878–E887,
1990.
15. Denne, S. C., E. A. Liechty, Y. M. Liu, G. Brechtel, and A. D.
Baron. Proteolysis in skeletal muscle and whole body in response to euglycemic hyperinsulinemia in normal adults. Am. J.
Physiol. 261 (Endocrinol. Metab. 24): E809–E814, 1991.
16. Flakoll, P. J., M. Kulaylat, M. Frexes-Steed, H. Hourani,
L. L. Brown, J. O. Hill, and N. N. Abumrad. Amino acids
augment insulin’s suppression of whole body proteolysis. Am. J.
Physiol. 257 (Endocrinol. Metab. 20): E839–E847, 1989.
17. Fluckey, J. D., T. C. Vary, L. S. Jefferson, W. J. Evans, and
P. A. Farrell. Insulin stimulation of protein synthesis in rat
skeletal muscle following resistance exercise is maintained with
E1073
E1074
HYPERINSULINEMIA STIMULATES PROTEIN SYNTHESIS IN HUMANS
48. Pessin, J. E., and A. L. Frattali. Molecular dynamics of
insulin/IGF-I receptor transmembrane signaling. Mol. Reprod.
Dev. 35: 339–345, 1993.
49. Pessin, J. E., and A. L. Frattali. Structure-function properties
of insulin/IGF-I hybrid receptors. In: Molecular Biology of Diabetes, edited by D. LeRoith and B. Draznin. Totowa, NJ: Humana,
1994, pt. II, p. 413–426.
50. Stirewalt, W. S., R. B. Low, and J. M. Slaiby. Insulin
sensitivity and responsiveness of epitrochlearis and soleus
muscles from fed and starved rats. Biochem. J. 227: 355–362,
1985.
51. Sugden, P. H., and S. J. Fuller. Regulation of protein turnover
in skeletal and cardiac muscle. Biochem. J. 273: 21–37, 1991.
52. Tack, C. J. J., A. E. P. Schefman, J. L. Willems, T. Thien, J. A.
Lutterman, and P. Smits. Direct vasodilator effects of physiological hyperinsulinemia in human skeletal muscle. Eur. J.
Clin. Invest. 26: 772–778, 1996.
53. Tessari, P., R. Trevisan, S. Inchiostro, G. Bioli, R. Nosadini,
S. V. De Kreutzenberg, E. Duner, A. Tiengo, and G. Crepaldi. Dose-response curves of effects of insulin on leucine
kinetics in humans. Am. J. Physiol. 251 (Endocrinol. Metab. 14):
E334–E342, 1986.
54. Utriainen, T., R. Malmstrom, S. Makimattila, and H. YkiJarvinen. Methodological aspects, dose-response characteristics and causes of interindividual variation in insulin stimulation
of limb blood flow in normal subjects. Diabetologia 38: 555–564,
1995.
55. Van Wyk, J., D. Graves, S. Casella, and S. Jacobs. Evidence
from monoclonal antibody studies that insulin stimulates deoxyribonucleic acid synthesis through the type I somatomedin
receptor. J. Clin. Endocrinol. Metab. 61: 639–643, 1985.
56. Watt, P., Y. Lindsay, C. Scrimgeour, P. Chien, J. Gibson, D.
Taylor, and M. Rennie. Isolation of aminoacyl-tRNA and its
labeling with stable-isotope tracers, use in studies of human
tissue protein synthesis. Proc. Natl. Acad. Sci. USA 88: 5892–
5896, 1991.
57. Watt, P. W., M. E. Corbett, and M. J. Rennie. Stimulation of
protein synthesis in pig skeletal muscle by infusion of amino
acids during constant insulin availability. Am. J. Physiol. 263
(Endocrinol. Metab. 26): E453–E460, 1992.
58. Werner, H., D. Beitner-Johnson, C. T. Roberts, and D.
LeRoith. Molecular comparisons of the insulin and IGF-I receptors. In: Molecular Biology of Diabetes, edited by B. Draznin and
D. LeRoith. Totowa, NJ: Humana, 1994, pt. II, p. 377–392.
59. Young, L. H., W. Stirewalt, P. H. McNulty, J. H. Revkin, and
E. J. Barrett. Effect of insulin on rat heart and skeletal muscle
phenylalanyl-tRNA labeling and protein synthesis in vivo. Am. J.
Physiol. 267 (Endocrinol. Metab. 30): E337–E342, 1994.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.2 on May 9, 2017
36. Laasko, M., S. V. Edelman, G. Brechtel, and A. D. Baron.
Decreased effect of insulin to stimulate skeletal muscle blood
flow in obese man. J. Clin. Invest. 85: 1844–1852, 1990.
37. Lin, T. A., X. Kong, T. A. J. Haystead, P. Arnim, G. Belsham,
N. Sonenberg, and J. C. Lawrence. PHAS-1 as a link between
mitogen-activated protein kinase and translation initiation. Science 266: 653–656, 1994.
38. Louard, R. J., D. A. Fryburg, R. A. Gelfand, and E. J.
Barrett. Insulin sensitivity of protein and glucose metabolism in
human forearm skeletal muscle. J. Clin. Invest. 90: 2348–2354,
1992.
39. McNulty, P. H., L. H. Young, and E. J. Barrett. Response of rat
heart and skeletal muscle protein in vivo to insulin and amino
acid infusion. Am. J. Physiol. 264 (Endocrinol. Metab. 27):
E958–E965, 1993.
40. McNurlan, M. A., P. Essen, A. Thorell, A. G. Calder, S. E.
Anderson, O. Ljungquist, A. Sandgren, I. Grant, I. Thader,
P. E. Ballmer, J. Wernerman, and P. J. Garlick. Response of
protein synthesis in human skeletal muscle to insulin: an
investigation with L-[ 2H5]phenylalanine. Am. J. Physiol. 267
(Endocrinol. Metab. 30): E102–E108, 1994.
41. Moller-Loswick, A. C., H. Zachrisson, A. Hyltander, U.
Korner, D. E. Matthews, and K. Lundholm. Insulin selectively attenuates breakdown of nonmyofibrillar proteins in peripheral tissues of normal men. Am. J. Physiol. 266 (Endocrinol.
Metab. 29): E645–E652, 1994.
42. Morgan, H. E., L. S. Jefferson, E. B. Wolpert, and D. E.
Rannels. Regulation of protein synthesis in heart muscle. II.
Effect of amino acid levels and insulin on ribosomal aggregation.
J. Biol. Chem. 246: 2163–2170, 1971.
43. Mosoni, L., M. Houlier, P. P. Mirand, G. Bayle, and J.
Grizard. Effect of amino acids alone or with insulin on muscle
and liver protein synthesis in adult and old rats. Am. J. Physiol.
264 (Endocrinol. Metab. 27): E614–E620, 1993.
44. Nair, K. S., G. C. Ford, and D. Halliday. Effect of intravenous
insulin treatment on in vivo whole body leucine kinetics and
oxygen consumption in insulin-deprived type I diabetic patients.
Metabolism 36: 491–495, 1987.
45. Natali, A., G. Buzzigoli, S. Taddei, D. Santoro, M. Cerri, P.
Pedrinelli, and E. Ferrannini. Effects of insulin on hemodynamics and metabolism in human forearm. Diabetes 39: 490–
500, 1990.
46. Newman, E., M. J. Heslin, R. F. Wolf, P. T. W. Pisters, and
M. F. Brennan. The effect of systemic hyperinsulinemia with
concomitant infusion of amino acids on skeletal muscle protein
turnover in the human forearm. Metabolism 43: 70–78, 1994.
47. Pacy, P. J., K. S. Nair, C. Ford, and D. Halliday. Failure of
insulin infusion to stimulate fractional muscle protein synthesis
in type I diabetic patients. Diabetes 338: 618–624, 1989.