Am J Physiol Endocrinol Metab 299: E80–E91, 2010.
First published April 20, 2010; doi:10.1152/ajpendo.00765.2009.
Effect of physical training on insulin secretion and action in skeletal muscle
and adipose tissue of first-degree relatives of type 2 diabetic patients
Flemming Dela and Bente Stallknecht
Center for Healthy Aging, Department of Biomedical Sciences, University of Copenhagen, Copenhagen, Denmark
Submitted 22 December 2009; accepted in final form 17 April 2010
glucagon-like peptide-1; hyperglycemic clamp; blood flow; intravenous glucose tolerance test; hyperinsulinemic euglycemic clamp;
microdialysis; arteriovenous balance
in the initial treatment of
patients with type 2 diabetes. In skeletal muscle, the insulinsensitizing effect of training in both healthy individuals and
patients with type 2 diabetes is well documented (6, 8, 10, 34,
46). In individuals with a genetic predisposition for type 2
diabetes, first-degree relatives (FDR) of patients with type 2
diabetes, only a few training studies have been carried out, and
none have directly examined the effect of physical training on
insulin sensitivity in skeletal muscle and adipose tissue. In one
study in a mixed group of FDR males and females, training
enhanced insulin-mediated whole body glucose uptake rates
(40), but there was no correlation between glucose uptake and
measures of fitness [maximal oxygen uptake (V̇O2max)]. In a
recent study in females (3), a training-induced increase in
whole body insulin sensitivity was found only in FDR subjects,
yet this could not be demonstrated in control subjects (CON)
despite similar increases in V̇O2max in the two groups.
PHYSICAL TRAINING IS A CORNERSTONE
Address for reprint requests and other correspondence: F. Dela, Center for
Healthy Aging, Dept. of Biomedical Sciences, Univ. of Copenhagen, Blegdamsvej 3, 2200 Copenhagen N, Denmark (e-mail: [email protected]).
E80
The major effect of training on insulin action is located in
skeletal muscle (7, 8), but adipose tissue can also be influenced
by exercise training, and it has been previously demonstrated
that insulin sensitivity and GLUT4 content in adipose tissue
increase in response to training (51, 52). The potential importance of adipose tissue in the development of type 2 diabetes is
indicated by the fact that mice with adipose tissue selective
reduction in GLUT4 develop insulin resistance in muscle and
liver (1).
Glucose- and arginine-stimulated ␤-cell secretion decreases
with a training-induced increase in insulin sensitivity (9, 28,
33, 35) in young, healthy subjects. However, in patients with
type 2 diabetes this relationship does not necessarily exist.
Patients with a well-preserved insulin-secretory capacity (typically those with the shortest duration of the disease) display an
increased rather than a decreased insulin secretion in response
to training (11). For this reason and because a potential ␤-cell
adaptation to physical training to our knowledge never has
been investigated in FDR, the present study included a test of
the ␤-cell response to stimulation. Glucagon-like peptide-1
(GLP-1) was used as a secretagogue to be superimposed upon
hyperglycemia to elicit a maximal insulin response with the
purpose of unmasking possible differences between CON and
FDR, since GLP-1 is the most potent insulinotropic hormone
known that is well tolerated in humans (60).
The inverse relationship between training-induced changes
in insulin action and secretion has usually, but not exclusively
(33), been demonstrated on the basis of cross-sectional and
observational data. In the present study, we sought to elucidate
the effect of endurance training on insulin action in both
skeletal muscle and adipose tissue, as well as on the whole
body level, together with measurements of insulin-secretory
capacity, in a comprehensive interventional study in male
subjects stratified according to their genetic disposition to type
2 diabetes.
We hypothesized that endurance training in both groups of
subjects would increase insulin-mediated glucose uptake rates
across the whole body, skeletal muscle, and adipose tissue.
Furthermore, we hypothesized that glucose- and GLP-1-stimulated insulin secretion would decrease after training in both
groups, i.e., that insulin action and secretion adapt to physical
training similarly in FDR and CON.
MATERIALS AND METHODS
Subjects
After ethical approval by the Ethics Committee for Copenhagen
and Frederiksberg (Project no. 01-289/00) healthy men with (n ⫽ 7)
and without (n ⫽ 8) a family history of type 2 diabetes, respectively,
were recruited via advertisements in newspapers. The study was
conducted in accordance with the Declaration of Helsinki, and informed written consent was obtained from each subject. The subjects
0193-1849/10 Copyright © 2010 the American Physiological Society
http://www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
Dela F, Stallknecht B. Effect of physical training on insulin
secretion and action in skeletal muscle and adipose tissue of firstdegree relatives of type 2 diabetic patients. Am J Physiol Endocrinol
Metab 299: E80 –E91, 2010. First published April 20, 2010;
doi:10.1152/ajpendo.00765.2009.—Physical training affects insulin
secretion and action, but there is a paucity of data on the direct effects
in skeletal muscle and adipose tissue and on the effect of training in
first-degree relatives (FDR) of patients with type 2 diabetes. We
studied insulin action at the whole body level and peripherally in
skeletal muscle and adipose tissue as well as insulin-secretory capacity in seven FDR and eight control (CON) subjects before and after 12
wk of endurance training. Training improved physical fitness. Insulinmediated glucose uptake (GU) increased (whole body and leg; P ⬍
0.05) after training in CON but not in FDR, whereas glucose-mediated
GU increased (P ⬍ 0.05) in both groups. Adipose tissue GU was not
affected by training, but it was higher (abdominal, P ⬍ 0.05; femoral,
P ⫽ 0.09) in FDR compared with CON. Training increased skeletal
muscle lipolysis (P ⬍ 0.05), and it was markedly higher (P ⬍ 0.05)
in subcutaneous abdominal than in femoral adipose tissue and quadriceps muscle with no difference between FDR and CON. Glucosestimulated insulin secretion was lower in FDR compared with CON,
but no effect of training was seen. Glucagon-like peptide-1 stimulated
insulin secretion five- to sevenfold. We conclude that insulin-secretory capacity is lower in FDR than in CON and that there is
dissociation between training-induced changes in insulin secretion
and insulin-mediated GU. Maximal GU rates are similar between
groups and increases with physical training.
E81
TRAINING: INSULIN SECRETION AND ACTION
with a family history of type 2 diabetes (FDR) had at least one
first-degree relative with confirmed (via hospital or general practitioner records) type 2 diabetes. All subjects had normal glucose tolerance
as assessed by a 75-g oral glucose tolerance test. The ages of the
subjects were 40 ⫾ 5 and 41 ⫾ 6 yr (means ⫾ SE) in FDR and CON,
respectively. Characteristics of the subjects are given in Table 1.
Training Protocol
Experimental Protocol
The study consisted of three experimental days and a physical
training period followed by a repetition of the three experimental
days. In addition, cardiorespiratory fitness [submax ergometer bicycle
test with incremental 2-min steps of increasing workloads (50
W/step)] was measured before, during, and at the end of the training
period. Pulmonary oxygen consumption (V̇O2) and carbon dioxide
excretion (V̇CO2) were measured using a Jaeger Champion (IntraMedic, Lyngby, Denmark). This test was followed by 5-min rest and
thereafter measurements of maximal V̇O2 (V̇O2max) using the plateau
criterion. On separate days, body composition was determined: thigh
circumference 20 cm proximal to the patella on both legs; leg volume
by water displacement; abdominal skinfold by calipers (model 01127;
Lafayette Instruments, IN); and percent body fat by DEXA scanning
(LUNAR DPX-IQ; Lunar, Madison, WI). Three days before any
experiment, the subjects consumed a diet containing at least 250 g of
Table 1. Subject characteristics of FDR and CON subjects before and after 3 mo of ergometer bicycle training
FDR (n ⫽ 7)
Before Training
CON (n ⫽ 8)
After Training
Before Training
After Training
90.8 ⫾ 2.9
28.1 ⫾ 0.8
26.8 ⫾ 2.3
14.5 ⫾ 0.9
59.9 ⫾ 1.7
21 ⫾ 4
886 ⫾ 75
89.5 ⫾ 3.5
27.7 ⫾ 1.0
26.0 ⫾ 1.7
14.7 ⫾ 0.8
59.1 ⫾ 1.9
18 ⫾ 4
673 ⫾ 35
Two-Way RM-ANOVA
Training
Group
Interaction
0.005
0.005
0.072
0.399
0.993
0.001
0.003
0.669
0.616
0.418
0.667
1.000
0.356
0.438
0.489
0.628
0.594
0.886
0.998
0.065
0.650
293 ⫾ 27
35.1 ⫾ 2.3
180 ⫾ 7
12.8 ⫾ 0.9
147 ⫾ 11
135 ⫾ 4
2.8 ⫾ 0.7*
50 ⫾ 8
0.012
0.002
0.026
0.063
0.005
0.017
⬍0.001
0.190
0.124
0.253
0.658
0.706
0.134
0.174
0.301
0.518
0.347
0.993
0.473
0.704
0.064
0.072
0.021
0.009
0.29 ⫾ 0.02
0.41 ⫾ 0.03
600 ⫾ 47
0.004
0.057
0.007
0.715
0.671
0.794
0.466
0.510
0.967
Body composition
Weight, kga
BMI, kg/m2
Body fat, %
Leg volume, litersb
Thigh circumference, cmc
Abdominal skinfold, mmd
Subcutaneous abdominal adipocyte volume, pl
93.5 ⫾ 4.3
27.4 ⫾ 1.3
24.8 ⫾ 1.7
15.0 ⫾ 0.5
58.8 ⫾ 1.6
27 ⫾ 4
775 ⫾ 144
Work load, max, W
VO2max, ml · min⫺1 · kg⫺1
Heart rate max, /lmin
Lactate, max, mM
VEmax, l/min
Heart rate, 150 W, l/min
Lactate, 150 W, mM
VE, 150 W, l/min
314 ⫾ 12
36.4 ⫾ 2.9
187 ⫾ 6
12.6 ⫾ 1.3
156 ⫾ 11
133 ⫾ 7
3.0 ⫾ 0.7
55 ⫾ 2
91.1 ⫾ 4.6
26.8 ⫾ 1.5
23.4 ⫾ 2.2
15.2 ⫾ 0.7
58.4 ⫾ 1.6
21 ⫾ 3
598 ⫾ 109
Cardiorespiratory fitness
330 ⫾ 14
39.5 ⫾ 3.6
183 ⫾ 6
13.5 ⫾ 1.9
176 ⫾ 24
131 ⫾ 4
2.4 ⫾ 0.6*
58 ⫾ 8
260 ⫾ 21
32.0 ⫾ 2.0
180 ⫾ 7
11.9 ⫾ 0.7
141 ⫾ 9
146 ⫾ 3
4.6 ⫾ 0.8
65 ⫾ 5
Muscle biopsy analyses
CS activity, ␮mol · min · mg protein
␤-HAD activity, ␮mol · min⫺1 · mg protein⫺1
Glycogen, nmol/mg dry wt
⫺1
⫺1
0.26 ⫾ 0.01
0.39 ⫾ 0.02
486 ⫾ 25
0.30 ⫾ 0.02
0.42 ⫾ 0.03
590 ⫾ 58
0.24 ⫾ 0.02
0.36 ⫾ 0.02
500 ⫾ 23
Data are means ⫾ SE. FDR, first-degree relatives of type 2 diabtetic; CON, control. CS: Citrate synthase; ␤-HAD: ␤-hydroxyacyl-CoA-dehydrogenase.
Measured on 2 different occasions; baverage of both legs; cmeasured 20 cm proximal to the patella; dmeasured laterally to the umbilicus. Statistical analysis by
2-way ANOVA for repeated measures. *Effect (P ⬍ 0.05) of training in subgroup analysis.
a
AJP-Endocrinol Metab • VOL
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
The training procotol included a home-based bicycle ergometer
(Monark 828E or 818E; Monark Exercise, Varberg, Sweden) training
for 12 wk, 6 days/wk, 45 min/day at 70% of V̇O2max. At every training
session, the participants filled out a training diary showing date,
workload and heart rate (POLAR S610i; Polar Electro Oy, Kempele,
Finland), which was mailed to the laboratory one or two times per
week. The participants were instructed in adjustments of the workload
according to the gradual increase in fitness. According to training
diaries, adherence to training sessions was 87 ⫾ 4% in CON and
82 ⫾ 9% in FDR (P ⬎ 0.05).
carbohydrates per day, and one day before, they abstained from
alcohol and caffeine intake.
The three experimental days consisted of a frequently sampled
intravenous glucose tolerance test (FSIVGTT), a hyperglycemic
clamp, and a hyperinsulinemic euglycemic clamp. All experimental
days were separated by 2 days. All experiments started at 8:00 in the
morning, with the subjects in a 10-h fasted state and in the posttraining
experiments 16 h after the last training bout.
FSIVGTT. An arterial catheter was inserted in the radial artery for
blood sampling. A venous cubital catheter was used for injection of a
bolus (300 mg/kg body wt; max 25 g) of glucose (20%). The bolus
was given within 1 min, immediately followed by 50 ml of saline flush
of the catheter. Blood was drawn for measurements of plasma glucose
and insulin concentration at depicted time points.
Hyperglycemic clamp. Glucose (20%) was infused (using exponentially decreasing infusion rates) into a cubital vein to reach a 20 mM
steady-state glucose concentration within 7–12 min, which was maintained via frequent blood sampling (heated hand vein) and glucose
analysis and subsequent adjustment of the glucose infusion. At t ⫽ 90
min, 2.5 nmol (20 ␮g) GLP-1 diluted in 50 ml of sterile saline was
injected over 1 min in the catheter in the cubital vein. Blood samples
were drawn for measurements of plasma insulin, C-peptide, and
proinsulin.
Hyperinsulinemic euglycemic clamp. Microdialysis catheters were
inserted in subcutaneous abdominal and femoral adipose tissue and in
the quadriceps muscle (vastus lateralis) as described below. Subsequently, biopsies were obtained from vastus lateralis and the subcutaneous abdominal adipose tissue using a Bergström needle. A venous
catheter for infusion of glucose (20%) and insulin was inserted in a
cubital vein. A catheter was inserted in the radial or the brachial artery
for drawing of blood samples and for monitoring and recording of
blood pressure. In a femoral vein a catheter was inserted for later
measurements of blood flow (by thermodilution) and drawing of
blood samples. A two-step (insulin infusion rates: 28 and 480
mU · min⫺1 · m⫺2) sequential clamp was performed, with each step
E82
TRAINING: INSULIN SECRETION AND ACTION
lasting 120 min. Before and at the end of each step, arterial and
femoral-venous blood was sampled for determination of substrate and
hormone concentrations. Unless otherwise noted, data are given as the
mean of the duplicate baseline and the mean of the triplicate endclamp values. Leg blood flow was measured three to five times
immediately after each blood sampling. V̇O2 and V̇CO2 were measured
by the ventilated hood technique (Jaeger Champion; IntraMedic,
Lyngby, Denmark) at baseline and during the final 30 min of each
clamp step.
Microdialysis
Subcutaneous Adipose Tissue Blood Flow
Subcutaneous adipose tissue blood flow (ATBF) was measured
with the local 133Xe washout technique (30). At least 30 min before
the start of the basal period, 0.5–1 MBq gaseous 133Xe (Amersham
Health UK) in a volume of 0.05– 0.1 ml was injected in the subcutaneous abdominal and femoral adipose tissue. The washout rate of
133
Xe was measured continuously with a scintillation counter system
(Oakfield Instruments, Eynsham, UK) strapped to the skin surface
above the 133Xe depot. ATBF was determined in 30-min periods and
calculated as ⫺k · ␭ · 100 (ml · 100 g⫺1 · min⫺1), where k is the rate
constant of the washout and ␭ is the tissue-to-blood partition coefficient for 133Xe at equilibrium, which was taken to be 10 ml/g.
Muscle Enzymes And Glycogen
Before biochemical analysis, muscle samples were freeze-dried.
With the use of a stereomicroscope, connective tissue, visible fat, and
blood were removed from the muscle samples. Muscle glycogen
concentration was determined as glucose residues after hydrolysis of
the muscle sample in 1 M HCl at 100°C for 2 h. Citrate synthase and
␤-hydroxyacyl-CoA-dehydrogenase (HAD) activities were measured
spectrophotometrically at 37°C. Enzyme activities are expressed as
micromoles substrate per minute per gram dry weight muscle tissue.
Adipocyte Size and In Vitro Glucose Incorporation
The adipose tissue was washed with sodium chloride, and adipocytes were isolated by collagenase treatment (0.6 mg/ml, 3.5% albumin, 0.55 mM glucose, pH 7.4, 37°C, 60 min), filtered, and washed
three times with glucose-free buffer. Fifty microliters of adipocytes
(lipocrit 20%) were mixed with 2 ml of osmium 2% and left for 24 –72
h. The radius (r) of each of 200 cells was measured in a stereomicroscope with scaled oculars, and adipocyte volume was calculated as
4/3␲r3. The remaining adipocytes were incubated in buffer (3.5%
albumin, pH 7.4, 37°C, lipocrit 1–2%) with trace amounts of
AJP-Endocrinol Metab • VOL
Sampling and Analyses
Dialysate was collected in 200-␮l capped microvials for 30 min
during the basal period and for the last 30 min of each of the two
clamp steps, immediately frozen, and kept at ⫺20°C until analysis.
Dialysate sampling was delayed by 2 min relative to the rest of the
experimental protocol to compensate for the transit time in the outlet
tubing. Blood was sampled into iced tubes. Blood for determination of
insulin, C-peptide, and proinsulin was stabilized with 500 KIE trasylol
and 1.5 mg EDTA per milliliter of blood and centrifuged immediately
at 4°C. All blood samples were kept at ⫺20°C until analysis. Microdialysate glucose and glycerol as well as plasma glycerol concentrations were determined using a CMA 600 microdialysis analyzer
(CMA/Microdialysis, Stockholm, Sweden). Plasma glucose and lactate concentrations were analyzed with an automatic analyzer (ABL
700; Radiometer, Copenhagen, Denmark). Plasma concentrations of
insulin, C-peptide, and proinsulin were determined using commercially available ELISA kits (DAKO, Glostrup, Denmark). Plasma free
fatty acids (FFA) were measured photometrically (Roche/Hitachi 912
System; Roche, Glostrup, Denmark) using a Wako NEFA-C test kit
(Wako Chemical, Neuss, Germany). All analyses were carried out in
duplicate.
Calculations
The first-phase insulin response (FPIR) to a glucose bolus in the
FSIVGTT experiments was calculated as the mean of the incremental
plasma insulin concentrations from 0 –10 min following the intravenous glucose bolus.
Glucose infusion rates (GIR) were averaged over 5-min intervals
throughout the two sequential 120-min hyperinsulinemic euglycemic
clamps. The steady-state GIR minus glucose space correction in the
final part (30 min) of each clamp step was taken as the M value. GIR
data are presented per fat free mass (FFM).
Insulin action (on glucose uptake rates) and ␤-cell secretory capacity vary inversely with each other in a hyperbolic relationship. An
index of this relationship was calculated as FPIR/M (clamp step I).
Leg glucose uptake was calculated as the arteriovenous blood
glucose concentration difference multiplied by leg blood flow, and
data are expressed relative to leg mass, assuming that one liter of leg
(water displacement) equals one kilogram of leg.
Estimates of adipose tissue and skeletal muscle venous metabolite
concentrations from interstitial metabolite concentrations were based
on various assumptions as described previously (53). Metabolite
exchange was calculated as arteriovenous metabolite concentration
multiplied by blood flow (53).
Statistical Analysis
Results are presented as means ⫾ SE. Data obtained during
hyperglycemic and during hyperinsulinemic glucose clamping were
analyzed by two-way ANOVA for repeated measurements (i.e., Before training, CON vs. FDR; after training, CON vs. FDR; CON,
before vs. after training; FDR, before vs. after training). Data on body
composition, fitness, and muscle biopsy analyses (i.e., data containing
one-factor repetition) were also analyzed by two-way ANOVA for
repeated measurements. The Student-Newman-Keuls test was used
post hoc to locate differences if statistical interaction was present.
Other data were tested by means of nonparametric ranking tests,
Mann-Whitney’s for unpaired data and Wilcoxon’s for paired data as
appropriate. Correlation between variables was tested by Pearsons
correlation test. P ⬍ 0.05 was considered significant in a two-tailed
test.
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
Microdialysis was performed in principle as described previously
(53). At least 60 min before start of the basal period and after local
anesthesia, three microdialysis catheters (CMA 60; CMA/Microdialysis, Stockholm, Sweden) were placed in the subcutaneous abdominal
and femoral adipose tissue and in the vastus lateralis part of the
quadriceps muscle. Catheters were perfused at a rate of 1.0 ␮l/min.
The perfusate consisted of Ringer acetate with 2 mM glucose, 5
kBq/ml [14C]glucose (NEN, Zaventem, Belgium), and 5 kBq/ml
[3H]glycerol (NEN).
The in vivo relative recovery (RR) for glucose and glycerol was
determined by the internal reference calibration technique (48). This
technique assumes that the relative loss of the labeled metabolite from
the microdialysis catheter to the interstitial fluid equals the RR of the
unlabeled metabolite from the interstitial fluid to the microdialysis
catheter. The RR of glucose and glycerol was calculated as (Dpmp ⫺
Dpmd)/Dpmp, where Dpmp is 14C or 3H disintegrations per minute in
5 ␮l of perfusate and Dpmd is 14C or 3H disintegrations per minute in
5 ␮l of dialysate. Interstitial concentrations of glucose and glycerol
were calculated as (Cd ⫺ Cp)/RR ⫹ Cp, where Cd is dialysate
concentration and Cp is perfusate concentration.
(1.5 ␮mol/l) with or without addition of insulin
(1,200 ␮U/ml). After 60 min, the lipid phase was extracted with
heptane, and the glucose incorporation in lipid was determined by
scintillation counting.
D-[U-14C]glucose
E83
TRAINING: INSULIN SECRETION AND ACTION
RESULTS
Body Composition and Training Markers
Whole Body and Leg Insulin Action
During the hyperinsulinemic euglycemic clamp experiments, insulin concentrations were not different before and
after training or between the groups (Table 2). Insulin-mediated whole body glucose uptake rates (M value) increased (P ⫽
0.006) in response to training in CON, whereas a significant
change was not seen in FDR (Fig. 1). Glucose infusion rates
were not different between FDR and CON at clamp step I, but
at the maximal insulin concentration (clamp step II) glucose
infusion rates were higher in FDR compared with CON (P ⬍
0.05). In the legs, representing mostly skeletal muscle, a
similar pattern in glucose uptake rates were seen in CON
(increase with training) and in FDR (no effect of training) (Fig.
1). The training-induced increase in glucose uptake rates in
CON was primarily due to an effect on leg blood flow, as
arteriovenous glucose extraction rates did not change significantly with training (data not shown). Leg glucose uptake rates
Adipose Tissue and Skeletal Muscle Metabolism
In vivo glucose uptake (Fig. 3) was higher (P ⬍ 0.05) in
subcutaneous abdominal adipose tissue and tended to be higher
(P ⫽ 0.09) in femoral adipose tissue in FDR compared with
CON, but quadriceps muscle glucose uptake did not differ
(P ⬍ 0.05) between groups. No effect of physical training on
in vivo glucose uptake was detected in any of the tissues (all,
P ⬍ 0.05). In adipose tissue, glucose uptake was increased
significantly only at a high insulin concentration (step II),
whereas in skeletal muscle glucose uptake was increased (P ⬍
0.05) already at a low insulin concentration (step I). Insulinstimulated glucose uptake was markedly higher (P ⬍ 0.05) in
quadriceps muscle than in subcutaneous abdominal and femoral adipose tissue, but insulin-stimulated glucose uptake did not
differ (P ⬎ 0.05) between the two adipose tissue depots.
Physical training increased (P ⬍ 0.05) basal quadriceps
muscle in vivo lipolysis (Fig. 4), but physical training did not
influence (P ⬎ 0.05) adipose tissue or quadriceps muscle
lipolysis during insulin infusion. Group (FDR/CON) did not
influence (P ⬎ 0.05) adipose tissue or quadriceps muscle
lipolysis. In subcutaneous abdominal and femoral adipose
Table 2. Plasma insulin and metabolite concentrations at baseline and during 2-step hyperinsulinemic euglycemic clamp in
FDR and CON subjects before and after 3 mo of ergometer bicycle training
FDR (n ⫽ 7)
Insulin, pmol/l
Before
After
Glucose, mmol/l
Before
After
Glycerol, ␮mol/l
Before
After
FFA, ␮mol/l
Before
After
CON (n ⫽ 8)
Basal
Clamp Step I
Clamp Step II
Basal
Clamp Step I
Clamp Step II
51 ⫾ 6
45 ⫾ 5
307 ⫾ 26*
292 ⫾ 18*
14,130 ⫾ 764*
13,687 ⫾ 787*
55 ⫾ 6
50 ⫾ 6
333 ⫾ 19*
303 ⫾ 24*
12,760 ⫾ 914*
11,751 ⫾ 1,067*
5.4 ⫾ 0.1
5.3 ⫾ 0.1
5.5 ⫾ 0.1
5.5 ⫾ 0.1
5.4 ⫾ 0.1
5.6 ⫾ 0.1
5.3 ⫾ 0.1
5.3 ⫾ 0.1
5.4 ⫾ 0.1
5.4 ⫾ 0.1
5.6 ⫾ 0.1
5.5 ⫾ 0.1
52 ⫾ 5
55 ⫾ 7
19 ⫾ 2*
17 ⫾ 1*
22 ⫾ 1
19 ⫾ 1
51 ⫾ 6
41 ⫾ 4
20 ⫾ 4*
21 ⫾ 3*
17 ⫾ 2
16 ⫾ 2
481 ⫾ 22
500 ⫾ 53
40 ⫾ 10*
29 ⫾ 6*
14 ⫾ 3*
16 ⫾ 4*
612 ⫾ 110
466 ⫾ 37
57 ⫾ 13*
48 ⫾ 12*
13 ⫾ 2*
11 ⫾ 1*
Values are means ⫾ SE. Within the clamp steps, insulin concentrations were not different between groups, but basal concentrations decreased with training
(main effect, P ⫽ 0.048). There was a significant suppression of FFA and glycerol concentrations with increasing insulin concentrations, but this effect was not
changed with training. *Significant (P ⬍ 0.05) difference from previous value.
AJP-Endocrinol Metab • VOL
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
The training program had a significant impact on body
composition in both FDR and CON. The training-induced
weight loss was primarily due to loss of fat as evidenced by a
lower body fat percentage, less abdominal skinfold thickness,
and a decrease in subcutaneous abdominal adipocyte size
(Table 1). Leg volume and thigh circumference were not
changed by training (Table 1), which may have been due to
opposite effects on leg skeletal muscle mass and subcutaneous
adipose tissue. V̇O2max increased similarly in FDR (⫹9 ⫾ 3%)
and in CON (⫹10 ⫾ 3%), and in all measured parameters of
cardiorespiratory fitness a main effect of physical training was
seen, with some minor differences between the two groups
(Table 1). A decreased heart rate and blood lactate concentration during the submaximal, incremental exercise test after
training (data not shown), as well as the changes in glycogen
content, CS, and HAD activity, also revealed the effects of the
training program (Table 1). At maximal exercise, a slight
decrease in maximal heart rate and a profound increase in
blood lactate (data not shown) indicated increased fitness as a
result of the training program.
were not significantly different between CON and FDR, either
before or after training.
The effect of training in CON was predominantly due to
increases in nonoxidative glucose disposal (in clamp steps I
and II, both P ⬍ 0.05) and to a minor extent a switch from
glucose to lipid oxidation (clamp step II, P ⬍ 0.05) (data not
shown).
M values correlated positively (P ⬍ 0.05) with V̇O2max in
CON (clamp step II: R2 ⫽ 0.57) but not in FDR (clamp step II:
R2 ⬍ 0.1) (data from before and after training in one analysis).
Likewise, a significant correlation was found between ⌬M
value and ⌬V̇O2max in CON (clamp step II, R2 ⫽ 0.38) but not
in FDR (clamp step II, R2 ⬍ 0.1).
During hyperglycemic clamping, the glucose infusion rates
necessary to maintain hyperglycemia at ⬇20 mM increased
significantly in both groups with training (Fig. 2). This infusion
rate is considered a measure of the combined effect of hyperinsulinemia and hyperglycemia on whole body glucose uptake
rates, given that plasma glucose concentrations are stable.
E84
TRAINING: INSULIN SECRETION AND ACTION
tissue, insulin suppressed lipolysis significantly already at a
low insulin concentration (step I), but insulin did not (P ⬎
0.05) suppress lipolysis in quadriceps muscle. Lipolysis was
markedly higher (P ⬍ 0.05) in subcutaneous abdominal adipose tissue than in femoral adipose tissue and quadriceps
muscle.
In vitro glucose incorporation into triacylglycerol in adipocytes did not differ (P ⬎ 0.05) between FDR and CON subjects
in either the basal state [before training: 7.6 ⫾ 2.4
fmol·min⫺1 ·␮l lipid⫺1 (FDR) vs. 5.8 ⫾ 0.8 fmol·min⫺1 ·␮l
lipid⫺1 (CON); after training: 5.3 ⫾ 1.3 (FDR) vs. 4.5 ⫾ 0.6
fmol·min⫺1 ·␮l lipid⫺1 (CON)], or during insulin stimulation
[before training: 9.1 ⫾ 2.3 (FDR) vs. 7.1 ⫾ 1.0 fmol·min⫺1 ·␮l
lipid⫺1 (CON); after training: 6.7 ⫾ 1.4 (FDR) vs. 5.3 ⫾ 0.8
fmol·min⫺1 ·␮l lipid⫺1 (CON)]. Physical training did not influence (P ⬎ 0.05) in vitro adipocyte glucose incorporation
into triacylglycerol when expressed per amount of lipid (and
hence corrected for the training-induced decrease in adipocyte
volume), but physical training decreased (P ⬍ 0.05) basal [before
training: 4.9 ⫾ 0.6 amol·min⫺1 ·adipocyte⫺1 (FDR) vs. 5.7 ⫾ 0.9
amol·min⫺1 ·adipocyte⫺1 (CON); after training: 3.1 ⫾ 0.4 (FDR)
vs. 3.2 ⫾ 0.3 amol·min⫺1 ·adipocyte⫺1 (CON)] and insulinAJP-Endocrinol Metab • VOL
stimulated [before training: 6.2 ⫾ 1.3 (FDR) vs. 6.7 ⫾ 0.9
amol·min⫺1 ·adipocyte⫺1 (CON); after training: 3.7 ⫾ 0.7 (FDR)
vs. 3.8 ⫾ 0.4 amol·min⫺1 ·adipocyte⫺1 (CON)] glucose incorporation when expressed per adipocyte in both FDR and CON.
Insulin increased (P ⬍ 0.05) in vitro glucose incorporation into
triacylglycerol in adipocytes with no difference (P ⬎ 0.05) between groups.
Insulin Secretion
FPIR (P ⬍ 0.05; Fig. 5) and the insulin area under the curve
for the entire 40-min FSIVGTT test (P ⫽ 0.051; data not
shown) was lower in FDR compared with CON, and training
did not change these parameters significantly in either group.
During hyperglycemic clamping, plasma glucose concentration was rapidly increased from fasting levels to ⬇20 mM (Fig.
2), and the coefficient of variation (CV%) of plasma glucose
concentrations throughout the clamp was not different between
the study groups or before and after 3 mo of training (CV%
from t ⫽ 15 min and onward, before and after training,
respectively: FDR, 5.2 ⫾ 0.8 and 4.8 ⫾ 0.6%; CON, 6.2 ⫾ 0.7
and 3.8 ⫾ 0.5%).
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
Fig. 1. Whole body (top) and leg (bottom) and
glucose uptake rates in 7 first-degree relatives
(FDR) to patients with type 2 diabetes and 8
control subjects (CON) before (open bars) and
after (filled bars) 3 mo of endurance training.
Measurements were done before (basal) and
during a two-step hyperinsulinemic euglycemic
clamp (step I and step II). Leg glucose uptake
rates did not differ significantly between CON
and FDR either before or after training. Values
are means ⫾ SE. *Effect of training (P ⬍ 0.05).
TRAINING: INSULIN SECRETION AND ACTION
Blood Metabolites
Fasting arterial glycerol and FFA concentrations did not
differ between the groups and did not change with training
(Table 2). With hyperinsulinemic euglycemic clamping, glycerol and FFA concentrations decreased (P ⫽ 0.05) in both
groups, and the decrease in response to hyperinsulinemia was
similar in the two groups and did not change significantly with
training (Table 2).
DISCUSSION
The main findings of the present study, which is relatively
small in number of subjects but experimentally comprehensive,
are that healthy men who are first-degree relatives to patients
with type 2 diabetes do not improve their insulin-mediated
glucose uptake at the whole body level or in skeletal muscle
and adipose tissue in response to physical training. However,
with combined hyperinsulinemia and hyperglycemia (i.e., hyperglycemic clamping) marked improvements of maximal glucose uptake rates were seen at the whole body level. Neither
AJP-Endocrinol Metab • VOL
FDR nor the healthy control subjects changed insulin-secretory
capacity in response to physical training. Experiments on
metabolite turnover of subcutaneous adipose tissue revealed
regional differences and also that in vivo insulin-stimulated
glucose uptake was higher in FDR than in CON, while insulin
inhibition of lipolysis was similar in both groups. The FDR
subjects were metabolically characterized by a decreased insulin-secretory capacity, which was present in parallel with a
normal to supranormal peripheral insulin action. In these subjects, a decrease in insulin action would therefore most likely
result in the development of overt type 2 diabetes, because
there is no reserve capacity in insulin secretion to compensate
for a reduced insulin action. This relationship between insulin
secretion and action can be expressed as the ratio of FPIR and
the M value, which indeed was lower in FDR compared with
CON.
Insulin Action
Despite similar adherence to the training intervention program and similar increases in cardiorespiratory fitness, FDR
subjects did not improve insulin-mediated glucose uptake significantly on the whole body level or specifically in the leg,
whereas improvements were seen in the CON subjects. Thus,
at clamp steps I and II in CON, training increased whole body
glucose uptake rates by 34 ⫾ 6 and 12 ⫾ 5%, respectively, and
leg glucose uptake rates by 88 ⫾ 55 and 35 ⫾ 13%, respectively. The lack of training effect on insulin-mediated glucose
uptake rates in FDR subjects are in keeping with a recent bed
rest study in which CON subjects displayed significantly larger
adaptations to (in)activity compared with FDR subjects (49).
During a hyperglycemic clamp, glucose is infused to maintain hyperglycemia. Thus, the glucose infusion rate during
hyperglycemic clamping may also be taken as a measure of
whole body glucose uptake (36) mediated partly by the accompanying increase in endogenous insulin and partly by the mass
action of glucose per se. In contrast to insulin sensitivity, which
is typically evaluated at low to medium insulin concentrations
during a hyperinsulinemic euglycemic clamp (26), during the
hyperglycemic clamp we found similar training-induced increases in glucose infusion rates between FDR and CON.
These rates were achieved in the face of insulin concentrations
of “only” 4 –5,000 pmol/l (which is high, but still less than half
of what was seen during step II of the hyperinsulinemic
clamp). Seemingly conflicting results between the hyperinsulinemic and the hyperglycemic clamps, but the extra glucose
uptake is explained by the accompanying hyperglycemia,
which, by mass-action, increases glucose uptake, i.e., the
glucose uptake that is due to the hyperglycemia per se. Our
hypothesis is that the training-induced increase in stimulated
glucose uptake rates are due to a mixture of insulin- and
glucose-mediated glucose uptake in CON, whereas in FDR the
training effect is exclusively related to the glucose-mediated
glucose uptake. This interpretation is compatible with an augmentation of glucose-mediated glucose uptake in FDR, as
shown in some (19, 20) but not all (39) previous studies.
The lack of a training effect on insulin-mediated glucose
uptake (whole body and specifically in the leg) in FDR points
to the possibility that the exercise-induced changes in proteins
and enzymes in the insulin signaling cascade in skeletal muscle
is impaired in people with a genetic disposition toward insulin
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
During the first 90 min, when only glucose was infused, plasma
insulin and C-peptide concentrations always increased significantly (P ⬍ 0.05), and no plateau was seen (Fig. 2). At t ⫽ 90
min, when a GLP-1 bolus was added, a substantial further increase was always seen occurring within 2 min after the bolus was
given (Fig. 2). Plasma insulin concentrations peaked at 5.8 ⫾ 0.7
and 6.3 ⫾ 0.8 min (FDR, before and after training, respectively)
and 5.6 ⫾ 0.8 and 6.1 ⫾ 0.7 min (CON, before and after training,
respectively) after the GLP-1 bolus was given, but the time to
peak was not different (P ⬎ 0.05) between groups or with
training. The GLP-1-induced increase in insulin concentrations
was marked (fold increase: 6.3 ⫾ 1.1 and 5.2 ⫾ 0.9 in FDR before
and after training, respectively; 6.9 ⫾ 1.1 and 6.8 ⫾ 1.2 in CON
before and after training, respectively), but the fold increase was
not different (P ⬎ 0.05) between groups or with training. Proinsulin concentrations mirrored insulin and C-peptide, with no
difference between groups at baseline (1.8 ⫾ 0.2 pmol/l), during
hyperglycemia (12.3 ⫾ 1.2 pmol/l) and during additional GLP-1
stimulation (38.4 ⫾ 4.2 pmol/l) (pooled mean data from both
groups before and after training).
Plasma insulin concentration and secretion (as judged by
C-peptide and proinsulin concentrations) did not change significantly with training in FDR or in CON subjects. However,
before training, plasma insulin concentrations during hyperglycemia alone tended (P ⫽ 0.066) to be lower in FDR than in
CON, and this difference became significant (P ⬍ 0.05) after
training. In support, after training, plasma C-peptide concentrations during hyperglycemia alone were also lower (P ⬍
0.05) in FDR than in CON. During the combined GLP-1
stimulation and hyperglycemia, insulin and C-peptide concentrations tended (P ⫽ 0.08 and P ⫽ 0.06, respectively) to be
lower in FDR than in CON (Fig. 2).
The integrated index of the ␤-cell capacity for insulin
production relative to peripheral insulin action on glucose
uptake rates, i.e., FPIR/M, did not change from before to after
training [FDR, 655 ⫾ 157 and 562 ⫾ 125; CON, 1,076 ⫾ 118
and 909 ⫾ 146 (pmol·liter⫺1 ·10 min⫺1)/(mg·min⫺1 · kg
FFM⫺1), respectively], but the index was lower in FDR than in
CON before training (P ⬍ 0.05) and tended to be lower after
training (P ⫽ 0.099).
E85
E86
TRAINING: INSULIN SECRETION AND ACTION
resistance. In the literature, most, but not all (32), have reported
impaired insulin signaling in skeletal muscle in FDR compared
with CON subjects [e.g., insulin receptor kinase activity, Akt
activation, PI 3-kinase activity, glycogen synthase activity (18,
27, 37, 54, 56)]. The effect of exercise training on these
parameters has not been measured in FDR, but in subjects
without familiar disposition to type 2 diabetes as well as in
patients with type 2 diabetes, strength (21) and endurance (16)
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
AJP-Endocrinol Metab • VOL
299 • JULY 2010 •
www.ajpendo.org
E87
TRAINING: INSULIN SECRETION AND ACTION
training increases skeletal muscle protein content of key proteins in insulin signaling. However, the predominant effect is a
training-induced enhancement of leg blood flow (16, 21), as
also found in CON, but not in FDR, in the present study. The
latter observation is in accord with recent findings of diminished insulin-mediated forearm blood flow in FDR (50).
In opposition to the present findings, in a recent study in
women, training increased an insulin sensitivity index in FDR
Fig. 2. Plasma glucose concentrations, glucose infusion rates, plasma insulin, and C-peptide concentrations during hyperglycemic clamps in 7 FDR (A–D) to
patients with type 2 diabetes and in 8 CON (E–H) subjects. Data are from before (open symbols) and after (filled symbols) 3 mo of endurance training. At t ⫽
90 min, a bolus (2.5 nmol) of glucagon-like peptide-1 (GLP-1) was given intravenously. Glucose infusion rates were significantly higher after training in both
groups (ANOVA for repeated measures). *Significant interaction between time and status of training, P ⫽ 0.002; difference (P ⬍ 0.05) from t ⫽ 45 min;
†difference between before and after training (main effect, P ⬍ 0.05). Before training, insulin concentrations tended (P ⫽ 0.066) to be lower in FDR than in
CON during hyperglycemia alone (t ⫽ 15–90 min). After training, insulin and C-peptide responses were significantly (P ⬍ 0.05) lower in FDR than in CON
during hyperglycemia (t ⫽ 15–90 min) and tended to be different between the groups during the addition of GLP-1 (P ⫽ 0.088 and P ⫽ 0.06, insulin and
C-peptide, respectively). Values are means ⫾ SE and displayed in 5-min intervals in A, B, E, and F. Note that the vertical scale changes from t ⫽ 90 min in
C, D, G, and H (use right axis from t ⫽ 90 min).
AJP-Endocrinol Metab • VOL
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
Fig. 3. Glucose uptake (by microdialysis technique) in 7 FDR and 7 CON subjects, before
(open bars) and after (filled bars) 3 mo of endurance training. At basal and during a 2-step
hyperinsulinemic euglycemic clamp (step I and
step II), glucose uptake was estimated in subcutaneous abdominal and femoral adipose tissue
and in quadriceps muscle. Insulin increased glucose uptake in subcutaneous abdominal (P ⬍
0.05; post hoc, basal vs. step II) and femoral
(P ⬍ 0.01; post hoc, basal vs. step II; step I vs.
step II) adipose tissue as well as in quadriceps
muscle (P ⬍ 0.001. post hoc: basal vs. step I and
II; step I vs. step II). Insulin-stimulated glucose
uptake was higher (P ⬍ 0.05) in quadriceps
muscle than in subcutaneous abdominal and
femoral adipose tissue. Glucose uptake was significantly higher in subcutaneous abdominal adipose tissue of FDR than in CON. Values are
means ⫾ SE. #P ⬍ 0.05 vs. CON (main effect).
E88
TRAINING: INSULIN SECRETION AND ACTION
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
Fig. 4. Lipolysis (glycerol release by microdialysis technique) in 7 FDR and 7 CON before
(open bars) and after (filled bars) 3 mo of endurance training. At basal and during a two-step
hyperinsulinemic euglycemic clamp (step I and
step II), lipolysis was estimated in subcutaneous
abdominal and femoral adipose tissue and in
quadriceps muscle. Insulin decreased lipolysis in
subcutaneous abdominal (P ⬍ 0.001; post hoc:
basal vs. steps I and II) and femoral (P ⬍ 0.001;
post hoc: basal vs. step I and II) adipose tissue.
Lipolysis was higher (P ⬍ 0.05) in subcutaneous
abdominal adipose tissue than in femoral adipose tissue and higher (P ⬍ 0.05) in the 2
adipose tissue depots than in quadriceps muscle.
Basal quadriceps muscle lipolysis was higher
after than before training. Values are means ⫾
SE. *P ⬍ 0.05 vs. before training.
but not in CON subjects despite similar increases in V̇O2max
(3), but similar increases in insulin-mediated glucose uptake in
FDR and CON subjects have also been shown (40). In the latter
study (40), insulin sensitivity correlated positively with V̇O2max
in CON but not in FDR, an observation that is confirmed by the
present study. The lack of a significant correlation between
insulin sensitivity and V̇O2max in FDR has been speculated (40)
to be due to an impaired oxidative capacity in the mitochondria
(i.e., a mitochondrial dysfunction) in FDR. However, skeletal
mitochondrial respiration is not decreased in insulin resistance
when the content of mitochondria is taken into account (4, 31,
45). Furthermore, mitochondrial biogenesis was stimulated in
both groups in the present study, as evidenced by traininginduced increases in CS activity (Table 1).
In this study, the FDR and CON groups were carefully
matched, their adherence to the training program was not
AJP-Endocrinol Metab • VOL
different, and the fitness outcome was identical. The fact that at
baseline FDR subjects were not insulin resistant compared with
CON cannot explain the lack of improvements in insulinmediated glucose uptake rates in FDR in response to the
training program. A wide heterogeneity in insulin sensitivity in
CON and FDR has been shown before (61). We did not
measure glucose transporter proteins, insulin signaling molecules, intramyocelluar lipids, etc., in muscle biopsies from the
present subjects, wherefore it cannot be excluded a priori that
differences between groups in these parameters may explain
the differences in insulin sensitivity outcomes. For example,
accumulation of intramyocellular lipids may impair insulin
action, and in FDR such an accumulation has previously been
demonstrated (23, 37, 42, 43, 55). Likewise, protein and gene
expression of phosphoprotein enriched in astrocytes 15
(PEA15), which interacts with the ERK-PI 3-kinase signaling
299 • JULY 2010 •
www.ajpendo.org
TRAINING: INSULIN SECRETION AND ACTION
E89
Insulin Secretion
Fig. 5. Plasma insulin concentrations before and following iv injection of
glucose (300 mg/kg body wt) before (open symbols) and after (filled symbols)
3 mo of endurance training in 7 FDR (circles) and 8 CON subjects (squares).
In the inset table, first-phase insulin response (FPIR) for 0 –10 min is given.
Values are mean ⫾ SE. *Different from FDR (P ⫽ 0.03, two-way ANOVA).
pathway and is negatively correlated with insulin sensitivity,
has been found to be overexpressed in FDR (57). The interaction of this protein with exercise training is not known.
Adipose Tissue and Skeletal Muscle Metabolism
Adipose tissue and skeletal muscle are important targets for
insulin action, and tissue-specific action can be examined in
vivo by the microdialysis technique and in vitro in tissue
biopsies. We found that increasing doses of insulin stimulate in
vivo glucose uptake in skeletal muscle in a stepwise fashion
but also that it takes a supraphysiological dose of insulin
(clamp step II) to stimulate subcutaneous adipose tissue glucose uptake. The relatively limited ability of insulin to stimulate glucose uptake in human adipose tissue has also been
found in previous studies using microdialysis (17) as well as
arteriovenous flux (5) techniques in subcutaneous abdominal
adipose tissue. In contrast, subcutaneous adipose tissue lipolysis is very sensitive to insulin and is suppressed at a physiological insulin concentration, whereas lipolysis in skeletal
muscle is not affected by insulin.
Interestingly, we found the in vivo glucose uptake in subcutaneous abdominal (P ⬍ 0.05) and femoral (P ⬍ 0.1)
adipose tissue to be higher in FDR than in CON. In vivo lactate
release from subcutaneous abdominal adipose tissue has previously been found to be higher in FDR rthan in CON (47), and
these findings suggest that an increased uptake and turnover of
glucose in adipose tissue is a primary defect in the development of type 2 diabetes. Furthermore, Eriksson et al. (14)
AJP-Endocrinol Metab • VOL
Glucose-stimulated insulin secretion was lower in FDR than
in CON. This is in accord with some (12, 13, 15, 25, 38, 41,
58), but not all (2, 24, 44, 59), previous reports. The addition
of GLP-1 stimulation to the hyperglycemic stimulus resulted in
a marked increase (5- to 7-fold) in insulin concentrations and
secretion in both groups; a fold increase that was somewhat
higher than previously seen in older healthy subjects and
patients with type 2 diabetes also stimulated with 2.5 nmol
GLP-1 and 15 mmol/l glucose, but the time to peak was similar
(60). Patients with type 2 diabetes have a reduced response to
combined hyperglycemia and GLP-1 stimulation (60), and in
the present study the response tended (P ⬍ 0.1) to be lower in
FDR than in CON subjects (Fig. 2). Thus, the ␤-cell response
to glucose is clearly diminished in FDR, and the response to
GLP-1 may be diminished. However, an isolated effect of
GLP-1 cannot be evaluated with the presently used experimental protocol.
It should be noted that the two independent tests of ␤-cell
secretion (IVGTT and hyperglycemic clamp) showed essentially the same results, i.e., lower insulin secretion in FDR than
in CON, and no effect of training. As for CON, the latter point
is in contrast to previous studies in young, healthy subjects (9,
28, 33, 35). In light of the observed increase in insulin action
on leg and whole body glucose uptake rates, we had expected
an adaptation of the glucose-stimulated insulin secretion. Furthermore, we did not find an effect of training on the GLP-1stimulated insulin response. This is in contrast to the training
effect, i.e., lower insulin response, that we previously found
with another non-glucose secretagogue, arginine (9). However,
a possible effect of training on GLP-1-stimulated insulin secretion may have been masked by the fact that insulin secretion
was already markedly stimulated by the hyperglycemia. A
likely explanation to the lack of training effect on insulin
secretion might be that CON subjects in the present study were
on average approximately 20 years older than those in the
previous studies (9, 35), and it is known that basal insulin
release declines with age (22) and that older individuals do not
exhibit the same compensatory increase in ␤-cell secretion as
young individuals after eccentric exercise training (29). In
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
found less efficient inhibition of in vivo lipolysis by insulin in
subcutaneous abdominal adipose tissue of FDR compared with
CON, and it thus seems that subcutaneous abdominal adipose
tissue metabolite turnover (glucose, lactate, lipids) in general is
increased in FDR. We found no significant differences in in
vitro basal and insulin-stimulated glucose incorporation into
triacylglycerol in adipocytes between FDR and CON, and a
previous in vitro study also did not detect metabolic defects in
adipocytes from FDR compared with CON (14). Apparently,
factors not present in vitro, e.g., blood flow, are of importance
for the differences in adipose tissue metabolism between FDR
and CON.
Endurance training increased in vivo skeletal muscle lipolysis, but no other clear effects of training were seen in adipose
tissue or skeletal muscle metabolism with the microdialysis
technique in either FDR or CON. This indicates that the
training program used in the present study was not sufficient to
markedly change resting metabolism in skeletal muscle or
adipose tissue.
E90
TRAINING: INSULIN SECRETION AND ACTION
FDR subjects, the ␤-cell adaptation to a training program is
more difficult to predict. In the present study, we found no
change in the response (in accord with the finding of no
increase in insulin-mediated glucose uptake rates); on the other
hand, it has previously been shown that in type 2 diabetic
patients with a relatively well-preserved ␤-cell capacity (i.e.,
“high-secretors”), the ␤-cell response to training is, in fact, an
increase in the ␤-cell secretory capacity (11). Thus, the training
intervention would tend to pull the ␤-cell secretory response
toward a decrease, whereas the genetic disposition to type 2
diabetes might pull the response toward an increase. Further
studies are needed to clarify this issue.
Summary
ACKNOWLEDGMENTS
We thank Regitze Kraunsøe, Jeppe Bach, and Lisbeth Kall for dedicated
technical assistance. We aso thank Claus Neumann MD for assistance with the
initial experiments.
GRANTS
The financial support from the Novo Nordisk Foundation, the Danish
Diabetes Association, the Lundbeck Foundation, the Jacob Madsens and Olga
Madsens Foundation, the Foundation of 1870, the Kathrine and Vigo Skovgaards Foundation, the Aase and Ejnar Danielsen Foundation, the Simon
Fougner Hartmanns Family Foundation, and a European Union grant (6th
Framework LSHM-CT-2004-005272, EXGENESIS) gratefully acknowledged.
DISCLOSURES
No conflicts of interest are reported by the authors.
REFERENCES
1. Abel ED, Peroni O, Kim JK, Kim YB, Boss O, Hadro E, Minnemann
T, Shulman GI, Kahn BB. Adipose-selective targeting of the GLUT4
gene impairs insulin action in muscle and liver. Nature 409: 729 –733,
2001.
2. Araújo-Vilar D, García-Estévez DA, Cabezas-Cerrato J. Insulin sensitivity, glucose effectiveness, and insulin secretion in nondiabetic offspring of patients with non-insulin-dependent diabetes mellitus: a crosssectional study. Metab Clin Exper 48: 978 –983, 1999.
3. Barwell ND, Malkova D, Moran CN, Cleland SJ, Packard CJ, Zammit
VA, Gill JM. Exercise training has greater effects on insulin sensitivity in
daughters of patients with type 2 diabetes than in women with no family
history of diabetes. Diabetologia 51: 1912–1919, 2008.
4. Boushel R, Gnaiger E, Schjerling P, Skovbro M, Kraunsoe R, Dela F.
Patients with type 2 diabetes have normal mitochondrial function in
skeletal muscle. Diabetologia 50: 790 –796, 2007.
5. Coppack SW, Frayn KN, Humphreys SM, Dhar H, Hockaday TD.
Effects of insulin on human adipose tissue metabolism in vivo. Clin Sci
77: 663–670, 1989.
AJP-Endocrinol Metab • VOL
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
We found that endurance exercise training for three months
in 40-year-old healthy men without a genetic disposition to
type 2 diabetes increased insulin-mediated glucose uptake rates
in whole body as well as the leg and that this increase was
dissociated from a change in the ␤-cell response to hyperglycemia with or without additional GLP-1 stimulation. In 40year-old men with a genetic disposition to type 2 diabetes, no
effect of training on insulin-mediated glucose uptake rates or
on the stimulated ␤-cell secretion was seen, but glucosemediated glucose uptake rates were preserved and increased
with training. Offspring subjects displayed impaired stimulated
␤-cell secretion and increased glucose uptake in subcutaneous
adipose tissue but not in quadriceps muscle.
6. Dela F, Handberg A, Mikines KJ, Vinten J, Galbo H. GLUT 4 and
insulin receptor binding and kinase activity in trained human muscle. J
Physiol 469: 615–624, 1993.
7. Dela F, Mikines KJ, Larsen JJ, Ploug T, Petersen LN, Galbo H. Insulin
stimulated muscle glucose clearance in patients with type 2 diabetes
mellitus. Effects of one-legged physical training. Diabetes 44: 1010 –1020,
1995.
8. Dela F, Mikines KJ, Linstow vM, Secher NH, Galbo H. Effect of
training on insulin mediated glucose uptake in human skeletal muscle. Am
J Physiol Endocrinol Metab 263: E1134 –E1143, 1992.
9. Dela F, Mikines KJ, Tronier B, Galbo H. Diminished arginine-stimulated insulin secretion in trained men. J Appl Physiol 69: 261–267, 1990.
10. Dela F, Ploug T, Handberg A, Petersen LN, Larsen JJ, Mikines KJ,
Galbo H. Physical training increases muscle GLUT-4 protein and mRNA
in patients with NIDDM. Diabetes 43: 862–865, 1994.
11. Dela F, Von Linstow ME, Mikines KJ, Galbo H. Physical training may
enhance ␤-cell function in type 2 diabetes. Am J Physiol Endocrinol
Metab 287: E1024 –E1031, 2004.
12. Doi K, Taniguchi A, Nakai Y, Kawamura H, Higaki Y, Yokoi H,
Tanaka H, Fujitani J, Suzuki M, Tokuyama K, Sakai M, Fukushima
M. Decreased glucose effectiveness but not insulin resistance in glucosetolerant offspring of Japanese non-insulin-dependent diabetic patients: a
minimal-model analysis. Metab Clin Exper 46: 880 –883, 1997.
13. Eriksson J, Franssila-Kallunki A, Ekstrand A, Saloranta C, Widen E,
Schalin C, Groop L. Early metabolic defects in persons at increased risk
for non-insulin-dependent diabetes mellitus. N Engl J Med 321: 337–343,
1989.
14. Eriksson JW, Smith U, Waagstein F, Wysocki M, Jansson, PA.
Glucose turnover and adipose tissue lipolysis are insulin-resistant in
healthy relatives of type 2 diabetes patients: is cellular insulin resistance a
secondary phenomenon? Diabetes 48: 1572–1578, 1999.
15. Fernandez-Castaner M, Biarnes J, Camps I, Ripolles J, Gomez N,
Soler J. Beta-cell dysfunction in first-degree relatives of patients with
non-insulin-dependent diabetes mellitus. Diabet Med 13: 953–959, 1996.
16. Frosig C, Rose AJ, Treebak JT, Kiens B, Richter EA, Wojtaszewski
JF. Effects of endurance exercise training on insulin signaling in human
skeletal muscle: interactions at the level of phosphatidylinositol 3-kinase,
Akt, and AS160. Diabetes 56: 2093–2102, 2007.
17. Hagstrom-Toft E, Arner P, Johansson U, Eriksson LS, Ungerstedt U,
Bolinder J. Effect of insulin on human adipose tissue metabolism in situ.
Interactions with beta-adrenoceptors. Diabetologia 35: 664 –670, 1992.
18. Handberg A, Vaag A, Vinten J, Beck-Nielsen H. Decreased tyrosine
kinase activity in partially purified insulin receptors from muscle of young,
non-obese first degreee relatives of patients with type 2 (non-insulindependent) diabetes mellitus. Diabetologia 36: 668 –674, 1993.
19. Henriksen JE, Alford F, Handberg A, Vaag A, Ward GM, Kalfas A,
Beck-Nielsen H. Increased glucose effectiveness in normoglycemic but
insulin- resistant relatives of patients with non-insulin- dependent diabetes
mellitus—a novel compensatory mechanism. J Clin Invest 94: 1196 –
1204, 1994.
20. Henriksen JE, Levin K, Thye-Ronn P, Alford F, Hother-Nielsen O,
Holst JJ, Beck-Nielsen H. Glucose-mediated glucose disposal in insulinresistant normoglycemic relatives of type 2 diabetic patients. Diabetes 49:
1209 –1218, 2000.
21. Holten MK, Zacho M, Gaster M, Juel C, Wojtaszewski JFP, Dela F.
Strength training increases insulin-mediated glucose uptake, GLUT4 content and insulin signaling in skeletal muscle in patients with Type 2
diabetes. Diabetes 53: 294 –305, 2004.
22. Iozzo P, Beck-Nielsen H, Laakso M, Smith U, Yki-Järvinen H, Ferrannini E. Independent influence of age on basal insulin secretion in
nondiabetic humans. European Group for the Study of Insulin Resistance.
J Clin Endocrinol Metab 84: 863–868, 1999.
23. Jacob S, Machann J, Rett K, Brechtel K, Volk A, Renn W, Maerker
E, Matthaei S, Schick F, Claussen CD, Haring HU. Association of
increased intramyocellular lipid content with insulin resistance in lean
nondiabetic offspring of type 2 diabetic subjects. Diabetes 48: 1113–1119,
1999.
24. Johanson EH, Jansson PA, Lonn L, Matsuzawa Y, Funahashi T,
Taskinen MR, Smith U, Axelsen M. Fat distribution, lipid accumulation
in the liver, and exercise capacity do not explain the insulin resistance in
healthy males with a family history for type 2 diabetes. J Clin Endocrinol
Metab 88: 4232–4238, 2003.
TRAINING: INSULIN SECRETION AND ACTION
AJP-Endocrinol Metab • VOL
44. Pimenta WP, Santos ML, Cruz NS, Aragon FF, Padovani CR, Gerich
JE. Insulin secretion, insulin sensitivity, and hepatic insulin extraction in
first-degree relatives of type 2 diabetic patients. Braz J Med Biol Res 36:
301–308, 2003.
45. Rabol R, Larsen S, Hojberg PM, Almdal T, Boushel R, Haugaard SB,
Andersen JL, Madsbad S, Dela F. Regional anatomic differences in
skeletal muscle mitochondrial respiration in type 2 diabetes and obesity. J
Clin Endocrinol Metab 95: 857–863, 2010.
46. Richter EA, Mikines KJ, Galbo H, Kiens B. Effect of exercise on insulin
action in human skeletal muscle. J Appl Physiol 66: 876 –885, 1989.
47. Sandqvist MM, Eriksson JW, Jansson PA. Increased lactate release per
fat cell in normoglycemic first-degree relatives of individuals with type 2
diabetes. Diabetes 50: 2344 –2348, 2001.
48. Scheller D, Kolb J. The internal reference technique in microdialysis: a
practical approach to monitoring dialysis efficiency and to calculating
tissue concentration from dialysate samples. J Neurosci Methods 40:
31–38, 1991.
49. Sonne MP, Alibegovic AC, Hojbjerre L, Vaag A, Stallknecht B, Dela
F. The effect of ten days of bed rest on metabolic and vascular insulin
action. A study in individuals at risk for type 2 diabetes. J Appl Physiol Jan
28 [Epub ahead of print].
50. Sonne MP, Hojbjerre L, Alibegovic AA, Vaag A, Stallknecht B, Dela
F. Impaired endothelial function and insulin action in first-degree relatives
of patients with type 2 diabetes mellitus. Metabolism 58: 93–101, 2009.
51. Stallknecht B, Andersen PH, Vinten J, Bendtsen LL, Sibbersen J,
Pedersen O, Galbo H. Effect of physical training on glucose transporter
protein and mRNA levels in rat adipocytes. Am J Physiol Endocrinol
Metab 265: E128 –E134, 1993.
52. Stallknecht B, Larsen JJ, Mikines KJ, Simonsen L, Bülow J, Galbo H.
Effect of training on insulin sensitivity of glucose uptake and lipolysis in
human adipose tissue. Am J Physiol Endocrinol Metab 279: E376 –E385,
2000.
53. Stallknecht B, Simonsen L, Bülow J, Vinten J, Galbo H. Effect of
training on epinephrine-stimulated lipolysis determined by microdialysis
in human adipose tissue. Am J Physiol Endocrinol Metab 269: E1059 –
E1066, 1995.
54. Storgaard H, Song XM, Jensen CB, Madsbad S, Björnholm M, Vaag
A, Zierath JR. Insulin signal transduction in skeletal muscle from
glucose-intolerant relatives of type 2 diabetic patients. Diabetes 50:
2770 –2778, 2001.
55. Straczkowski M, Kowalska I, Baranowski M, Nikolajuk A, Otziomek
E, Zabielski P, Adamska A, Blachnio A, Gorski J, Gorska M. Increased skeletal muscle ceramide level in men at risk of developing type
2 diabetes. Diabetologia 50: 2366 –2373, 2007.
56. Vaag A, Henriksen JE, Beck-Nielsen H. Decreased insulin activation of
glycogen synthase in skeletal muscles in young nonobese caucasian
1st-degree relatives of patients with non-insulin-dependent diabetes mellitus. J Clin Invest 89: 782–788, 1992.
57. Valentino R, Lupoli GA, Raciti GA, Oriente F, Farinaro E, Della VE,
Salomone M, Riccardi G, Vaccaro O, Donnarumma G, Sesti G, Hribal
ML, Cardellini M, Miele C, Formisano P, Beguinot F. The PEA15 gene
is overexpressed and related to insulin resistance in healthy first-degree
relatives of patients with type 2 diabetes. Diabetologia 49: 3058 –3066,
2006.
58. Van Haeften TW, Dubbeldam S, Zonderland ML, Erkelens DW.
Insulin secretion in normal glucose-tolerant relatives of type 2 diabetic
subjects. Assessments using hyperglycemic glucose clamps and oral
glucose tolerance tests. Diabetes Care 21: 278 –282, 1998.
59. Vauhkonen I, Niskanen L, Vanninen E, Kainulainen S, Uusitupa M,
Laakso M. Defects in insulin secretion and insulin action in non-insulindependent diabetes mellitus are inherited. Metabolic studies on offspring
of diabetic probands. J Clin Invest 101: 86 –96, 1998.
60. Vilsboll T, Toft-Nielsen MB, Krarup T, Madsbad S, Dinesen B, Holst
JJ. Evaluation of beta-cell secretory capacity using glucagon-like peptide
1. Diabetes Care 23: 807–812, 2000.
61. Volk A, Renn W, Overkamp D, Mehnert B, Maerker E, Jacob
Balletshofer B, Haring HU, Rett K. Insulin action and secretion in
healthy, glucose tolerant first degree relatives of patients with type 2
diabetes mellitus. Influence of body weight. Exper Clin Endocrinol Diabet
107: 140 –147, 1999.
299 • JULY 2010 •
www.ajpendo.org
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.4 on June 17, 2017
25. Johnston C, Ward WK, Beard JC, McKnight, Porte D Jr. Islet
function and insulin sensitivity in the non-diabetic offspring of conjugal
type 2 diabetic patients. Diabet Med 7: 119 –125, 1990.
26. Kahn CR. Insulin resistance, insulin insensitivity, and insulin unresponsiveness: a necessary distinction. Metabolism 27: 1893–1902, 1978.
27. Kashyap SR, Belfort R, Berria R, Suraamornkul S, Pratipranawatr T,
Finlayson J, Barrentine A, Bajaj M, Mandarino L, DeFronzo R, Cusi
K. Discordant effects of a chronic physiological increase in plasma FFA
on insulin signaling in healthy subjects with or without a family history of
type 2 diabetes. Am J Physiol Endocrinol Metab 287: E537–E546, 2004.
28. King DS, Dalsky GP, Staten MA, Clutter WE, Van Houten DR,
Holloszy JO. Insulin action and secretion in endurance-trained and untrained humans. J Appl Physiol 63: 2247–2252, 1987.
29. Krishnan RK, Hernandez JM, Williamson DL, O’Gorman DJ, Evans
WJ, Kirwan JP. Age-related differences in the pancreatic ␤-cell response
to hyperglycemia after eccentric exercise. Am J Physiol Endocrinol Metab
275: E463–E470, 1998.
30. Larsen OA, Lassen NA, Quaade F. Blood flow through human adipose
tissue determined with radioactive xenon. Acta Physiol Scand 66: 337–
345, 1966.
31. Larsen S, Ara I, Rabol R, Andersen JL, Boushel R, Dela F, Helge JW.
Are substrate use during exercise and mitochondrial respiratory capacity
decreased in arm and leg muscle in type 2 diabetes? Diabetologia 52:
1400 –1408, 2009.
32. Meyer MM, Levin K, Grimmsmann T, Beck-Nielsen H, Klein HH.
Insulin signalling in skeletal muscle of subjects with or without Type II
diabetes and first degree relatives of patients with the disease. Diabetologia 45: 813–822, 2002.
33. Mikines KJ. The influence of physical activity and inactivity on insulin
action and secretion in man. Acta Physiol Scand 146, Suppl 609: 1–43,
1992.
34. Mikines KJ, Sonne B, Farrell PA, Tronier B, Galbo H. Effect of
physical exercise on sensitivity and responsiveness to insulin in humans.
Am J Physiol Endocrinol Metab 254: E248 –E259, 1988.
35. Mikines KJ, Sonne B, Tronier B, Galbo H. Effects of training and
detraining on dose-response relationship between glucose and insulin
secretion. Am J Physiol Endocrinol Metab 256: E588 –E596, 1989.
36. Mitrakou A, Vuorinen-Markkola H, Raptis G, Toft I, Mokan M,
Strumph P, Pimenta W, Veneman T, Jenssen T, Bolli G, Korytkowski
M, Yki-Järvinen H, Gerich J. Simultaneous Assessment of Insulin
Secretion and Insulin Sensitivity Using a Hyperglycemic Clamp. J Clin
Endocrinol Metab 75: 379 –382, 1992.
37. Morino K, Petersen KF, Dufour S, Befroy D, Frattini J, Shatzkes N,
Neschen S, White MF, Bilz S, Sono S, Pypaert M, Shulman GI.
Reduced mitochondrial density and increased IRS-1 serine phosphorylation in muscle of insulin-resistant offspring of type 2 diabetic parents. J
Clin Invest 115: 3587–3593, 2005.
38. Nyholm B, Porksen N, Juhl CB, Gravholt CH, Butler PC, Weeke J,
Veldhuis JD, Pincus SM, Schmitz O. Assessment of insulin secretion in
relatives of patients with type 2 (non-insulin-dependent) diabetes mellitus:
evidence of early beta-cell dysfunction. Metab Clin Exper 49: 896 –905,
2000.
39. Osei K, Cottrell DA, Orabella MM. Insulin sensitivity, glucose effectiveness, and body fat distribution pattern in nondiabetic offspring of
patients with NIDDM. Diabetes Care 14: 890 –896, 1991.
40. Østergård T, Andersen JL, Nyholm B, Lund S, Nair KS, Saltin B,
Schmitz O. Impact of exercise training on insulin sensitivity, physical
fitness, and muscle oxidative capacity in first-degree relatives of type 2
diabetic patients. Am J Physiol Endocrinol Metab 290: E998 –E1005,
2006.
41. Perseghin G, Ghosh S, Gerow K, Shulman GI. Metabolic defects in lean
nondiabetic offspring of NIDDM parents: a cross-sectional study. Diabetes 46: 1001–1009, 1997.
42. Perseghin G, Scifo P, De Cobelli F, Pagliato E, Battezzati A, Arcelloni
C, Vanzulli A, Testolin G, Pozza G, Del M, Luzi L. Intramyocellular
triglyceride content is a determinant of in vivo insulin resistance in
humans: a 1H-13C nuclear magnetic resonance spectroscopy assessment
in offspring of type 2 diabetic parents. Diabetes 48: 1600 –1606, 1999.
43. Petersen KF, Dufour S, Befroy D, Garcia R, Shulman GI. Impaired
mitochondrial activity in the insulin-resistant offspring of patients with
type 2 diabetes. N Engl J Med 350: 664 –671, 2004.
E91