1
Adrenal Medulla And Exercise Training
Michael Kjær
Department Of Internal Medicine Tta, The State University Hospital And Department
Of Medical Physiology, The Panum Institute, University Of Copenhagen, Denmark
3 Blegdamsvej, Dk-2200 Copenhagen N, Denmark. Phone: +45 35327414. Fax: +45
35327420.
Epinephrine or adrenaline is almost exclusively secreted from the adrenal
medulla during exercise. In exercising adrenodemedullated rats plasma epinephrine
was undetectable (Sonne et al. 1985, Scheurink et al. 1989) and in
adrenalectomized but mineralo- and glucocorticoid replaced human subjects the
plasma levels of epinephrine during exercise were less than 10% of what was found
in healthy control subjects at the same relative work intensity (% 02max)(Barwich et
al. 1981, Hoelzer et al. 1986).
The adrenal medulla is stimulated via sympathetic nerves (Elliott 1913)
running through the celiac ganglion, and during exercise in humans it can be
demonstrated that injection of local anesthesia around the celiac ganglion can
reduce the exercise induced rise in arterial plasma epinephrine concentrations by up
to 90% (Kjær et al. 1993). Furthermore, in tetraplegic subjects who due to spinal cord
injury have impaired sympathetic nerve activity, no rise in plasma epinephrine is
found during electrically induced bicycling of a high intensity (Kjær et al. submitted).
The stimulation of the adrenal medulla during exercise mainly results in a release of
epinephrine from chromaffin cells, that results in well described peripheral metabolic
and circulatory effects. However also norepinephrine, DOPA and various
neuropeptide such as, substance P, enkephalins, neuropeptide Y, vasoactive
intestinal polypeptide, neurotensin and calcitonin gene-related peptide are released
from the adrenal medulla during physical activity. For the majority of these
substances however, a precise physiological role during exercise has not been
2
described, and the present review will therefore focus on epinephrine in relation to
physical training.
The Effect Of Training On Epinephrine Secretion During Exercise
Arterial plasma concentrations of epinephrine rises markedly during exercise
especially at high exercise intensities, the rise being less pronounced in endurance
trained athletes (high maximal oxygen uptake) as compared to sedentary subjects at
the same absolute work load (Fig 1)(Kjær et al 1988a, Deuster et al. 1989). This
effect of training on the plasma epinephrine response occurs already over the first 35 weeks of training following the increase in maximal oxygen uptake (Winder et al.
1978), and the exercise-induced rise in plasma epinephrine is closely correlated to
the relative work load (% of maximal oxygen uptake) in the individual.
To evaluate whether changes in plasma epinephrine actually reflects changes
in adrenal medullary secretion rates, turnover studies are important. When
epinephrine clearance (3H-epinephrine infusion) is determined in both endurance
athletes and in sedentary controls, a significant inverse relationship is found between
clearance and exercise intensity expressed relative to individual maximal capacity
(% 02max)(Fig 2)(Kjær et al. 1985). Although epinephrine clearance increases 15%
at mild exercise and decreases 20% at heavy exercise (Fig 2), the important point is
that these small changes cannot explain the five to tenfold increase in plasma
epinephrine levels often seen during exercise (Kjær et al. 1985). Therefore changes
in plasma epinephrine levels during exercise in man do reflect changes in adrenal
medullary secretion of epinephrine. Furthermore, differences in epinephrine
clearance cannot explain differences in plasma epinephrine between subjects with
different training status. However, since clearance varies with the relative work load
and therefore at a given work load might increase in trained (low % 02max) but
decrease in untrained subjects (high % 02max) one may overestimate differences in
epinephrine secretion between trained and untrained by only relying on plasma
3
concentrations (Kjær et al. 1985).
The mechanism behind the lower epinephrine response in trained compared
to untrained subjects when compared at identical absolute work loads (at a given
02) is probably a reduction in stimuli to increase the autonomic neuroendocrine
activity during exercise. It has been shown that not only peripheral feed back (Sato et
al. 1986), but also activity in motor centers in the brain can directly elicit autonomic
neuroendocrine activity during exercise (Kjær et al. 1987). By the use of
tubocurarine, skeletal muscles were weakened and the voluntary effort (and thereby
the motor center activity) to perform a certain work output was increased, and this
resulted in a larger epinephrine response compared to control experiments (Fig 3).
When work load was reduced during curarization to achieve identical perceived
exertion as compared to control (Fig 3, right part) identical epinephrine responses
were found during exercise. Therefore a training status dependent change in
perceived exertion in a given individual alters the stimulation of adrenomedullary
secretory activity during exercise.
Training-Induced Alterations In Secretory Capacity Of The Adrenal Medulla.
The diminished epinephrine response to a given workload in athletes
compared to untrained subjects may reflect that the signals influencing autonomic
neuroendocrine activity are modified with physical training. However, this does not
exclude adaptations in the tissue of the adrenal medulla leading to a change in the
secretory capacity of the endocrine gland.
At the end of prolonged exhausting exercise, the epinephrine concentration in
arterial plasma as well as estimated epinephrine secretion were higher in athletes
than in sedentary subjects (Kjær et al., 1985). This difference was probably not due
to differences in stimuli in the two groups causing a higher nervous drive on the
adrenal medulla in the trained subjects, because at exhaustion the relative work load
4
(% 02max), heart rate, plasma glucose concentration and hematocrit were identical
in trained and untrained subjects (Kjær et al., 1985). In another study, an exercise
protocol which aimed at identical stimulation of the adrenal medulla in trained and
untrained subjects was used (Kjær et al., 1986). Catecholamine concentrations in
arterialized blood were measured during graded exercise including work intensities
above maximal oxygen uptake. Duration and relative intensity (% 02max) of the
different stages were identical for athletes and untrained subjects and so were heart
rates and norepinephrine concentrations. Nevertheless, epinephrine responses were
markedly higher in the athletes than in the sedentary subjects.
This pattern of epinephrine responses in trained vs untrained individuals was
also found in studies where the adrenal medulla was specifically stimulated with
various non-exercise stimuli (Yamaguchi 1992, Cheung 1990). In response to similar
degrees of insulin-induced hypoglycemia (Kjær et a., 1984) as well as after
stimulation with hypoxia (Kjær et al., 1988a, 1988b), hypercapnia (Kjær et al.,
1988b), glucagon (Kjær et al., 1988b) or caffeine (LeBlanc et al. 1985) athletes had
more pronounced increases in plasma epinephrine concentrations than untrained
individuals. Although a full dose-response relationship between the various stimuli
and epinephrine secretion was not obtained in these experiments, e.g. the exercise
stimulus as well as the glucagon dose was close to maximal, and thereby reflected
maximal epinephrine secretion capacity for those stimuli. Therefore, these findings
support the hypothesis that long-term endurance trained athletes have an increased
adrenal medullary secretory capacity.
One study has however not been able to find a larger epinephrine response in
trained subjects during hypoglycemia (Tremblay et al. 1990). This could be due to
shorter training anamnesis in that study and due to the fact that subjects were
studied early after a training bout, in a period were sensitivity to both epinephrine and
insulin was markedly enlarged.
This adaptation of the secretory capacity of the adrenal medulla probably
requires a very long period of intense physical training. In short term training studies
5
in humans, it has not been possible to demonstrate any rise in adrenal medullary
secretory capacity (Svendenhag 1985), and in highly trained athletes short-term (5
weeks) periods with inactivity due to an injury did not result in any significant
alteration in the epinephrine response to hypoglycemia compared to the trained state
(Kjær et al. 1992).
The development of a "sports adrenal medulla" might be caused by the
repeated stimulation of epinephrine secretion occurring with training. This higher
secretion capacity may also be reflected by significant higher epinephrine levels in
endurance athletes(Kjær et al. 1984,1985, 1986) although some studies have only
found tendencies towards this (Lehmann et al. 1984, Svedenhag 1985). In one study
performing 12-16 weeks intense bicycle training Devalon et al.(1989) actually found
increased basal values of epinephrine as well as for DOPA, whereas basal levels of
norepinephrine was unchanged. Finally, in a recent study the 24-hour release of
epinephrine, defined as the total elevation over baseline values (during everyday
activity including training sessions), was also found to be greater in well trained
endurance athletes compared with sedentary controls (Dela et al. 1992).
Adrenal Medullary Volume And Epinephrine Content After Physical Training
The fact that short term alterations in level of physical activity does not bring
foreword any change in epinephrine secretion capacity in humans, could speak in
favor of a selection phenomenon (subjects with large epinephrine secretion capacity
being more inclined to perform well in endurance sports) rather than a training
induced adaptation. However, in 10-weeks swim training rats (up to 6 h daily) it has
been demonstrated that not only the adrenal gland (Song et al. 1973, Östman et al.,
1971) but also the adrenal medulla volume as well as the adrenal content of
epinephrine was larger/higher in endurance trained young animals compared to
control rats who were either weight matched, sham trained or cold stressed (Fig
5)(Stallknecht et al. 1990). The training induced enlargement of the adrenal medulla
6
was mainly a hypertrophy, and the increased epinephrine content was due to the
enlargement of the gland rather than due to increased medullary epinephrine
concentration (Stallknecht 1990, Schmidt et al. 1992). Therefore it is likely that in
humans adrenal medulla adapts as a result of training, but that several years of
training is required to bring foreword this adaptation. At present, it has not been
possible to find methods (MRI etc.) that are accurate and precise enough to pick up
possible anatomical differences in adrenal gland size between trained and untrained
subjects (Stallknecht and Kjær, unpublished observations).
Interestingly, the adaptations in rats were more pronounced in males
compared to females, indicating gender differences (Stallknecht et al. 1990). Studies
evaluating gender differences in adreno-medullary adaptation to training in humans
have not been carried out. The hypertrophy of the adrenal medulla in response to
training can also occur in old rats training on a treadmill for 10 weeks (Schmidt et al.
1992) indicating that endocrine tissue, as is the case for skeletal muscle, can adapt
to physical training also at an older age (Hagberg et al. 1988). It has however to be
noted that the adaptation of the adrenal medulla to training was markedly lower in
the old compared to the young animals (Schmidt et al. 1992). In the young animals
the training may have had a more pronounced influence on adreno-medullary
function. As an example, Stallknecht et al (1990) found that the epinephrine
response to hypoglycemia in the swim-training rats (with adrenal medulla
hypertrophy) was markedly diminished. Probably the fact that the rats were
hypoglycemic not only at the end of each exercise bout, but also between training
sessions, caused adaptations within the central nervous system that are specifically
related to a young age, and that are not found in humans with free access to food
and less pronounced hypoglycemia.
Conclusion
Physical training induces at least three different adaptations of the adrenal
7
medulla. First, endurance training that increases 02max, results in a lower plasma
epinephrine concentration at a given absolute submaximal exercise intensity,
probably related to a diminished sympathetic innervation of the adrenal medulla.
Secondly however, compared at identical relative work loads as well as in response
to numerous non-exercise stimuli, athletes have a higher epinephrine secretion
capacity compared to sedentary subjects, indicating that a long-term adaptation of
endocrine glands to endurance training occurs - the development of a so-called
"sports adrenal medulla". Thirdly, longitudinal studies in animals indicate that both
adrenal medullary volume as well as adrenal epinephrine content increases as a
result of long-term physical training (Fig 5).
Since epinephrine has several advantageous effects on heart, skeletal muscle
and the central nervous system in relation to exercise performance the higher
capacity to secrete epinephrine in well-trained subjects is beneficial in competitive
sports. Finally, taking into account that the secretion capacity for epinephrine
decreases with age, physical training can cause a biological rejuvenation of the
adrenal medulla similar to events taking place in heart and skeletal muscle.
References:
1. Barwich, D., Hägele, H., Weiss, M., Weicker, H. Hormonal and metabolic
adjustment in patients with central Cushing´s disease after adrenalectomy. Int. J.
Sports Med. 2: 220-227, 1981.
2. Cheung, C.Y. Fetal adrenal medulla catecholamine response to hypoxia-direct
and neural components. Am. J. Physiol. 258: R1340-R1346, 1990.
3. Dela, F., Mikines, K.J., Linstow, M.V., Galbo, H. Heart rate and plasma
catecholamines during 24 h everyday life in trained and untrained men.
J.Appl.Physiol. 73: 2389-2394, 1992.
4. Deuster, P.A., Chrousos, G.P., Luger, A., DeBolt, J.E., Trostmann, U.H., Kyle,
8
S.B., Montgomery, L.C., Loriaux, D.L. Hormonal and metabolic responses of
untrained, moderately trained, and highly trained men to three exercise
intensities. Metabolism 38: 141-148, 1989.
5. Devalon, M.L., Miller, T.D., Squires, R.W., Rogers, P.J., Bove, A.A., Tyce, G.M.
Dopa in plasma increases during acute exercise and after exercise training. J.
Lab. Clin. Med. 114: 321-327, 1989.
6. Elliott, T.R. The innervation of the adrenal glands. J. Physiol.(Lond.) 46: 285-290,
1913.
7. Hagberg, J.M., Seals, D.R., Yerg, J.E., Gavin, J., Gingerich, R., Premachandra,
B., Holloszy, J.O. Metabolic responses to exercise in young and older athletes
and sedentary men. J.Appl.Physiol. 65: 900-908, 1988.
8. Hoelzer, D.R., Dalsky, G.P., Schwartz, N.S., Clutter, W.E., Shah, S.D., Holloszy,
J.O., Cryer, P.E. Epinephrine is not critical to prevention of hypoglycemia during
exercise in humans. Am. J. Physiol. 251: E104-E110, 1986.
9. Kjær, M., Mikines, K.J., Christensen, N.J., Tronier, B., Vinten, J., Sonne, B.,
Richter, E.A., Galbo, H. Glucose turnover and hormonal changes during insulininduced hypoglycemia in trained humans. J.Appl.Physiol. 57: 21-27, 1984.
10. Kjær, M., Christensen, N.J., Sonne, B., Richter, E.A., Galbo, H. Effect of exercise
on epinephrine turnover in trained and untrained male subjects. J.Appl.Physiol.
59: 1061-1067, 1985.
11. Kjær, M., Farrell, P.A., Christensen, N.J., Galbo, H. Increased epinephrine
response and inaccurate glucoregulation in exercising athletes. J.Appl.Physiol.
61: 1693-1700, 1986.
12. Kjær, M., Secher, N.H., Bach, F.W., Galbo, H. Role of motor center activity for
hormonal changes and substrate mobilization in exercising man. Am. J. Physiol.
253: R687-R695, 1987.
13. Kjær, M., Bangsbo, J., Lortie, G., Galbo, H. Hormonal response to exercise in
man: influence of hypoxia and physical training. Am. J. Physiol. 254: R197-R203,
1988a.
9
14. Kjær, M, Galbo, H. Effect of physical training on the capacity to secrete
epinephrine. J.Appl.Physiol. 64: 11-16, 1988b.
15. Kjær, M., Mikines, K.J., Linstow, M., Galbo, H., Nicolaisen, T. Effect of 5 weeks
detraining on epinephrine response to insulin induced hypoglycemia in athletes.
J. Appl.Physiol. 72: 1201-1204, 1992.
16. Kjær, M., Engfred, K., Fernandes, A., Secher, N.H., Galbo, H. Regulation of
hepatic glucose production during exercise in humans: role of
sympathoadrenergic activity. Am. J. Physiol. 265: E275-E283, 1993.
17. LeBlanc, J., Jobin, M., Cote, J., Samson, P., Labrie, A. Enhanced metabolic
response to caffeine in exercise-trained human subjects. J.Appl.Physiol. 59: 832837, 1985.
18. Lehmann, M., Dickhuth, H.H., Schmid, P., Porzig, H., Keul, J. Plasma
catecholamines, beta-adrenergic receptors and isoproterenol sensitivity in
endurance trained and non-endurance-trained volunteers. Eur. J. Appl. Physiol.
52: 362-369, 1984.
19. Sato, A., Sato, Y., Schmidt, R.F. Catecholamine secretion and adrenal nerve
activity in response to movements of normal and inflamed knee joints in cats. J.
Physiol.(Lond.) 375: 611-624, 1986.
20. Schuerink, A.J.W., Steffens, A.B., Bouritius, H., Dreteler, G.H., Bruntink, R.,
Remie, R., Zaagsma, J. Adrenal and sympathetic catecholamines in exercising
rats. Am. J. Physiol. 256: R155-R160, 1989.
21. Schmidt, K.N., Gosselin, L.E., Stanley, W.C. Endurance exercise training causes
adrenal medullary hypertrophy in young and old Fischer 344 rats. Horm. Metab.
Res. 24: 511-515, 1992.
22. Song, M.K., Ianuzzo, C.D., Saubert, C.W., Gollnick, P.D. The mode of adrenal
gland enlargement in the rat in response to exercise training. Pflügers Arch. 339:
59-68, 1973.
23. Sonne, B., Mikines, K.J., Richter, E.A., Christensen, N.J., Galbo, H. Role of liver
nerves and adrenal medulla in glucose turnover of running rats. J.Appl.Physiol.
10
59: 1640-1646, 1985.
24. Stallknecht, B., Kjær, M., Ploug, T., Maroun, L., Ohkuwa, T., Vinten, J., Mikines,
K.J., Galbo, H. Diminished epinephrine response to hypoglycemia despite
enlarged adrenal medulla in trained rats. Am. J. Physiol. 259: R998-R1003, 1990.
25. Svedenhag, J. The sympatho-adrenal system in physical conditioning. Acta
Physiol. Scand. 125,suppl.543: 1-74, 1985.
26. Tremblay, A., Pinsard, D., Coveney, S., Catellier, C., Laferriere, G., Richard, D.,
Nadeau, A. Counterregulatory response to insulin-induced hypoglycemia in
trained and nontrained humans. Metabolism 39: 1138-1143, 1990.
27. Winder, W.W., Hagberg, J.M., Hickson, R.C., Ehsani, A.A., McLane, J.A. Time
course of sympathoadrenal adaptation to endurance training in man.
J.Appl.Physiol. 45: 370-374, 1978.
28. Östman, J., Sjöstrand, N.O. Effect of heavy physical training on the
catecholamine levels of the heart and the adrenals of the rat. Acta Physiol.
Scand. 82: 202-208, 1971.
29. Yamaguchi, N. Sympathoadrenal system in neuroendocrine control of glucose:
mechanisms involved in the liver, pancreas, and adrenal gland under
hemorrhagic and hypoglycemic stress. Can. J. Physiol. Pharmacol. 70: 167-206,
1992.
Legends To Figures:
Fig 1. Epinephrine concentrations in arterial plasma during bicycle exercise (210 W)
in 7 endurance trained and 7 sedentary control males. Means and SE are given.
Modified from Kjær et al. 1988a.
Fig 2. Epinephrine clearance at rest and exhausting bicycle exercise in 6 endurance
trained athletes and 6 untrained subjects. The exercise intensity is expressed relative
11
to individual maximal oxygen uptake and epinephrine clearance determined by tracer
infusion. At rest clearance was identical in the two groups (28.8 ml/kg/min). Individual
values during exercise are given as % of individual resting values. Modified from
Kjær et al. 1985.
Fig 3. Epinephrine concentration in arterial plasma at rest and during 2 exercise
periods. 8 subjects were studied with and without partial neuromuscular blockade
with tubocurarine. The value 8 (5-13) min before start of exercise represents a
sample taken immediately before drug administration. Values are mean and SE. Star
and triangle denotes difference (p<0.05 and p<0.01, respectively) between curare
and control experiments. Modified from Kjær et al. 1987.
Fig 4. The adrenal medulla volume in response to 10 weeks of swim training in rats.
Means and SE are given. Stars denote difference (p<0.05) between swim training
group and control experiments. Modified from Stallknecht et al 1990.
Fig 5. no text