Growth Hormone & IGF Research 20 (2010) 380385

Contents lists available at ScienceDirect

Growth Hormone & IGF Research

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / g h i r

Effect of high- and low-intensity exercise and metabolic acidosis on levels of GH,
IGF-I, IGFBP-3 and cortisol
Patrick Wahl a,b,c,, Christoph Zinner a, Silvia Achtzehn a, Wilhelm Bloch b,c, Joachim Mester a,c
a
Institute of Training Science and Sport Informatics, German Sport University, Cologne, Germany
b
Department of Molecular and Cellular Sport Medicine, Institute of Cardiovascular Research and Sport Medicine, German Sport University Cologne, Germany
c
The German Research Centre of Elite Sport, German Sport University Cologne, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Objective: The purpose of the present study was to examine the acute hormonal response of a short term
Received 2 June 2010 high-intensity training (HIT) versus a high volume endurance training (HVT) and to determine the
Received in revised form 28 July 2010 contribution of the metabolic acidosis as a stimulus for possibly different reactions of circulating hGH, IGF-1,
Accepted 2 August 2010 IGFBP-3 and cortisol.
Available online 30 August 2010
Design: Eleven subjects participated in three experimental trials separated by one week. Two times subjects
performed four 30 s maximal effort exercise bouts on a cycle ergometer separated by 5 min rest each. Before
Keywords:
Acidosis
the exercise subjects either received (single-blinded) bicarbonate (HIT (B)) or a placebo (HIT (P)). The third
Hormone secretion exercise trail consisted of a constant load exercise for 1 h at 50% VO2max (HVT). Venous blood samples were
Cortisol taken under resting conditions, 10 min, 60 min and 240 min after each exercise condition to determine hGH,
hGH-IGF-1 axis IGF-1, IGFBP-3 and cortisol serum concentrations. Capillary blood samples were taken to determine lactate
High-intensity training concentrations and blood gas parameters.
High volume training Results: Power output, mean lactate concentrations and mean pH values were signicantly higher during HIT (B)
compared to HIT (P). Serum cortisol and hGH concentrations were signicantly increased 10 min post exercise in
both HIT interventions. IGFBP-3 was only signicantly increased after HIT (P), whereas IGF-1 was not affected by
any of the interventions. HVT showed no signicant effects on cortisol, hGH, IGF-1 and IGFBP-3 levels. Additionally
it was shown that the diminished acidosis during HIT (B) attenuates the cortisol and hGH response.
Conclusions: The present study suggests that HIT/acidosis is a stimulus for exercise-induced cortisol/hGH secretion,
but not for IGF-1 and IGFBP-3 under these experimental conditions. These ndings might be relevant for
arrangements of interval training, due to the fact that active or passive recovery during rest periods inuence the
acid base status and may therefore inuence the hormonal response.
2010 Growth Hormone Research Society. Published by Elsevier Ltd. All rights reserved.

1. Introduction contribution of local and systemic IGF-1 in mediating the outcomes of

Growth hormone (GH) is the principal regulator of the hepatic
synthesis of insulin-like growth factor I (IGF-1) and IGF-binding
protein 3 (IGFBP-3), which is the major IGF carrier and modulator of
IGF-1 action in plasma [1]. IGF-1 itself is the primary downstream
mediator of GH actions, and circulating IGF-1 plays an important role
in the feedback regulation of GH secretion [2]. On the one hand, IGF-1
is produced by the liver in an endocrine fashion, which acts
systemically, on the other hand IGF-1 is produced locally (i.e. in
skeletal muscle) to act in an autocrine/paracrine fashion. However
IGF-1, produced in skeletal muscle during exercise, is also released
into the circulation which might contribute to an increase in
circulating IGF-1 levels as well [36]. However, the relationship and
Corresponding author. Institute of Training Science and Sport Informatics, German
Sport University Cologne, Am Sportpark Mngersdorf 6, 50933 Cologne, Germany.
Tel.: + 49 221 49826071; fax: + 49 221 49828180.
E-mail address: Wahl@dshs-koeln.de (P. Wahl).
physical activity are still not fully understood [7]. IGF-1 has
widespread anabolic and insulin-sensitizing effects, and plays a
critical role in formation, maintenance, and regeneration of skeletal
muscles [1]. IGF-1 also plays a direct role in whole body glucose
homeostasis primarily by stimulating skeletal muscle glucose uptake
[3]. In contrast GH is a counterregulatory hormone that antagonizes
the hepatic and peripheral effects of insulin on glucose metabolism
[8]. Nindl et al. stated that IGF-1 is an important metabolic biomarker
associated with a variety of health- and exercise-related outcomes [5].
In the circulation only ~ 2% of IGF-1 circulates in free form. Most of
IGF-1 is bound to one of six IGF-1 binding proteins (IGFBP). An altered
afnity and association/disassociation constants under different
environmental conditions (e.g. exercise) inuence the bioavailability
of IGF-1 and the extent of interaction with IGF-1 receptors on the cell
surface. Furthermore only free IGF-1 can pass from the vasculature
to the interstitium [7].
It is well recognized that exercise has a signicant impact on the
GH/IGF system. Endurance training can induce an increase of hGH,

1096-6374/$ see front matter 2010 Growth Hormone Research Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ghir.2010.08.001
 P. Wahl et al. / Growth Hormone & IGF Research 20 (2010) 380385 381

IGF-1, and IGFBP-3 levels in the circulation, but less is known about
the effects of different training intensities (e. g. high-intensity training
(HIT) or high volume low-intensity training (HVT)) on circulating
levels of these growth factors [3]. Only a few studies have addressed
this issue in a strictly experimental way [9,10].
Although pH is generally well regulated, a more dramatic increase
in the acidity of the circulating blood and the skeletal muscle occurs
when performing HIT compared to HVT. One can speculate that these
systemic and local changes in the extracellular environment might
inuence the release, the afnity, and association/dissociation of GH,
IGF-1 and IGFBP-3. The extracellular pH has been recognized to
regulate the IGF-1 interactions with different cells, components of the
extracellular matrix and with IGFBP, which could modulate IGF-1
activity and bioavailability [11]. But only few in vivo data are available.
The present study therefore focused on the one hand on different
exercise intensities (HIT versus HVT) and on the other hand on the
effects of the acid base status on circulating GH, IGF-1, and IGFBP-3
concentrations. Furthermore, to quantify the induced stress by the
different exercise protocols circulating cortisol levels were measured.
2. Materials and methods
2.1. Subjects
Eleven healthy male subjects (age: 26.5 5.6 yr; height: 182
7.2 cm; weight: 76.1 8.2 kg; VO2max 64.6 7.1 mL kg1 min1)
participated in this study. The subjects were drawn from the
population of healthy male sport students. All subjects gave written
informed consent to the contribution to the study and the research
protocol was approved by a local ethical committee.
Before the participation, subjects performed a ramp test to
determine VO2max (Zan 600, Zan Messgerte, Oberthulba, Germany),
maximal performance and the proper warming-up intensity for each
subject for the experiments. The ramp test consisted of a 10 min
warm-up at 1.5 W kg1 body weight followed by a 5 min rest period.
Afterwards the ramp started with 2 min at 2.0 W kg1 body weight
followed by an increase of 20 W every 30 s until exhaustion.
2.2. Exercise study protocol
Subjects participated in three experimental trials, each separated
by one week. In the rst two weeks, the same two high-intensity
interval training (HIT) interventions (1. HIT bicarbonate [HIT (B)] and
2. HIT placebo [HIT (P)]) were carried out in a randomised order,
followed by the high volume endurance training (HVT) intervention
in the third week.
The HIT interventions consisted of four 30 s maximal effort (all-
out) exercise bouts on a cycle ergometer (Schoberer Rad Metechnik
SRM GmbH, Jlich, Germany) adjusted to an isokinetic mode set
to a cadence of 120 rpm, separated by 5 min passive rest each
(whereas subjects either received bicarbonate or a placebo be-
fore the exercise) (Fig. 1). Subjects were instructed to perform
the tests in a sitting position on the ergometer. During the HIT
interventions all subjects were vocally encouraged to achieve
maximal power output in the same way. The third exercise
trail (HVT) consisted of a constant load exercise for 1 h at 50%
VO2max.
For each 30 s all-out exercise bout of both HIT interventions, mean
power output (MP) and the fatigue index (FI) (power decline) were
calculated. FI was calculated by the following formula: FI = [(Peak
Power Lowest Power) 100] Peak Power1.
For both HIT and the HVT intervention, total energy expenditure
[kJ] and total work [kJ] were determined by the following formula:
energy expenditure: kJ = 4.1868 [(VO2 3.9) + (VCO2 1.1)]; total
work: kJ = (MP [W] exercise time [s]) 10001 [12].
Before each experimental trial, subjects warmed up for 10 min at
an intensity of 50% VO2max, followed by a 5 min passive rest on the
cycle ergometer. During each session environmental conditions
(temperature and humidity) were kept constant. Always the same
two investigators attended the tests. During each session spirometric
data were determined.
2.3. Ingestion/experimental treatment
The subjects refrained from strenuous physical exercise for 1 d
before each experimental session. The subjects were also advised to
avoid coffee and/or carbonated/acid beverages/food.
After the arrival participants ingested either a sodium bicarbonate
(NaHCO3) solution (0.3 g kg1 body wt suspended in 0.02 mL wa-
ter kg1 body wt) or a placebo (CaCO3) over a 90 min period before
the two HIT interventions [13]. The administration was randomised
and blind for the subjects, but not for the testers. Before the HVT
intervention no substance was ingested.

Fig. 1. The HIT protocol lasted a total of 42 min (whereas absolute exercise time was 2 min) and consisted of a 10 min warming up (WU) at 50% of VO2max followed by four 30 s all-
out exercise bouts separated by 5 min passive rest and a passive recovery period of 10 min after the last bout. Capillary blood samples for blood gas analysis were taken before the
ingestion (90), after ingestion (pre), after warming up (WU), and 30 s (30) and 4 min (4) after each 30 s bout. Blood samples for lactate analysis were taken before the ingestion
(90), after ingestion (pre), after warming up (WU), directly (0), 3 min (3), and 4:30 min (4:30) after each 30 s bout. Venous blood samples for hormone analysis were taken
before ingestion (90), before exercise (pre), 10, 60 and 240 post exercise.
382 P. Wahl et al. / Growth Hormone & IGF Research 20 (2010) 380385

2.4. Measurements 3. Results

Capillary samples from the earlobe were collected for blood gas (AVL
Omni 6; Roche Diagnostics GmbH, Mannheim, Germany) and lactate
analysis (EBIOplus; EKF Diagnostic Sales, Magdeburg, Germany) at the
following points of time for the HIT interventions (Fig. 1):
1. blood gas: pre ingestion, post ingestion, post warming-up (WU)
and 30 s/4 min after each 30 s exercise bout.
2. lactate: pre ingestion, post ingestion, post WU and 0 s/3 min/
4:30 min after each 30 s exercise bout.
For the HVT intervention blood samples were taken:
1. blood gas/lactate: pre, post WU, 15 min, 30 min, 45 min, 60 min.
Lactate concentrations [La] and blood gas parameter (pH, HCO
base excess (BE)) were measured in plasma. For lactate determination
in plasma 115 L of blood were withdrawn from the earlobe and
directly centrifuged for 1 min at 6000 rpm with EBA20 (Hettich
Zentrifugen, Tuttlingen, Germany). It took less than 30 s to take the
blood sample and to start centrifugation. Afterwards 20 L of plasma
were analysed. All 20 L samples were directly mixed with 1 mL of the
3.1. Power output, total work, energy expenditure, oxygen uptake and
RER
The results for the mean power (MP) output for the four HIT
intervals as well as for the fatigue index (FI) (power decline) are
shown in Table 1.
MP was signicantly higher in third (p = 0.007) and fourth
(p = 0.000) bouts after bicarbonate ingestion. The power decline
was signicantly lower during the last interval (p = 0.041). Total work
performed [kJ], total energy expenditure [kJ] and total oxygen
consumption [L] were signicantly higher in the HVT intervention
compared to HIT. Total work and energy expenditure tended to be
higher in HIT (B) (p = 0.059 and p = 0.055) compared to HIT (P)
without reaching statistical signicance, most likely due to the
3 ,
signicantly higher power output in the 3rd and 4th bouts after
bicarbonate ingestion. Mean respiratory exchange ratio [RER] was
signicantly higher in both HIT interventions compared to HVT, but
showed no differences among each other (Table 1).
3.2. Lactate and blood gas analysis (pH, HCO
3 , BE)

EBIO plus system solution. For blood gas analysis 140 L of blood was
withdrawn from the earlobe and directly measured.
Venous samples were collected for the determination of cortisol,
hGH, IGF-1 and IGFBP-3. One venous blood sample was taken before
ingestion (90; only HIT), before exercise (pre), and three samples
were taken at 10 min (10), 60 min (60), and 240 min (240) post
exercise. Nine and a half millilitres of blood was collected by the
Mean [La]plasma was signicantly higher in HIT (B) (17.9 2.1) and
HIT (P) (16.4 1.9) versus HVT (1.7 0.8) and signicantly higher in
HIT (B) versus HIT (P) at respective points of time. Fig. 2A shows the
time course of [La]plasma for HIT (B) and HIT (P). No changes in
[La]plasma were found during HVT.
Capillary pH, HCO3 , and BE are shown in Fig. 2BD. Prior to the start

Vacutainer blood withdrawal system (Becton Dickinson). After of ingestion (90) there were no signicant differences between the
storage at 7 C for ~ 30 min for deactivation of coagulation factors, trials in any of the blood gas parameters. Signicant increases in pH,
the blood samples were centrifuged for 10 min at 1.861 g and 4 C HCO 3 and BE were found after ingestion of bicarbonate but not after

(Rotixa 50, Hettich Zentrifugen, Mhlheim, Germany). The serum was placebo. Exercise induced a signicant decrease in capillary pH, HCO 3 ,

stored at 80 C till analysis. Serum levels of cortisol (ngmL1), hGH BE in both trials. Signicantly higher pH, HCO 3 , and BE were

(mLUmL1), IGF-1 (ngmL1) and IGFBP-3 (ngmL1) were
determined by using human ELISA kits (Cortisol ELISA EIA-1887,
hGH ELISA EIA-3552, and IGFBP-3 ELISA EIA-3300; DRG Instruments
GmbH Germany, Quantikine Human IGF-1 Immunoassay (Cata-
lognr.: DG100), R&D Systems, Inc. USA).
2.5. Statistical analysis
Statistical analyses of the data were performed by using a statistics
software package (Statistica for Windows, 7.0, Statsoft, Tulsa, OK).
Descriptive statistics of the data are presented as means SD. For the
comparison of different terms, repeated-measures ANOVA with
Bonferroni post hoc test was used, unless described otherwise.
Statistical differences were considered to be signicant for p b 0.05.
found during and after HIT (B) in comparison to HIT (P) (Table 1 and
Fig. 2BD). Blood gas parameter did not change during HVT.
3.3. Growth factor analysis
3.3.1. Cortisol
The ingestion of bicarbonate/placebo or rather the resting phase
signicantly decreased cortisol levels to the same extent (90 pre;
HIT (B): 45 37 pg mL1 (p = 0.011); HIT (P): 47 34 pg mL1
(p = 0.006)). Pre-values of all three interventions were similar and
showed no signicant differences. Both HIT interventions showed a
signicant increase in cortisol levels pre 10 post exercise (pre 10;
HIT (B): +61 54 pg mL1 (p = 0.000); HIT (P): +101 37 pg mL1
(p = 0.000)), whereas the HVT intervention showed no signicant

Table 1
Overview of the three exercise interventions. Values are means SD. *signicantly different from HIT (P) (p b 0.05); signicantly different from HVT (p b 0.05).

HIT (B) HIT (P) HVT

Total Time 4 * 30 s (= 2 min) 4 * 30 s (= 2 min) 60 min

Intensity All out All out 50% VO2max

Mean Bout 1 742 221 757 236 192 21

Power Bout 2 674 206 678 219
[Watt] Bout 3 605 194* 560 207
Bout 4 597 201* 525 180
Fatigue Index [%] Bout 1 55.9 10.1 57.2 9.1
Bout 2 57.0 8.4 60.3 10.2
Bout 3 59.2 9.9 62.8 11.6
Bout 4 59.8 12.1* 66.3 13.9
Total work [kJ] 78.6 0.2 75.5 0.2 692.5 77.1
Total energy expenditure [kJ] 1022 134 967 109 3672 662
Total oxygen consumption [L] 39.3 5.5 37.4 4.3 179.4 32.6
Mean RER 1.28 0.03 1.26 0.05 0.90 0.05
 P. Wahl et al. / Growth Hormone & IGF Research 20 (2010) 380385 383

Fig. 2. A: [La]plasma and BD: pH, HCO3 , and BE during HIT (B) (squares, broken line) and HIT (P) (circles, solid line).
+
Signicant (+p b 0.05) increase/decrease compare to the
sampling point pre; *signicant difference (*p b 0.05) between the trials at respective sampling times. White bars represent the 30 s exercise bouts. Sampling times for lactate and
blood gases see Fig. 1. Values are presented as means SD.

changes in cortisol levels (19 51 pg mL1). Cortisol concentrations 3.3.2. hGH

10 post exercise, were signicantly different between all three
conditions. 60 post exercise cortisol levels were still signicantly
higher after HIT (P) compared to pre values, whereas no signicant
difference was found for HIT (B) compared to pre values anymore.
Cortisol concentrations were signicantly decreased below pre values
60 post exercise after HVT. After 240 cortisol levels were signicantly
decreased below pre values in all three conditions (Fig. 3).
The ingestion of bicarbonate or placebo had no inuences on resting
hGH levels. The resting values (pre) of all three interventions were
similar and showed no signicant differences. Both HIT interventions
showed a signicant increase in hGH levels 10 post exercise (pre 10;
HIT (B): +59 25 pg mL1 (p = 0.000); HIT (P): +72 21 pg mL1
(p = 0.000)), whereas the HVT intervention showed no signicant

Fig. 3. Changes in circulating cortisol protein content at different time points for HIT (P) Fig. 4. Changes in circulating hGH protein content at different time points for HIT (P)
(circles, solid line), HIT (B) (squares, broken line), and HVT (triangles, dotted line). (circles, solid line), HIT (B) (squares, broken line), and HVT (triangles, dotted line). hGH
Cortisol concentrations pre ingestion (90) were only determined for both HIT concentrations pre ingestion (90) were only determined for both HIT interventions.
interventions. Due to the ingestion/resting phase, cortisol levels signicantly decrease Signicant increases (*p b 0.05) pre 10 post exercise were only observed for HIT (P)
in both HIT interventions (90 pre) (+p b 0.05). Signicant difference compared to and (B). hGH concentrations 10 post exercise were signicantly different between all
pre values (*p b 0.05). Signicant difference between interventions at respective time three conditions (#p b 0.05) (HIT (B) versus HIT (P) was calculated with Fisher post hoc
points (#p b 0.05). Values are presented as means + or SD. test; p = 0.003). Values are presented as means + or SD.
384 P. Wahl et al. / Growth Hormone & IGF Research 20 (2010) 380385
changes in hGH levels (+17 15 pg mL1). hGH concentrations 10
post exercise were signicantly higher in HIT (P) and (B) compared to
HVT and signicantly different between HIT (B) and HIT (P). 60 after
both HIT interventions, hGH levels decreased signicantly, whereas
hGH levels after HIT (P) were still higher than hGH levels 10 after HVT
(Fig. 4). 240 after exercise hGH returned back to baseline levels.
3.3.3. IGF-1 and IGFBP-3
IGF-1 showed neither signicant changes for both HIT interven-
tions, nor for HVT.
The ingestion of bicarbonate/placebo had no signicant inuences
on resting IGFBP-3 levels. The resting values (pre) of all three
interventions showed no signicant differences. HIT (P) showed a
signicant increase (p = 0.002) in IGFBP-3 levels pre 10 post
exercise (HIT (P): +411 184 pg mL1), whereas HIT (B) and HVT
showed no signicant changes in IGFBP-3 levels (HIT (B): +310
406 pg mL 1; HVT: +139 238 pg mL1). The overall comparison
from ANOVA showed no signicant differences (p = 0.43) between the
three different interventions.
3.3.4. Inuence of haemoconcentration on growth factor levels
In the present study, increases in haematocrit were observed during
the exercise sessions compared to pre values (pre 10: +3.7 1.3%
HIT (P) and +2.9 1.4% (HIT (B)). If the present results are adjusted for
changes in plasma volume (PV) [PV changes in percentage of pre
values = {(Hbpre / Hbpost) (100 Hctpost) / [(100 Hctpre] 1)} 100]
[14], the observed statistical differences remained unchanged for all IGF-
1, cortisol and hGH values, except that cortisol values 10 post between
HIT (B) and HIT (P) (p = 0.085) and hGH values 10 post between HIT
(B) and HVT (p = 0.103) did not differ anymore. In contrast, no
signicant changes can be observed for IGFBP-3 anymore, if the values
are adjusted for changes in PV; neither for each intervention, nor
between interventions at respective points of measurement.
4. Discussion
Endurance training can induce an increase of hGH, IGF-1, and
IGFBP-3 levels in the circulation, but less is known about the effects of
different training intensities on the release of aforementioned
hormones [3,9,10]. Since a dramatic increase in the acidity of the
circulating blood results when HIT is performed [15], we hypothesized
that HIT and HVT and the physiological cue(s) of acidosis might have
(different) effects on the serum concentrations of cortisol, hGH, IGF-I,
and IGFBP-3. The major nding from the present study was that HIT
stimulates a transient increase in circulating levels of hGH, IGFBP-3,
and cortisol whereas HVT had no inuence on these parameters,
despite a much higher training volume and time commitment.
Additionally it was shown that alterations in the acid base status
blunt the hGH and cortisol response to exercise. IGF-1 was not affected
by any of the three interventions. By design, the absolute exercise time
was 97% lower in the HIT group (2 min versus 60 min) and
necessitated a training time commitment that was only ~one-half of
that of the HVT training (37 versus 70 min). Most of the training time
during the HIT intervention was spent in recovery between short,
intense bursts of all-out cycling.
Exercise stress, evaluated by cortisol levels, was highest in HIT (P)
showing an increase of cortisol levels of +218 106%. Bicarbonate
ingestion (HIT (B)) signicantly attenuated the cortisol response with
increases of +177 80%, suggesting that changes in the acid base status
partly caused the cortisol response to high-intensity exercise. HVT
even decreased circulating cortisol levels. These results indicate
that exercise intensity positively deects cortisol concentrations,
whereas exercise duration has no cumulative effect on the cortisol
response. This is in accordance with previously published studies
[1618].
Several studies provided evidence for a physiological role of pH in
the hormonal responses to exercise [15,19]. Gordon et al. [15]
demonstrated that with a 90 s maximal exercise bout, the post-
exercise hGH responses was reduced by an orally induced alkalosis
which is in consistence with our results. An decrease in the blood pH
may therefore be partly responsible for the hGH response to acute
high-intensity anaerobic exercise. In contrast Sutton et al. found no
relationship between hGH secretion and the blood pH [20]. Further-
more a previous study showed that the exercise-related GH release, to
a large extent, is induced by synchronous hyperthermia, rather than
by exercise itself [21]. As we did not measure core temperature in this
study, we can only speculate that the rise in core temperature was
similar in both HIT interventions, which does not explain the
differences between placebo and bicarbonate. Possibly, both stimuli
hyperthermia and pH partly induce the release of hGH. Another
limitation of the present study is the lack of a no exercise control
group, as both cortisol and hGH have distinct diurnal rhythms. As all
three trials were carried out at the same time of day, diurnal
differences were avoided between the conditions but this does not
take diurnal increases/decreases into account. Even though the
physiological effects of hGH and cortisol (growth, turnover of muscle,
bone and collagen, regulation of metabolic function) are well known,
we can only speculate that an acute increase, as shown in the present
study, might play a positive role in optimising training adaptation and
in eliciting health benets [22,23]. Both hormones may contribute to
the positive effects of HIT as it has been shown by previous studies
using similar exercise protocols [24].
The quantication of (total) IGF-1 levels in serum or plasma has
yielded inconsistent results, with levels being reported to decline [25],
to increase [10,26,27], or to remain unchanged [28,29] after the onset
of exercise. Thereby studies either investigated low-intensity or high-
intensity exercise. Only a few studies compared the IGF-1 response to
different exercise protocols in a strictly experimental way [9,10].
Schwartz et al. demonstrated that 10 min of high-intensity cycle
ergometry resulted in a more pronounced increase in IGF-1 (13.3
3.2%) and IGFBP-3 (23 6%) than 10 min of low-intensity exercise
(7.7 2.7% and 12.5 3.3% respectively) [10]. Copeland et al. com-
pared continuous moderate-intensity exercise (6065% VO2max) and
high-intensity interval exercise (8085% VO2max) of equal duration
(20 min) and found similar increases for IGF-1 and IGFBP-3 [9]. Both
interventions, however, do not reect commonly used training
regimes. On the one hand 1020 min are very short for low-intensity
exercise, on the other hand the absolute training time of moderate
continuous exercise normal differs from high-intensity interval
training, in contrast to the studies of Copeland et al. [9] and Schwarz
et al. [10]. Therefore the present study compared a HVT (60 min) with
a HIT (2 min). As aforementioned, no signicant changes in IGF-1
levels, neither during HIT nor during HVT, and a 12.4 6.1% increase in
IGFBP-3 only after HIT (P) were found in the present study.
The buffering of hydrogen ions with bicarbonate reduces the degree
of acidaemia. Therefore the ingestion of bicarbonate is considered to be
functional in maintaining homeostasis. In the present study, pH was
signicantly higher during HIT (B) at respective points in time
compared to HIT (P). From present data it might be assumed that
metabolic acidosis slightly affects the IGFBP-3 in humans, as the only
signicant increase was found after HIT (P) but not after HIT (B) or HVT.
In fact it was shown that acidosis consistently affects the binding of IGF-
1 to cells and the ECM, which is altered under acidic conditions [3]. Cell
culture experiments showed on the one hand, that IGF-1 binding to IGF-
IR is decreased at low pH [30], on the other hand cell association of
IGFBP-3 was increased at low pH which binds IGF-1. IGF-1 binding at
reduced pH is therefore dominated by IGFBPs and not IGF-IR [11].
Furthermore other studies demonstrated that lowering of pH may
be associated with the activation of IGFBP-3 proteases such as
cathepsin D [31,32]. Cathepsin D seems to play a role in the regulation
of cellular IGF action by its potential to alter the structure and function
 P. Wahl et al. / Growth Hormone & IGF Research 20 (2010) 380385
of IGFBP-3 [31]. Acute exercise increased IGFBP-3 proteolytic activity
by 44 14% which may contribute to anabolic effects of physical
activity by increasing the bioavailability of IGF-1 [10]. In contrast Dall
et al. provided evidence that IGFBP-3 proteolysis was unchanged after
a maximal exercise test in elite rowers [33]. If a proteolysis of IGFBP-3
by cathepsin D after exercise is really existent, this might explain the
increase in IGFBP-3 found in this investigation. But anyhow Kraemer
et al. showed that changing the magnitude of blood acidosis by using
voluntary bicarbonate ingestion to induce alkalosis did not reduce/
change the serum concentrations of IGF-1 and IGFBP-3 after volitional,
maximal cycle exercise of 90 s [32].
Redistribution of body water during exercise may contribute to the
increases in circulating hormones as well [3]. Previous results
demonstrated that IGF-1/IGFBP-3 levels determined at various time-
points during exercise and corrected for changes in haematocrit
display only modest or no increase in IGF-1 and IGFBP-3 [3,6,9,3335].
In the present study, increases in haematocrit were observed, but
anyhow these facts did not change the observation that the
diminished acidosis during HIT (B) slightly attenuates the response
of hGH and cortisol. In contrast, no signicant changes can be
observed for IGFBP-3 anymore. So it can be concluded that circulatory
acidosis associated with high-intensity exercise stress appears to have
limited inuence on the potential mechanism(s) that mediate
increases in circulating concentrations of IGF-1 and IGFBP-3, which
is in line with the results of Kraemer et al. as described previously [32].
However we decided to present the absolute concentrations of
growth factors and binding protein as suggested by others [3,9,34], with
no correction for haematocrit changes, as there are errors associated
with estimating plasma volume changes from haematocrit values
during exercise [36], which questions the validity of the haematocrit
correction factor [34]. Furthermore, the resulting increase in growth
factor concentration will increase the probability of interaction with
receptors in target tissues, which might be the reason why previous
studies have chosen not to correct for changes in haematocrit [10,34,35].
Berg et al. suggested that if the focus is the possible changes in tissue
actions of IGF-1 during exercise it is not meaningful to correct changes in
circulating IGF-1 for changes in haematocrit [3].
Our results demonstrate that the consequence of an increase in
circulating cortisol and GH levels is in part a decreased blood pH. How-
ever the present study does not provide a mechanistic understanding
of the role of a change in pH. Nevertheless these ndings might be
relevant for training arrangements, especially the planning of loadrest
intervals. Normally, rest periods during interval training are carried out
actively, performing low-intensity exercise to decrease lactate levels
and to normalize pH faster compared to passive rest. But the faster
elimination of lactate and the faster normalization of pH decreases
the time of exposure to these potent metabolic stimuli, possibly leading
to different stress/hormonal responses to the same training load. It
can be speculated that also long term adaptations could vary depending
on the arrangement of interval training and training intensity. Future
studies should therefore focus on the long term adaptations and effects
of different congurations/designs of interval training programs.
Conict of interest
None declared.
References
[1] J. Frystyk, Exercise and the growth hormone-insulin-like growth factor axis, Med.
Sci. Sports Exerc. 42 (2010) 5866.
[2] J. Frystyk, Free insulin-like growth factors measurements and relationships to
growth hormone secretion and glucose homeostasis, Growth Horm. IGF Res. 14
(2004) 337375.
[3] U. Berg, P. Bang, Exercise and circulating insulin-like growth factor I, Horm. Res.
62 (Suppl 1) (2004) 5058.
[4] G.R. Adams, Invited review: autocrine/paracrine IGF-I and skeletal muscle
adaptation, J. Appl. Physiol. 93 (2002) 11591167.
[5] B.C. Nindl, Insulin-like growth factor-I, physical activity, and control of cellular
anabolism, Med. Sci. Sports Exerc. 42 (2010) 3538.
385
[6] U. Berg, B. Saltin, K. Hall, Net uxes of IGF-1, IGFBP-1, insulin, GH and energy
substrates over contracting muscle during prolonged exercise in healthy men,
Endocrinol. Metab. 4 (1997) 313320.
[7] B.C. Nindl, J.R. Pierce, Insulin-like growth factor I as a biomarker of health, tness,
and training status, Med. Sci. Sports Exerc. 42 (2010) 3949.
[8] N. Moller, J.O. Jorgensen, Effects of growth hormone on glucose, lipid, and protein
metabolism in human subjects, Endocr. Rev. 30 (2009) 152177.
[9] J.L. Copeland, L. Heggie, IGF-I and IGFBP-3 during continuous and interval
exercise, Int. J. Sports Med. 29 (2008) 182187.
[10] A.J. Schwarz, J.A. Brasel, R.L. Hintz, S. Mohan, D.M. Cooper, Acute effect of brief
low- and high-intensity exercise on circulating insulin-like growth factor (IGF) I,
II, and IGF-binding protein-3 and its proteolysis in young healthy men, J. Clin.
Endocrinol. Metab. 81 (1996) 34923497.
[11] K. Forsten-Williams, T.R. Cassino, L.J. Delo, A.D. Bellis, A.S. Robinson, T.E. Ryan,
Enhanced insulin-like growth factor-I (IGF-I) cell association at reduced pH is
dependent on IGF binding protein-3 (IGFBP-3) interaction, J. Cell. Physiol. 210
(2007) 298308.
[12] J.B. WEIR, New methods for calculating metabolic rate with special reference to
protein metabolism, J. Physiol. 109 (1949) 19.
[13] L.R. McNaughton, Bicarbonate ingestion: effects of dosage on 60 s cycle
ergometry, J. Sports Sci. 10 (1992) 415423.
[14] W. Van Beaumont, Evaluation of hemoconcentration from hematocrit measure-
ments, J. Appl. Physiol. 32 (1972) 712713.
[15] S.E. Gordon, W.J. Kraemer, N.H. Vos, J.M. Lynch, H.G. Knuttgen, Effect of acidbase
balance on the growth hormone response to acute high-intensity cycle exercise,
J. Appl. Physiol. 76 (1994) 821829.
[16] E.E. Hill, E. Zack, C. Battaglini, M. Viru, A. Viru, A.C. Hackney, Exercise and
circulating cortisol levels: the intensity threshold effect, J. Endocrinol. Invest 31
(2008) 587591.
[17] M. Viru, A.C. Hackney, T. Janson, K. Karelson, A. Viru, Characterization of the
cortisol response to incremental exercise in physically active young men, Acta
Physiol. Hung. 95 (2008) 219227.
[18] M.S. Tremblay, J.L. Copeland, W. Van Helder, Inuence of exercise duration on
post-exercise steroid hormone responses in trained males, Eur. J. Appl. Physiol. 94
(2005) 505513.
[19] V.S. Rojas, H.K. Struder, B.V. Wahrmann, W. Bloch, W. Hollmann, Bicarbonate
reduces serum prolactin increase induced by exercise to exhaustion, Med. Sci.
Sports Exerc. 38 (2006) 675680.
[20] J. Sutton, L. Lazarus, Growth hormone in exercise: comparison of physiological
and pharmacological stimuli, J. Appl. Physiol. 41 (1976) 523527.
[21] S.E. Christensen, O.L. Jorgensen, N. Moller, H. Orskov, Characterization of growth
hormone release in response to external heating. Comparison to exercise induced
release, Acta Endocrinol. (Copenh) (1984) 295301.
[22] R.J. Godfrey, Z. Madgwick, G.P. Whyte, The exercise-induced growth hormone
response in athletes, Sports Med. 33 (2003) 599613.
[23] A. Viru, M. Viru, Cortisolessential adaptation hormone in exercise, Int. J. Sports
Med. 25 (2004) 461464.
[24] M.J. Gibala, J.P. Little, M. van Essen, et al., Short-term sprint interval versus
traditional endurance training: similar initial adaptations in human skeletal
muscle and exercise performance, J. Physiol. 575 (2006) 901911.
[25] H. Koistinen, R. Koistinen, L. Selenius, Q. Ylikorkala, M. Seppala, Effect of marathon
run on serum IGF-I and IGF-binding protein 1 and 3 levels, J. Appl. Physiol. 80
(1996) 760764.
[26] J.A. Alemany, B.C. Nindl, M.D. Kellogg, W.J. Tharion, A.J. Young, S.J. Montain, Effects
of dietary protein content on IGF-I, testosterone, and body composition during
8 days of severe energy decit and arduous physical activity, J. Appl. Physiol. 105
(2008) 5864.
[27] E.F. De Palo, G. Antonelli, R. Gatti, S. Chiappin, P. Spinella, E. Cappellin, Effects of
two different types of exercise on GH/IGF axis in athletes. Is the free/total IGF-I
ratio a new investigative approach? Clin. Chim. Acta 387 (2008) 7174.
[28] J.A. Kanaley, J. Frystyk, N. Moller, et al., The effect of submaximal exercise on
immuno- and bioassayable IGF-I activity in patients with GH-deciency and
healthy subjects, Growth Horm. IGF Res. 15 (2005) 283290.
[29] K. Stokes, M. Nevill, J. Frystyk, H. Lakomy, G. Hall, Human growth hormone
responses to repeated bouts of sprint exercise with different recovery periods
between bouts, J. Appl. Physiol. 99 (2005) 12541261.
[30] K.E. Forsten, R.M. Akers, J.D. San Antonio, Insulin-like growth factor (IGF) binding
protein-3 regulation of IGF-I is altered in an acidic extracellular environment,
J. Cell. Physiol. 189 (2001) 356365.
[31] C.A. Conover, D.D. De Leon, Acid-activated insulin-like growth factor-binding
protein-3 proteolysis in normal and transformed cells. Role of cathepsin D, J. Biol.
Chem. 269 (1994) 70767080.
[32] W.J. Kraemer, F.S. Harman, N.H. Vos, et al., Effects of exercise and alkalosis on
serum insulin-like growth factor I and IGF-binding protein-3, Can. J. Appl. Physiol.
25 (2000) 127138.
[33] R. Dall, K.H. Lange, M. Kjaer, et al., No evidence of insulin-like growth factor-
binding protein 3 proteolysis during a maximal exercise test in elite athletes,
J. Clin. Endocrinol. Metab. 86 (2001) 669674.
[34] J. Cappon, J.A. Brasel, S. Mohan, D.M. Cooper, Effect of brief exercise on circulating
insulin-like growth factor I, J. Appl. Physiol. 76 (1994) 24902496.
[35] A. Eliakim, Y. Oh, D.M. Cooper, Effect of single wrist exercise on broblast growth
factor-2, insulin-like growth factor, and growth hormone, Am. J. Physiol. Regul.
Integr. Comp. Physiol. 279 (2000) R548R553.
[36] J. Lundvall, P. Lindgren, F-cell shift and protein loss strongly affect validity of PV
reductions indicated by Hb/Hct and plasma proteins, J. Appl. Physiol. 84 (1998)
822829.