HG1
HG1
KRAEMER, WILLIAM J., LOUIS MARCHITELLI, SCOTT E. To date, studies have generally indicated that acute
GORDON, EVERETT HARMAN, JOSEPH E. DZIADOS, ROBERT heavy-resistance exercise stimulates an increase in pe-
MELLO, PETER FRYKMAN, DINI MCCURRY, AND STEVEN J. ripheral blood concentrations of testosterone in males
FLECK. Hormonal and growth factor responses to heavy resist- (9, 13, 19, 29). Furthermore, it has been suggested that
ance exercise protocols. J. Appl. Physiol. 69(4): 1442-1450,
training may influence resting values of testosterone (14-
1990.-To examine endogenousanabolic hormone and growth
factor responsesto various heavy resistanceexerciseprotocols 16). Limited data also indicate that human growth hor-
(HREPs), nine male subjectsperformed each of six randomly mone may increase in response to an acute bout of
assignedHREPs, which consisted of identically ordered exer- resistance exercise (25, 29, 33). VanHelder et al. (33)
cisescarefully designedto control for load [5 vs. 10 repetitions have demonstrated that human growth hormone eleva-
maximum (RM)], rest period length (1 vs. 3 min), and total tions may be dependent on specific exercise characteris-
work effects. Serum human growth hormone (hGH), testoster- tics such as the load utilized and frequency of lifting
one (T), somatomedin-C (SM-C), glucose, and whole blood exercise. To our knowledge, no data exist regarding so-
lactate (HLa) concentrations were determined preexercise, matomedin-C responses to heavy resistance exercise pro-
midexercise(i.e., after 4 of 8 exercises),and at 0, 5, 15, 30, 60, tocols. The purpose of this investigation was to deter-
90, and 120 min postexercise.All HREPs produced significant mine the impact of load, rest period length, and total
(P < 0.05) temporal increases in serum T concentrations,
although the magnitude and time point of occurrence above work on serum testosterone, human growth hormone,
resting values varied acrossHREPs. No differences were ob- and somatomedin-C response patterns during and after
served for T when integrated areas under the curve (AUCs) different heavy resistance exercise protocols.
were compared.Although not all HREPs produced increasesin
serum hGH, the highest responseswere observed consequent METHODS
to the HlO/l exerciseprotocol (high total work, 1 min rest, lo-
RM load) for both temporal and time integrated (AUC) re- Nine healthy male subjects gave informed written con-
sponses.The pattern of SM-C increasesvaried among HREPs sent to participate in this investigation. The physical
and did not consistently follow hGH changes.Whereas tem- characteristics of the subjects were the following: age,
poral changeswere observed,no integrated time (AUC) differ- 24.66 t 4.27 (SD) yr; height, 178.41 t 7.77 cm; body
encesbetween exerciseprotocols occurred. These data indicate
mass, 81.08 t 12.03 kg; maximal oxygen consumption,
that the releasepatterns (temporal or time integrated) observed
are complex functions of the type of HREPs utilized and the 54.17 t 4.63 mlokgW1~min-‘; and body fat, 15.96 t 4.18%.
physiological mechanismsinvolved with determining periph- All subjects had recreational experience with resistance
eral circulatory concentrations (e.g., clearancerates, transport, training but none were competitive lifters. None of the
receptor binding). All HREPs may not affect muscleand con- subjects had a medical history of any endocrine disorder
nective tissue growth in the samemanner becauseof possible and none were on medications or hormonal therapy.
differences in hormonal and growth factor release. Furthermore, subjects reported no history of anabolic
steroid use. Urine testing of all subjects verified these
anaerobic; insulin-like growth factor; anabolic; males;lactate reports during the study.
A minimum of 2 wk was utilized for experimental
protocol familiarization, descriptive testing, and load
IN VIVO AND IN VITRO investigations have demonstrated verifications for each experimental exercise protocol.
that several hormones (e.g., growth hormone, testoster- Percent body fat was determined using hydrostatic
one) influence growth and development (11). Yet, the weighing with a computer-interfaced load cell and stand-
exact mechanism(s) by which these growth-promoting ard body composition methodology as previously de-
actions occur remains unclear. Heavy resistance training scribed (10, 37). Maximal oxygen consumption was de-
has been shown to be a potent stimulus for muscle cell termined utilizing a continuous treadmill protocol and
hypertrophy (26, 31). This may be due, in part, to exer- plateau criteria as previously described (5).
cise-induced increases in endogenous anabolic hormones Experimental design and exercise protocols. Each of the
and growth factors. six heavy resistance exercise protocols was performed in
1442
HORMONAL RESPONSES TO RESISTANCE EXERCISE 1443
random order and by all nine subjects. Subsequent sta- differences in responses occurred as a result of changes
tistical analysis demonstrated no order effects. The ex- in total work, load, and rest period lengths. The exercises
perimental design is shown in Fig. 1. This design allowed utilized, the order used, and the number of sets for the
for a more quantitative approach to examine responses primary workouts are shown in Table 1. Thus two exer-
to heavy resistance exercise because of specific changes cise series were utilized in this study, with each series
in the exercise protocols. Two exercise series were used having a primary workout (S5/3 and HlO/l), a load
in this study, each consisting of three workouts (i.e., a control (S10/3 and H5/1), and a rest control (S5/1 and
primary workout, a rest control, and a load control). H10/3). Control workouts were matched for the total
Series 1 had significantly (P < 0.05) lower total work work of the primary workout.
(49,161 t 10,100 vs. 59,859 t 12,675 J) compared with All exercises were structured proportionally for each
series 2. subject with grip widths and positions marked and kept
As can be seen in Fig. 1, the “strength workout” was constant for each exercise. The matching of the total
the primary exercise protocol for series I, which used a work between workouts was performed by a computer
5-repetitions maximum (RM) load and a 3-min rest program, which, given a specific exercise, weight, and
period length between sets and exercises. It was desig- number of repetitions, calculated the number of repeti-
nated as S5/3, meaning S for series 1 (lower total work), tions required to produce the same total work using a
5 for the 5-RM load used, and /3 for the 3-min rest different weight. Lifting work was calculated as weight
periods. The load control workout for series 1 used a lo- times the vertical distance moved per repetition times
RM load and was designated as SlO/3. The rest control the number of repetitions. The program took into con-
workout for series 1 used a 1-min rest period length and sideration the vertical distance moved of both the iron
was designated as S5/l. The same type of terminology plates and the centers of gravity of the lifters’ body
was utilized for designating each exercise protocol in segments. These distances were obtained from measure-
series 2. The “hypertrophy workout” was the primary ments on the subjects and equipment when they were in
protocol for series 2 and was designated as HlO/l, mean- the starting and ending exercise positions. Anthropo-
ing H for series 2 (higher total work), 10 for the lo-RM metric tables were used to locate body segment centers
load used, and /l for the 1-min rest periods used between of gravity and estimate body segment weights from total
sets and exercises. Similarly, the load control workout body weight (37).
used a 5-RM load and was designated as H5/1, and the Experimental protocol. One week separated each ran-
rest control workout used 3-min rest period lengths and domized experimental protocol. Subjects refrained from
was designated as Hl0/3. Whereas both workouts de- ingestion of alcohol or caffeine for 24 h and did not
velop strength and hypertrophy, the S5/3 workout pro- perform any strenuous exercise for 48 h before the ex-
tocol is typical of weight-training protocols used primar- perimental exercise session. Resistance exercise sessions
ily for “strength” development and the HlO/l workout during the entire investigation were controlled as sub-
protocol is primarily used by bodybuilders for increases jects were only allowed to perform one other light resist-
in muscular hypertrophy (20). The variations of the ance exercise workout (i.e., 12-15 RM for 4 exercises)
primary workouts were examined to help determine if during each 7-day cycle. One RM testing every other
week demonstrated that no strength changes occurred
0 over the course of the study. In addition, all aerobic
exercise was also controlled to allow only two sessions
per week, again with no training effects observed over
5 RM, 1 min (SS/l) 5 RM, 1 min (H5/1)
(Series 1 Rest Control) (Series 2 Load Control) the course of the study. Thus all subjects were housed
(US Army billets), fed (US Army dining facility), and
Decreased Rest Increased Load
exercised under similar conditions for the entire duration
of the study. Dietary analysis (Nutri-Calc, PCD System,
Penn Yan, NY) for the 3 days before each experimental
~umEw(3iuw ll!mmmJu Mgpmm?@~B~ f!!lmmBm(l session demonstrated normal Recommended Daily Al-
(Primary Workout) (Primary Workout) lowance caloric, nutrient, vitamin, and mineral intakes.
5 RM, 3min (S5/3) 10 RM, 1 min (HlO/l)
TABLE] 1. Experimental heavy resistance
exercise protocols
Decreased Load Increased Rest
Repetition Maximum and No. of Sets
Exercise Order
Series I (S5/3) Series 2 (HlO/l)
10 RM, 3 min (S10/3) 10 RM, 3 min (Hl O/3)
(Series 1 Load Control) (Series 2 Rest Control) 1. Bench press 5 RM x 5 sets 10 RM x 3 sets
2. Double-leg extension 5 RM x 5 sets 10 RM x 3 sets
3. Military press 5 RM x 3 sets 10 RM x 3 sets
TOTAL WORK TOTAL WORK 4. Bent leg, incline sit-ups 5 RM x 3 sets 10 RM x 3 sets
5. Seated rows 5 RM x 3 sets 10 RM x 3 sets
All Series 1 Workouts All Series 2 UJorkouts 6. Latissimus dorsi pull down 5 RM X 4 sets 10 RM x 3 sets
(49,161 J) (59,859 J) 7. Arm curls 5 RM x 3 sets 10 RM x 3 sets
FIG. 1. Experimental design consisting of 2 series of 3 exercise 8. Leg press 5 RM x 5 sets 10 RM x 3 sets
protocols, each with primary workout, load control, and rest control. All exercises were performed on a Universal weight machine except
Total work for series 1 was significantly (P < 0.05) less than series 2. for exercises 4 and 7, which utilized free weights.
1444 HORMONAL RESPONSES TO RESISTANCE EXERCISE
Prior urine nitrogen determinations verified all subjects were computed after preexercise values were subtracted
to be in the normal range for positive nitrogen balance from each time point. Differences between areas under
before each test session. the curve were analyzed using an analysis of variance
Subjects reported for the experimental session, and with repeated measures and Tukey’s post hoc analyses
venous blood samples were obtained with the subjects in to determine pairwise differences. Significance in this
a slightly reclined seated position, which was used for all study was chosen at P < 0.05.
samples. All testing was conducted at the same time of
day to reduce the effects of any diurnal variations on the RESULTS
hormonal concentrations. Before a resting blood sample
was obtained, a ZO-min equilibration period was utilized. In general, this investigation demonstrated variations
Subjects knew they would not immediately start to ex- in response patterns between the different heavy resist-
ercise after the resting blood sample was obtained. The ance exercise protocols. The most prominent differences
exercise protocol started 10 min after the resting blood in response patterns were observed for serum human
sample was drawn. During pilot testing, this procedure growth hormone and whole blood lactate values, which
was shown to eliminate any significant anticipatory in- maintained sustained elevations over time. In contrast,
creases in hormonal responses previously thought to alterations in serum testosterone and somatomedin-C
affect the examination of exercise responses (19). Water response patterns were more irregular. At the onset, it is
intake was allowed ad libitum throughout the exercise also important to note that between the six heavy resist-
protocols and recovery. The venous blood samples were ance exercise protocols, no significant differences in rest-
obtained from an indwelling cannula in a superficial arm ing baseline values were observed for any of the blood
vein kept patent with isotonic saline (30 ml/h). Blood variables measured.
samples were obtained preexercise, midexercise (i.e., Figure 2 shows the responses of serum glucose (A) and
after 4 exercises), and at 0 (immediately postexercise), 5, whole blood lactate (B). No exercise protocol in series 1
15, 30, 60, 90, and 120 min after each exercise protocol. or 2 elicited any significant changes in serum glucose
All blood samples were processed and stored at -120°C concentrations. For whole blood lactate significant group
until analysis. and time effects were observed. Subsequent post hoc
Biochemical analyses. Whole blood lactate and serum analyses with preexercise values demonstrated signifi-
glucose concentrations were determined in duplicate via cant increases above rest at various time points in both
a Lactate Analyzer-640 (Wolverine Medical, Grand Rap- exercise series, which are shown in Fig. 2B.
ids, MI) and a 23-Glucose Analyzer (Yellow Springs Total work comparisons (i.e., S5/1 vs. H5/1 and SlO/
Instruments, Yellow Springs, OH). Hemoglobin was ana- 3 vs. H10/3) showed no pairwise differences in whole
lyzed in triplicate using the cyanmethemoglobin method blood lactate between S5/1 and H5/1, but H10/3 values
(Sigma Chemical, St. Louis, MO), and hematocrit was at midexercise, 0,5, and 15 min postexercise were signif-
analyzed in triplicate utilizing standard microcapillary icantly greater than S10/3 at those time points.
technique. The percent changes in plasma volume were Evaluation of the effects of rest period changes in
calculated according to equations by Dill and Costill (7). series 1 showed that decreasing the rest period length to
Serum testosterone, human growth hormone, and so- 1 min (SS/l) significantly increased the concentrations
matomedin-C concentrations were determined by radio- over S5/3 values at midexercise and 0 and 5 min post-
immunoassay. All samples were run in duplicate and exercise. Similarly, in series 2, increasing the rest period
were decoded only after analyses were completed. Tes- length from 1 to 3 min (H10/3) significantly decreased
tosterone was measured using an 1251solid-phase radio- whole blood lactate values at midexercise and at 0,5,15,
immunoassay (Diagnostic Products, Los Angeles, CA) 30, and 60 min postexercise. A significant interaction
with a detection limit of 0.38 nmol/l. Intra- and inter- effect in whole blood lactate was seen between total work
assay variances were calculated to be ~3.2 and 4.9%, and rest period changes.
respectively. Human growth hormone was measured uti- In series 1, no significant differences were observed in
lizing a 1251liquid-phase radioimmunoassay with double- whole blood lactate when the load was lightened [ (S5/3)
antibody technique (Cambridge Medical Diagnostics, 5 RM vs. 10 RM (S10/3)]. Conversely, in series 2, when
Billerica, MA) with a limit of detection of 0.24 pg/l. the load was increased from 10 RM to 5 RM (HlO/l vs.
Intra- and interassay variances were calculated to be H5/1) significant decreases were observed in whole blood
~4.4 and 4.9%, respectively. Somatomedin-C (insulin- lactate at midexercise and at 0, 5, 15, 30, and 60 min
like growth factor 1) was measured using a 12? double- postexercise. Again, a significant interaction with total
antibody disequilibrium radioimmunoassay with a pre- work was observed in load changes.
liminary octadecasilyl-silica (ODS) extraction procedure Figure 3 shows the responses of serum testosterone to
(IncStar, Stillwater, MN), with the limit of detection the various heavy resistance exercise protocols. Signifi-
being ~2.0 nmol/l. Intra- and interassay variances were cant group and time effects were observed. No differences
~4.6 and 4.8%, respectively. were observed for area under the curve comparisons
Statistical anaZyses. Statistical evaluation of these data (inset). The temporal increases in testosterone above
was accomplished utilizing a multivariate analysis of resting values are shown for each heavy resistance exer-
variance with repeated measures. Subsequent post hoc cise series.
pairwise differences were determined using Tukey’s tests. When total work comparisons (i.e., S5/1 vs. H5/1 and
The resulting integrated areas under the curve for tes- S10/3 vs. H10/3) were examined, post hoc analyses dem-
tosterone, human growth hormone, and somatomedin-C onstrated the only difference was that S5/1 was greater
HORMONAL RESPONSES TO RESISTANCE EXERCISE 1445
SERIES 1
ES 2
. . S5/3 (primary workout)
- HlO/l (primary workout)
A O-----• s10/3 (loed control)
.--w-v.
H5/1 (load control)
e --4 s5/1 (rest control)
e -- 4
Hf O/3 (rest control)
z3 01 PREI MIDI I I
0 5 15
1 I
30 60
I I
90 120
I 1 I I
PRE MID 0 5
I
15
I 1 I I
30 60 90
I
120
EXERCISE - POST-EXERCISE - EXERC’SE - POST-EXERCISE -
(mm mm
FIG. 2. Serum glucose (A) and whole blood lactate (B) concentrations (means -t- SE) for various time points and
resistance exercise protocols. * P < 0.05 from corresponding resting values.
than H5/1 at 90 min. No differences as a result of total Subsequent post hoc comparisons to determine the
work were observed when area under the curve analyses effects of total work (i.e., S5/1 vs. H5/1 and SlO/3 vs.
were examined. Hl0/3) on serum human growth hormone showed that
Changing the rest period length in series 1 (S5/1 vs. there were significant pairwise differences at 0 and 5 min
S5/3) demonstrated only one significant pairwise differ- postexercise for the 5-RM load (SS/l > H5/1). No dif-
ence at 30 min postexercise where S5/3 was greater than ferences were observed between the SlO/3 and Hl0/3
S5/l. In series 2, increasing the rest period length to 3 protocols. Furthermore, no differences for total work
min (HlO/l vs. Hl0/3) resulted in significantly lower comparisons were observed when area under the curve
values at midexercise and at 0, 5, and 15 min postexer- responses w.ere examined.
cise. In addition, significantly lower serum testosterone Changing rest period length in series 1 resulted in
concentrations for HlO/l vs. Hl0/3 were observed at 120 serum human growth hormone responses for S5/1 that
min postexercise. No significant interactions with total were significantly greater than S5/3 at 5 min postexer-
work were observed for changes in serum testosterone cise. In series 2, increasing the rest period length to 3
concentrations. Again, no significant differences were min (Hl0/3) significantly reduced the serum concentra-
observed for area under the curve analyses. tions of human growth hormone at midexercise and at 0,
Alterations in load for series 1 (S5/3 vs. SlO/3) dem- 5, 15, and 30 min postexercise. Consistent with the
onstrated that there were significant pairwise differences temporal findings, the integrated area under the curve
at 30 and 60 min postexercise. No significant differences, analyses supported the finding that increasing the rest
were observed for load changes (HlO/l vs. H5/1) in series period length to 3 min significantly reduced the serum
2. No significant interactions were observed with total human growth hormone levels (HlO/l > Hl0/3). No
work. Area under the curve analyses demonstrated no significant interactions with total work were observed
differences between heavy resistance exercise protocols. for human growth hormone responses.
Figure 4 shows the responses of human growth hor- No significant differences in serum human growth
mone to the various heavy resistance exercise protocols. hormone responses were observed in series 1 when the
Significant group and time effects were observed. The load was changed (S5/3 vs. SlO/3). In series 2 when the
various significant increases above resting concentra- load was changed from 10 to 5 RM (HlO/l vs. H5/1),
tions for each exercise protocol are shown. The inte- significantly lower serum human growth hormone con-
grated area under the curve results demonstrated that centrations were observed at midexercise and at 0, 5, 15,
the HlO/l protocol response was significantly greater and 30 min postexercise. The integrated area under the
than that of any of the other exercise protocols examined curve analyses supported the trend observed in the tem-
(inset). poral responses as no differences were observed in series
1446 HORMONAL RESPONSES TO RESISTANCE EXERCISE
-200 ’
fS/S 510/J M/l HlO/l HO/I HIO/J
SERIES 2
s 1
ES 2
. S5/3 (primary workout)
0
.-. Hl O/l (primary work out)
.--a--* S 1 O/3 (load control) @--e-w@ H5/1 (load control)
b --4 S5/1 (rest control) :: + -- -4 Hl O/3 (rest control)
*
31.t *
* * *
* TTf * I
- $A L T i I
15.0
.
, 1 I t I I I 1 1 1
PRE MID 0 5 15 30 60 96 120 PRE MID 0 5 15 30 60 90 120
EXERCISE I - POST-EXERCISE -I EXERCISE t - POST~EXERCISE -
(min) (min)
FIG. 3. Serum testosterone concentrations (means k SE) for various time points and resistance exercise protocols.
* P c 0.05 from corresponding resting values. Inset, area under the curve (AUC) testosterone responses (means t
SE).
1. In series 2, a significant reduction in human growth significantly higher serum somatomedin-C concentra-
hormone responses resulted when the load was changed tions than HlO/l at 5 min postexercise (H10/3 > HlO/
(i.e., H5/1 c HlO/l). No significant interactions with 1). No significant interactions were observed with total
total work were observed with load changes. work and no differences were observed for integrated
Figure 5 shows the responses of serum somatomedin- area under the curve comparisons.
C (insulinlike growth factor 1) to the various heavy When the load was changed in series 1, only one
resistance exercise protocols. Significant group and time pairwise difference was observed. S5/3 somatomedin-C
effects were observed. The various significant increases values were significantly lower than S10/3 values at 90
above resting concentrations are shown. No significant min postexercise. No differences were observed for load
differences were observed in somatomedin-C when the changes in series 2. No interactions with total work were
integrated areas under the curve for the various heavy observed and no differences were demonstrated when
resistance exercise protocols were compared (inset). integrated areas under the curve were examined.
Total work produced two pairwise differences in so- The hormonal responses were corrected for percent
matomedin-C responses, one at 30 min postexercise changes in plasma volume to reduce the influence of
where the S5/1 protocol was significantly greater than plasma volume shifts on serum concentrations during
the H5/1 protocol and one at 90 min postexercise where the different exercise protocols. Furthermore, normali-
the SlO/3 protocol was significantly greater than the zation of these data did not yield different results. Peak
H10/3 protocol. No integrated area under the curve values were not correlated with initial concentrations of
differences were observed. the hormones. Changes in plasma volume shifts during
When the rest period length was changed in series 1, recovery were negligible. The greatest percent changes
significant differences were observed at 30 min postex- in plasma volume were observed pre- to postexercise and
ercise (S5/3 < S5/1). In series 2, the H10/3 protocol had were as follows (means t SD): series 1: S5/3, -7.68 t
HORMONAL RESPONSES TO RESISTANCE EXERCISE 1447
,oooAUC
GROWTH
HORNONE
(j&l)
400
SERIES 2
200 - HlO/l (primary workout)
0 . ..----* M/l (load control)
ws SlO/S W HlO/l W HO/S
SERIES 1 SERIES 2 G -- 4 Hl O/3 (rest control)
*-
. IES 1
-f 32.0
b 0 . S5/3 (primary workout)
. *
3 28.0 -
.c----. S 1013 (load control) *
s
$j 24.0 . 0- -0-0 S5/1 (rest control)
E .
I= 20.0
E
3 16.0 .
0
Qt
z" 12.0
4:
5 8.0
=t
If 4.0
FIG. 4. Serum human growth hormone concentrations (means t SE) for various time points and resistance exercise
protocols. * .P c 0.05 from corresponding resting values. Inset, integrated AUC responses (means 2 SE). For integrated
AUC responses, * P < 0.05 from all other exercise protocols.
4.28%; S10/3, -3.58 t 4.25%; and S5/1, -4.86 t 2.54%; sponses. Thus, although many acute temporal changes
series 2: HlO/l, -8.22 t 5.52%; H5/1, -3.37 -1- 2.74%; were observed, the magnitude, direction of change, and
and H10/3, -2.77 t 2.31%. variability of these changes at the various time points
did not result in many significant time-integrated area
DISCUSSION under the curve response pattern differences. The type
of heavy resistance exercise protocol utilized in training
Heavy resistance exercise has been shown to be a may have important ramifications on subsequent adap-
potent stimulus for increases in muscle cell size (26, 31). tations (20). Still, whether any subsequent adaptations
Still, the link between the concomitant in vivo changes are related to the acute temporal increases or the more
of endogenous anabolic hormones and tissue growth has prolonged time integrated increases during a recovery
not been specifically determined, and increased concen- time period remains unknown.
trations of circulating anabolic hormones may not reflect Testosterone is a potent anabolic hormone affecting
anabolism at the tissue’ level. This investigation was a (i.e., direct and indirect effects) muscle tissue growth (11,
first step in the future examination of such relationships 21, 23). Testosterone has a direct effect on skeletal
and improves our present understanding of the circula- muscle, and these actions are not mediated by a second-
tory responses of such anabolic hormones and growth ary hormone (11). Each heavy resistance exercise pro-
factors to various heavy resistance exercise protocols. tocol in this study increased serum testosterone concen-
These data demonstrate that heavy resistance exercise trations above resting concentrations. However, all pro-
protocols elicit various temporal increases in peripheral tocols did not elicit the same magnitude or duration of
concentrations of anabolic hormones and growth factors. serum testosterone increases, even when the identical
In general, the most striking differences between exercise total work was performed (Fig. 3). Increasing the rest in
protocols were demonstrated by changes in human series 2 and decreasing the load in series 1 reduced serum
growth hormone and whole blood lactate. This contrasts testosterone concentrations independently of total work
sharply with the less obvious differences between exer- differences. This suggests that such serum testosterone
cise protocols for testosterone and somatomedin-C re- alterations consequent to changing single factor exercise
1448 HORMONAL RESPONSES TO RESISTANCE EXERCISE
ooo AUC SOMTOUEDIN-C (nmol/L)
r
ES 1 SERIES 2 -
-; 26.0
%
g 24.0
0'
t 22.0
B
s 20.0
co
9 18.0
5:
s 16.0
z
g 14.0
12.0
10.0 I P
.4 I I I I I 1 I I 4- I I I I -I) -
1 I I
PRE MID 0 5 15 30 60 90 120 PRE MID 0 5 1s 30 60 90 120
EXERCISE - POST-EXERCISE - EXERCISE I- POST-EXERCISE -
(mW (mm
FIG. 5. Serum somatomedin-C concentrations (means t SE) for various time points and resistance exercise
protocols. * P < 0.05 from corresponding resting values. Inset, integrated AUC responses (means A SE).
variables may influence differential metabolic and cel- result of decreased muscle utilization of testosterone.
lular changes. Still, because of the lack of any large Increases in testosterone during exercise and into re-
sustained changes, such differential effects in anabolism covery have been commonly observed in other studies
and cell growth remain speculative and need to be di- after high-intensity aerobic and anaerobic exercise (6,
rectly documented. Although increases in testosterone 22,25,35). Typically, changes in the plasma volume and
consequent to an acute exercise session utilizing several a reduction in clearance rates secondary to reductions in
heavy resistance exercises have been previously observed hepatic blood flow have been used to explain such hor-
(9, 35), these data now demonstrate that the manipula- monal responses to acute exercise (4, 28, 35). These
tion of exercise variables can impact the response pat- mechanisms could theoretically contribute to the exer-
terns observed. cise responses of each of the hormones examined in this
In this investigation a longer recovery period was investigation. The magnitude of the plasma volume shifts
examined, and, although possibly as a result of random in this investigation was small during the exercise ses-
variations, these data demonstrated that increases in sions and almost negligible during the recovery period.
testosterone may occur in a rebound fashion later into It has been previously reported that heavy resistance
recovery (i.e., 90 or 120 min) after certain heavy resist- exercise produces elevations in human growth hormone
ante exercise protocols. Still, the extent to which hor- concentrations (25, 29, 33). In this study, only certain
monal pulsations may have influenced these data is not heavy resistance exercise protocols produced significant
clear. It might be hypothesized that, regardless of the elevations in human growth hormone, which confirms
mechanism(s) of testosterone increase, the skeletal mus- the findings of VanHelder et al. (33), who showed that
cle will be exposed to an elevated peripheral testosterone exercise variables in resistance training may have an
concentration. This could increase the likelihood of pos- important role in determining the serum responses of
sible interactions with potential muscle cell receptors. human growth hormone. It is apparent from these data
Alternatively, increased levels could theoretically be a that within a series that keeps total work constant,
HORMONAL RESPONSES TO RESISTANCE EXERCISE 1449
changing one program variable alters the serum human We thank Drs. Ronald Portanova, Carl Maresh, and Karl Fried1 for
growth hormone response patterns. A variety of possible helpful comments on the manuscript. We also are grateful to Charles
Cruthirds, Andrew Damokosh, and Alan Vela for help in data collection
mechanisms may have influenced these changes such as and analyses. We thank Joann Ruble and Andrew C. Fry for help in
metabolic demands, hyperventilation, breath holding, the preparation of the manuscript. We also thank a dedicated group of
and hypoxia (8,32,34). These data extend the results of test subjects without whom this project would not have been possible.
VanHelder et al. and support suggestions that exercise We especially thank Dr. James Vogel for intense support and encour-
agement.
that produces greater demands on anaerobic glycolysis Human subjects participated in these studies after giving their free
stimulates serum human growth hormone elevations. and informed voluntary consent. Investigators adhered to AR 70-25
Still, it needs to be pointed out that no consistent sys- and US Army Medical Research and Development Command Regula-
tematic relationships to blood lactate were observed in tion 70-25 on Use of Volunteers in Research.
this study, suggesting that a combination of factors re- The views, opinions, and/or findings contained in this report are
those of the author(s) and should not be construed as an official
lated to anaerobic metabolism is involved. It was dem- Department of the Army position, policy, or decision, unless so desig-
onstrated in this study that the blood lactate responses nated by other official documentation.
are significantly influenced by the rest period length and Address for reprint requests: W. J. Kraemer, Center for Sports
duration of exercise. The combination of short rest pe- Medicine, Greenberg Sports Complex, The Pennsylvania State Uni-
versity, University Park, PA 16802.
riods and long duration (i.e., time to perform a lo-RM
set was longer than that to perform a 5-RM set) resulted Received 13 March 1989; accepted in final form 7 June 1990.
in greater blood lactate responses to heavy resistance
exercise. When exercise duration was reduced or rest REFERENCES
period length increased, the resulting blood lactate levels 1. ALLEN, R. E., R. A. MERKEL, AND R. B. YOUNG. Cellular aspects
were lowered. of muscle growth: myogenic cell proliferation. J. Anim. Sci. 49:
It has been demonstrated that growth effects attrib- 115-127,1979.
uted to human growth hormone are likely mediated 2. BERELOWITZ, M., M. SZABO, L. A. FROOHMAN, S. FIRESTONE,
AND L. CHU. Somatomedin-C mediates growth hormone negative
through the effects of secondary hormones (i.e., soma- feedback by effects on both the hypothalamus and the pituitary.
tomedins) (2, 30). With the recent biochemical charac- Science Wash. DC 212: 1279-1281,198l.
terization of the growth hormone receptor, the direct 3. BLUM, W. F. Insulin-like growth factor I (IGF-I)-binding protein
effects of growth hormone-receptor interactions conse- complex is better mitogen than free IGF-I. Endocrinology 125: 766-
quent to exercise remain to be demonstrated (1, 24, 30). 772,1989.
4. CADOUX-HUDSON, T. A., J. D. FEW, AND F. J. IMMS. The effects
Studies by Florini et al. (11, 12) have suggested that of exercise on the production and clearance of testosterone in well
longer integrated time periods (i.e., 24 h) of human trained young men. Eur. J. Appl. Physiol. Occup. Physiol. 54: 321-
growth hormone levels are better correlated to somato- 325,1985.
medin-C levels. Somatomedins have been hypothesized 5. COSTILL, D. L., AND E. L. FOX. Energetics of marathon running.
Med. Sci. Sports 1: 81-86, 1969.
to be involved with a variety of physiological roles, rang- 6. CUMMING, D. C., L. A. BRUNSTING, G. STRICH, A. L. RIES, AND
ing from muscle, bone, and connective tissue growth to R. W. REBAR. Reproductive hormone increases in response to
changes in muscle strength as a result of the aging acute exercise, in men. Med. Sci. Sports Exercise 18: 369-373,1986.
process (12, 19, 27). Furthermore, in vitro studies have 7. DILL, D. B., AND D. L. COSTILL. Calculation of percentage change
shown that somatomedin-C is one of the most potent in volume of blood, plasma, and red cells in dehydration. J. Appl.
Physiol. 37: 247-248, 1974.
anabolic ligands mediating muscle cell growth (11, 12). 8. DJAROVA, R., A. ILKOV, A. VARBANOVA, A. KIKIFOROVA, AND G.
To our knowledge, no other study has evaluated the MATEEV. Human growth hormone, cortisol and acid-base balance
acute responses of serum somatomedin-C to heavy re- changes after hyperventilation and breath-holding. Int. J. Sports
sistance exercise. Only random acute temporal increases Med. 7: 311-315,1986.
9. FAHEY, T. D., R. ROLPH, P. MOUNGMEE, J. NAGEL, AND S.
were observed for somatomedin-C, and no changes were MORTARA. Serum testosterone, body composition and strength of
observed in the area under the curve response patterns. young adults. Med. Sci. Sports 8: 31-34, 1976.
Such results probably reflect a more complex set of 10. FITZGERALD, P. I., J. A. VOGEL, J. MILETTI, AND J. M. FOSTER.
regulatory mechanisms related to cellular localization, An Improved Portable Hydrostatic Weighting System for Body
stimulated release, transporter protein binding and re- Composition. Natick, MA: US Army Res. Inst. Environ. Med.,
1987. (Tech. Rep. T4-88)
lease, and receptor interactions (17, 30). Blum (3) has 11. FLORINI, J. R. Hormonal control of muscle cell growth. J. Anim.
recently demonstrated that binding proteins may act as Sci. 61: 21-37, 1985.
a reservoir, constantly releasing small amounts of so- 12. FLORINI, J. R., P. N. PRINZ, M. V. VITIELLO, AND R. L. HINTZ.
matomedin-C. This has been hypothesized to create a Somatomedin-C levels in healthy young and old men: relationship
to peak and 24 hour integrated levels of growth hormone. J.
steady-state situation for receptor occupancy by a pul- Gerontol. 40: 2-7, 1985.
satile release of somatomedin-C from binding proteins 13. GUEZENNEC, Y., L. LEGER, F. LHOSTE, M. AYMONOD, AND P. C.
(3). This appears to be more effective than exposing the PESQUIES. Hormone and metabolite response to weight-lifting
receptor to large temporary concentrations of somato- training sessions. Int. J. Sports Med. 7: 100-105, 1986.
medin-C. Such mechanisms might explain the apparent 14. HAKKINEN, K., A. PAKARINEN, M. ALEN, H. KAUHANEN, AND P.
V. KOMI. Relationships between training volume, physical per-
lack of changes in response to the various heavy resist- formance capacity and serum hormone concentrations during pro-
ance exercise protocols. The large temporal increases longed training in elite weight lifters. Int. J. Sports Med. 8: 61-65,
observed at various time points may represent an over- 1987.
load on this reservoir-type mechanism. Thus the phys- 15. HAKKINEN, K., A. PAKARINEN, M. ALEN, H. KAUHANEN, AND P.
V. KOMI. Neuromuscular and hormonal adaptations in athletes to
iological significance and source of such large temporal strength training in two years. J. Appl. Physiol. 65: 2406-2412,
increases in serum somatomedin-C observed in this study 1988.
after resistance exercise remain to be elucidated. HAKKINEN. K.. A. PAKARINEN. M. ALEN. AND P. V. KOMI. Serum
1450 HORMONAL RESPONSES TO RESISTANCE EXERCISE
hormones during prolonged training of neuromuscular perform- Central Nervous System, edited by M. K. Raizada, M. I. Phillips,
ance. Eur. J. Appl. Physiol. Occup. Physiol. 53: 287-293, 1985. and D. LeRoith. New York: Plenum, 1987, p. 277-295.
17. HAN, V. K. M., A. J. D’ERCOLE, AND P. K. LUND. Cellular 28. ROWELL, L. B., J. R. BLACKMAN, AND R. A. BRUCE. Indocyanine
localization of somatomedin (insulin-like growth factor) messenger green clearance and estimated hepatic blood flow during mild to
RNA in the human fetus. Science Wash. DC 236: 193-197,1987. maximal exercise in upright man. J. Clin. Invest. 43: 1677-1690,
18. KLIMES, I., M. VIGAS, J. JURCORICOVA, AND S. NEMETH. Lack of 1964.
effect of acid-base alterations on growth hormone secretion in man. 29. SKIERSKA, E., J. USTUPSKA, B. BICZOWA, AND J. LUKASZEWSKA.
EndocrinoZ. Exp. 11: 155-162,1977. Effect of physical exercise on plasma cortisol, testosterone and
19. KRAEMER, W. J. Endocrine responses to resistance exercise. Med. growth hormone levels in weight lifters. Endokrynol. Pal. 27: 159-
Sci. Sports Exercise 20, Suppl.: Sl52-S157, 1988. 165,1976.
20. KRAEMER, W. J., B. J. NOBEL, M. J. CLARK, AND B. W. CULVER. 30. SONNTAG, W. E., L. J. FORMAN, N. MIKI, AND J. MEITES. Growth
Physiologic responses to heavy-resistance exercise with very short hormone secretion and neuroendocrine regulation. In: Handbook
rest periods. Int. J. Sports Med. 8: 247-252, 1987. for Endocrinology, edited by G. H. Bass and H. M. Kaplan. Boca
21. KUOPPASALMI, K., AND H. ALDERCREUTZ. Interaction between Raton, FL: CRC, 1982, p. 35-59.
catabolic and anabolic steroid hormones in muscular exercise. In: 31. STARON, R. S., R. S. HIKIDA, F. C. HAGERMAN, G. A. DUDLEY,
Exercise Endocrinology, edited by K. Fotherby and S. B. Pal. Berlin: AND R. F. MURRAY. Human skeletal muscle fiber type adaptability
de Gruyter, 1985, p. 65-98. to various workloads. J. Histochem. Cytochem. 32: 146-152,1984.
22. KUOPPASALMI, K., H. NAVERI, S. REHUNEN, M. HARKONEN, AND 32. SUTTON, J. R. Effect of acute hypoxia on the hormonal response
H. ALDERCREUTZ. Effect of strenuous anaerobic running exercise to exercise. J. Appl. Physiol. 42: 587-592, 1977.
on plasma growth hormone, cortisol, luteinizing hormone, testos- 33. VAN HELDER, W. P., M. W. RADOMSKI, AND R. C. GOODE. Growth
terone, androstenedione, estrone and estradiol. J. Steroid Biochem. hormone responses during intermittent weight lifting exercise in
7: 823-829,1976. men. Eur. J. Appl. Physiol. Occup. Physiol. 53: 31-34, 1984.
23. LAMB, D. R. Androgens and exercise. Med. Sci. Sports 7: l-5,1975. 34. VANHELDER, W. P., K. CASEY, R. C. GOODE, AND W. M. RADOM-
24. LEUNG, D. W., S. A. SPENCER, G. CACHIANES, R. G. HAMMONDS, SKI. Growth hormone regulation in two types of aerobic exercise
C. COLLINS, W. J. HENZEL, R. BARNARD, M. J. WATERS, AND W. of equal oxygen uptake. Eur. J. Appl. Physiol. Occup. Physiol. 55:
I. WOOD. Growth hormone receptor and serum binding protein: 236-239,1986.
purification, cloning and expression. Nature Land. 330: 537-543, 35. WEISS, L. W., K. J. CURETON, AND F. N. THOMPSON. Comparison
1987. of serum testosterone and androstenedione responses to weight
25. LUKASZEWSKA, J., B. BICZOWA, D. BOLIEWICZ, M. WILK, AND B. lifting in men and women. Eur. J. Appl. Physiol. Occup. Physiol.
OBUCHOWICA-FEDELUS. Effect of physical exercise on plasma cor- 50:413-419,1983.
tisol and growth hormone levels in young weight lifters. Endokry- 36. WILKERSON, J. E., S. M. HORVATH, AND B. GUTIN. Plasma tes-
nol. Pal. 27: 149-158, 1976. tosterone during treadmill exercise. J. Appl. Physiol. 49: 249-253,
26. MACDOUGALL, J. D., G. R. WARD, D. G. SALE, AND J. R. SUTTON. 1980.
Biochemical adaptation of human skeletal muscle to heavy resist- 37. WILMORE, J. H., P. A. VODAK, R. B. PARR, R. N. GIRONDOLF,
ance training and immobilization. J. Appl. Physiol. 43: 700-703, AND J. E. BEHIRG. Further simplification of a method for deter-
1977. mination of residual lung volume. Med. Sci. Sports Exercise 12:
27. ROSENFELD, R. G., AND A. R. HOFFMAN. Insulin-like growth 216-218,198O.
factors and their receptors in the pituitary and hypothalamus. In: 38. WINTER, D. A. Biomechanics of Human Movement. New York:
Insulin, Insulin-Like Growth Factors, and Their Receptors in the Wiley, 1979, p. 151-152.