There is substantial evidence for mood effects, whether positive(e.g., antidepressant, anxiolytic) (34,36) or negative (e.g., overtraining-induced depression) (37) of acute/chronic physical exertion in humans. For obvious reasons, the neurochemical substrates of these psychological outcomes of exercise cannot be directly investigated in humans; on the other hand, indirect analyses of neurotransmitter biosynthesis have been frequently achieved through blood, and to a lesser extent, cerebrospinal fluid collection. For instance, the measurement of the effects of physical exercise upon tryptophan disposition in blood (9,22,27) and the concentration of 5-hydroxyindoleacetic acid (5-HIAA, the metabolite of 5-HT) in cerebrospinal fluid (38) has provided indirect indices favoring the hypothesis that exercise affects the central serotonergic systems. On the other hand, direct evidence for an effect of acute/repeated exercise on central serotonergic systems can be reached through animal models of exercise, especially rodents. However, animal models are far from being perfect since the reasons for the need to exercise in humans, including those emanating from social and health programs, cannot be reproduced in animal models. In addition, what is true for the reasons to exercise is also true for its consequences, especially those related to mood. Thus, rodents rarely (not to say never) fill out questionaries related to their mood changes! Hopefully, the rapid progress in neurosciences, especially in the field of psychopharmacology, has facilitated the usefulness of animal models in the study of anxiety- and depression-like behaviors. Taken together, the data obtained with these models have shown that numerous antidepressant and anxiolytic drugs have selective effects upon serotonergic systems, whether through blockade of 5-HT reuptake (e.g., selective serotonin reuptake inhibitors: SSRI) or through agonist/antagonist properties at 5-HT receptors(e.g., 5-HT1A receptor partial agonists)(6,29). This thus suggests that central serotonergic systems could be (one of) the link(s) between the exercising task and its positive mood effects. A prerequisite to this hypothesis is that acute exercise affects central serotonergic systems: this review will try to show that although numerous points remain obscure available data favor this hypothesis, at least in rodents.
ACUTE PHYSICAL EXERCISE AND ENTRY OF TRYPTOPHAN INTO THE BRAIN
Brain 5-HT synthesis depends on two main variables, the neuronal concentration of its precursor, tryptophan, and the activity of its rate-limiting enzyme, tryptophan hydroxylase (which converts tryptophan into 5-hydroxytryptophan) (11). Thus, increases in tryptophan availability increase 5-HT synthesis whereas the opposite treatments known to reduce tryptophan availability diminish 5-HT synthesis rate. In addition, depolarization and, in turn phosphorylation-induced increases in tryptophan hydroxylase activity, leads to increases in 5-HT synthesis independent from changes in tryptophan availability. We are thus confronted with two regulatory and complementary mechanisms that govern 5-HT synthesis rate, the first linked to precursor availability and the second intrinsic to the electrical properties of the serotonergic neurones. To date, most studies have focused on the influence of exercise upon tryptophan entry into the brain (and hence into the serotonergic neurones) rather than on tryptophan hydroxylase activity.
I have stated that tryptophan-induced changes in 5-HT synthesis find their origins in peripheral metabolism. Actually, brain proteolysis is weak and tryptophan is an essential amino acid, two points that underlie the key control of brain tryptophan levels by blood tryptophan levels. One illustration of this link is provided by the observation that peripheral(intraperitoneal or intravenous) tryptophan administration increases brain tryptophan levels and then 5-HT synthesis (11,16). However, things are rendered more complicated than suggested because blood tryptophan exists as two forms, with a great majority of tryptophan molecules being bound to albumin and only a small fraction being under a“free” form (35). Moreover, the equilibrium between bound and free tryptophan is displaced by free fatty acids in favor of the free form, owing to the competition between free fatty acids and tryptophan for albumin binding (35). For instance, fasting, which promotes high rates of lipolysis and hence high circulating levels of free fatty acids, is associated with increases in free tryptophan and decreases in bound tryptophan, respectively. However, beside these tryptophan-dependent regulatory mechanisms, brain 5-HT synthesis is also regulated by the circulating levels of the so-called “large neutral” amino acids, which include aromatic amino acids and branched-chain amino acids. Thus, these amino acids compete with tryptophan at the level of a transporter located on the blood-brain barrier, and increases or decreases in the levels of these competing amino acids can promote(independently from changes in circulating tryptophan levels) decreases and increases, respectively, in brain tryptophan levels and then 5-HT synthesis(26). For instance, peripheral administration of the branched-chain amino acid valine leads to a decrease in brain tryptophan availability and 5-HT synthesis. Lastly, one additional variable to be taken into account is the intrinsic concentration of circulating (free/bound) tryptophan. Actually, the latter is controlled by the activity of tryptophan pyrrolase, an enzyme located in the liver which rapidly metabolizes tryptophan via the so-called “kynurenine pathway” (2). If tryptophan pyrrolase is activated (e.g., by glucocorticoids), circulating free and bound tryptophan levels will decrease, thereby reducing brain tryptophan availability few hours later (12).
In keeping with the well-documented influence of physical exercise upon lipolysis, acute physical exercise increases blood free tryptophan and decreases albuminbound tryptophan both in (trained) animals(10,14,15) and humans(9,22,27). Because in most cases circulating total (free plus bound) tryptophan level is not affected by exercise, it is likely that the increase in free tryptophan during exercise is solely due to lipolysis and that hepatic tryptophan pyrrolase activity undergoes weak changes, if any, during acute physical exercise. This may seem paradoxical in view of the activation of the corticotropic axis during exercise; actually, the hypothesis that glucocorticoid-elicited hyperactivity of tryptophan pyrrolase is counterbalanced by exercise-related inhibitory factors cannot be discarded. Besides, several studies have analyzed the impact of acute physical exercise on the other regulators of tryptophan entry into the brain. For instance, whether acute physical exercise affects the blood levels of those amino acids competing with tryptophan for entry into the brain is a question that has received some attention. Unfortunately, animal and human studies have led to the proposal that exercise decreases(7-9,23), increases(1,10,25,32), or does not affect(15,21,22) branched-chain amino acid levels. On the other hand, most of these studies have reported a stimulatory effect of exercise upon the ratio of free tryptophan to the sum of the other competing amino acids, i.e., exercise stimulates tryptophan influx into the brain compartment. If the hypothesis that blood free tryptophan (rather than total tryptophan) governs brain tryptophan levels is correct (see below), increases in the ratio of free tryptophan over the sum of the competing amino acids should lead to increases in brain tryptophan levels. Actually, analyses of whole brains (14,15), brain regions(10,16), or cerebrospinal fluid samples(17) from exercising rats confirm the above hypothesis. Nevertheless, one could argue that exercise-induced increases in blood free tryptophan and brain tryptophan are not causally related. For instance, if exercise increases in an unspecific manner blood-brain barrier permeability(i.e., through cerebrovascular changes or alterations in the kinetics of the neutral amino acid transporter) (12), brain levels of all amino acids (including tryptophan) are increased independently from exercise-elicited changes in the blood levels of these amino acids. One study conducted with trained animals allows us to reject this hypothesis because beside increases in blood free tryptophan and brain tryptophan, neither the blood levels nor the brain levels of most neutral amino acids were affected by acute treadmill exercise (15). In this study, blood total tryptophan level remained unaffected by acute exercise, thereby indicating that lipolysis-elicited increases in blood free tryptophan directly lead to increases in brain tryptophan levels (16). This result thus shows that acute exercise differs from other models (e.g., carbohydrate ingestion, immobilization stress) where increases in brain tryptophan are dependent upon 1) increases in blood total (rather than free) tryptophan, 2) decreases in the competition for entry into the brain between tryptophan and the other neutral amino acids, or 3) unspecific increases in blood-brain barrier permeability (12).
I would like to point out that this absolute relationship between blood free tryptophan and brain tryptophan in exercising animals is only valid under basal conditions. For instance, the relationship between blood tryptophan and brain tryptophan is more complex if animals display increased sympathetic tone prior to their acute exercise task. Thus, pretreating the animals with a norepinephrine reuptake inhibitor will amplify exercise-induced increases in brain tryptophan levels compared with undrugged exercising animals(14), an amplification possibly due to the stimulatory effects of high sympathetic tone upon lipolysis and blood-brain barrier permeability and to the inhibitory effects of high sympathetic tone upon the competition between tryptophan and the other neutral amino acids for entry into the brain (12).
ACUTE PHYSICAL EXERCISE AND BRAIN 5-HT SYNTHESIS
Earlier studies have shown that administration of tryptophan increases brain tryptophan levels and 5-HT synthesis. Indeed, this direct relationship between brain tryptophan and 5-HT synthesis relies on the lack of saturation of the rate-limiting enzyme in 5-HT synthesis (tryptophan hydroxylase) under physiological situations (11). This also holds true for exercise-induced increases in blood free tryptophan which thereby stimulate tryptophan entry into the brain and then 5-HT synthesis, as assessed either through the analysis of 5-HIAA (in whole brain, brain regions, or cerebrospinal fluid)(4,10,14,16,17,30) or the analysis of 5-HT turnover (16). In general, exercise-induced increases in 5-HT synthesis do not increase 5-HT itself(4,10,14,16,19,30) because most of tissue 5-HT levels are constituted by neuronal 5-HT stored in vesicles whereas 5-HT newly formed from tryptophan is rapidly metabolized into 5-HIAA. In keeping with this remark, it is noteworthy that the exercise-induced increase in 5-HIAA levels in different brain regions concerns regions particularly enriched either in serotonergic cell bodies (i.e., in midbrain) or in serotonergic nerve terminals (e.g., in hippocampus, striatum, cortex, and hypothalamus).
One study (19), but not another(10), reported that the extent to which acute exercise increases brain tryptophan and, in turn, 5-HIAA levels depends on the duration of the training period. Thus, acute treadmill exercise had weak, albeit significant, stimulatory effects on brain tryptophan and 5-HIAA levels in 8-wk trained rats compared with the stimulatory effects of exercise in 1-wk trained rats. Surprisingly, 5-HT levels, which are not modified by acute exercise in 1-wk trained rats, were decreased by acute exercise in 8-wk trained rats(19). This suggested that under conditions of low(exercise-induced) elevations in neuronal tryptophan levels, such as those measured following an 8-wk training, 5-HT utilization/metabolism into 5-HIAA is not compensated by concomitant increases in 5-HT synthesis (thereby reducing 5-HT levels). Whether this lack of compensation arises solely from the weak increases in brain tryptophan entry is not certain, and it is possible that a diminished activity of tryptophan hydroxylase during acute exercise participates in the aforementioned imbalance. This hypothesis is supported by the following observations in 1-wk trained rats: first, the respective amplitudes of acute exercise-elicited increases in tryptophan, 5-HT and 5-HIAA (in studies where both were measured), reveals that acute exercise has differential effects upon these indoles(14,17). Thus, although tryptophan rises are large, those regarding 5-HIAA are relatively weak, suggesting that one step in 5-HT biosynthesis is partly inhibited during exertion. Second, in one study in which control rats and acutely exercised rats were pretreated with different doses of tryptophan, it was found that tryptophan utilization toward synthesis of 5-HT (and 5-HIAA) was diminished by acute exercise, possibly through an inhibitory effect of exercise upon the affinity of tryptophan hydroxylase rather than through an early saturation of the enzyme(18). Third, in another study, we compared 5-HT and 5-HIAA levels in different brain regions of control and acutely exercised rats that were pretreated with high doses of tryptophan (so that exercise-induced increases in central tryptophan levels were no more detectable): indeed, tryptophan-induced increases in 5-HT and 5-HIAA levels were lower in the hippocampus and the striatum of exercising rats compared with those measured in control animals (16). Taken together, all these data suggest that under certain conditions exercise-induced increases in brain tryptophan entry may be associated with a partial inhibition of 5-HT synthesis. Whether this last phenomenon results from increased 5-HT release, stimulation of 5-HT(1A/1B) autoreceptors and feedback inhibitory processes upon tryptophan hydroxylase activity is an interesting possibility that remains to be verified.
ACUTE PHYSICAL EXERCISE AND 5-HT FUNCTIONS
Because acute exercise increases 5-HT synthesis in several brain regions, it could be assumed that this paradigm increases 5-HT release and then several 5-HT receptor-mediated functions. Actually, the relationships between synthesis and release are not so clear-cut, and in many instances, increases in 5-HT synthesis lead to increases in intraneuronal metabolism in 5-HIAA without altering the number of 5-HT molecules effectively released in the synaptic clefts (12). Moreover, another possibility is that under several stressful situations increases in 5-HT synthesis take place after 5-HT has been released to avoid a release-induced depletion in neuronal 5-HT stores (12). With regard to this uncertainty, one interesting approach concerns the measurement of extracellular (synaptic) 5-HT levels using relevant monitoring techniques (such as microdialysis). To date, only one study reported that acute exercise in rats is associated with a lack of variation in extracellular 5-HT levels; alternatively, these levels begin to decrease when exercise stops (28). In keeping with electrophysiological evidence that treadmill locomotion is associated with an increased discharge rate in serotonergic cell bodies (at least those located in the raphe obscurus) and the suggestion that locomotion is associated with a rapid 5-HT reuptake in serotonergic nerves (39), it is likely that this apparent lack of change in extracellular 5-HT during exercise represents the sum of increased 5-HT release plus increased 5-HT reuptake. In addition, some authors have reported an increase in extracellular 5-HT levels in the frontal cortex (where serotonergic nerve terminals are abundant) of(untrained) walking rats (33), whereas others have reported increases in extracellular 5-HIAA levels in the frontal cortex, but not the raphe dorsalis (a nucleus enriched in serotonergic cell bodies) of exercising rats (20). However, the putative impact of stress in untrained animals and the possibility that 5-HIAA levels do not reflect 5-HT release impede any conclusion.
Another means to analyze whether the effects of acute exercise upon serotonergic systems have a functional impact is linked to the analysis of 5-HT receptor-mediated functions. However, it is important to note here that this tool does not provide any clear-cut information regarding the net effect of acute exercise on 5-HT release. For instance, glucocorticoids affect gene expression, binding, and/or function of several postsynaptic 5-HT receptors without affecting 5-HT release from presynaptic serotonergic cells(12). As far as 5-HT receptor-mediated functions are concerned, a recent study has found that neither training nor acute exercise in trained animals affected the hyperphagia that follows the acute stimulation of 5-HT1A autoreceptors, which are located on the cell bodies of serotonergic neurones and are the targets of numerous antidepressant/anxiolytic compounds (13). Moreover, the magnitude of the so-called “flat body posture” and “forepaw treading” responses to the stimulation of postsynaptic 5-HT1A receptors located in the brainstem and/or the spinal cord (i.e., behavioral responses that are often taken as indices of central 5-HT1A receptor function) were not affected by training and/or acute exercise(13). Lastly, some studies have used 5-HT receptor blockade in vivo to show that different physiological responses triggered by exercise may be modulated by serotonergic systems. In humans, it has been shown that administration of the 5-HT2A receptor blocker ketanserin prevents exercise-elicited prolactin release(24). This result suggests that exercise actually triggers 5-HT release and stimulation of those 5-HT2A receptors controlling prolactin release. Alternatively, because it is unknown whether ketanserin, at the doses used, affected only 5-HT2A receptors, a clearcut conclusion is difficult to draw. Two animal studies have reported that 5-HT2A/5-HT2C receptor blockade increases time to reach exhaustion (4,5). Beside suggesting either that exercise releases 5-HT, which in turn stimulates 5-HT2A and/or 5-HT2C receptors involved in the control of fatigue, or that 5-HT, under basal conditions, exerts a tonic control upon these receptors (the blockade of which will diminish fatigue), these studies may provide evidence for a negative influence of 5-HT, and in turn, 5-HT2A/5-HT2C receptor stimulation upon exercise endurance. That 5-HT2A/5-HT2C receptor stimulation increases treadmill fatigue(3,22) supports the involvement of 5-HT2A and 5-HT2C receptors in the control of endurance. However, the side effects of the drugs used in these studies (e.g., anxiety, hypolocomotion) render any strong statement problematic.
This survey has tried to delineate the influence of acute exercise upon central serotonergic systems. Clearly, one key issue concerns the functional impact of acute exercise on 5-HT transmission/function. Another issue concerns the possible links between central serotonergic systems and exercise-elicited mood changes. Recent progress in 5-HT receptor pharmacology has helped to define part of the mechanisms through which antidepressant and anxiolytic therapies exert their positive effects. In keeping with this progress and in view of the significant available information regarding dysfunctions of the serotonergic system in psychiatric patients, it is likely that 5-HT plays a pivotal role in the etiology of some forms of depression and anxiety. However, the real link between central 5-HT and the positive effects of exercise on mood, if any, is still to be determined. It is my belief that understanding of this issue should increase in the near future provided that fundamental and clinical scientists collaborate constructively.
1. Ahlborg, G., P. Felig, L. Hagenfeldt, R. Hendler, and J. Wahren. Substrate turnover during prolonged exercise in man. J. Clin. Invest.
2. Badawy, A. A. B. The functions and regulation of tryptophan
pyrrolase. Life Sci.
3. Bailey, S. P., J. M. Davis, and E. N. Ahlborn. Effect of increased brain serotonergic (5-HT1C
) activity on endurance performance in the rat
. Acta Physiol. Scand.
4. Bailey, S. P., J. M. Davis, and E. N. Ahlborn. Neuroendocrine and substrate responses to altered brain 5-HT activity during prolonged exercise to fatigue. J. Applied Physiol.
5. Bailey, S. P., J. M. Davis, and E. N. Ahlborn. Serotonergic agonists and antagonists affect endurance performance.Int. J. Sports Med.
6. Blier, P. and C. De Montigny. Current advances and trends in the treatment of depression. Trends Pharmacol. Sci.
7. Blomstrand, E., F. Celsing, and E. A. Newsholme. Changes in plasma concentrations of aromatic and branched-chain amino acids during sustained exercise in man and their possible role in fatigue. Acta Physiol. Scand.
8. Blomstrand, E., P. Hassmen, B. Ekblom, and E. A. Newsholme. Administration of branched-chain amino acids during sustained exercise: effects on performance and on plasma concentration of some amino acids. Eur. J. Appl. Physiol.
9. Blomstrand, E., P. Hassmen, and E. A. Newsholme. Effect of branched-chain amino acid supplementation on mental performance. Acta Physiol. Scand.
10. Blomstrand, E., D. Perrett, M. Parry-Billings, E. A. Newsholme. Effect of sustained exercise on plasma amino acid concentrations and on 5-hydroxytryptamine metabolism in six different regions of the rat
.Acta Physiol. Scand.
11. Carlsson, A. and M. Lindqvist. The effect of L-tryptophan
and psychotropic drugs on the formation of 5-hydroxytryptophan in the mouse brain in vivo. J. Neural Transm.
12. Chaouloff, F. Physiopharmacological interactions between stress hormones and central serotonergic systems. Brain Res. Rev.
13. Chaouloff, F. Influence of physical exercise on 5-HT1A
receptorand anxiety-related behaviours. Neurosci. Lett.
14. Chaouloff, F., J. L. Elghozi, Y. Guezennec, and D. Laude. Effects of conditioned running on plasma, liver and brain tryptophan
and on brain 5-hydroxytryptamine metabolism of the rat
. Br. J. Pharmacol.
15. Chaouloff, F., G. A. Kennett, B. Serrurier, D. Merino, and G. Curzon. Amino acid analysis demonstrates that increased plasma free tryptophan
causes the increase of brain tryptophan
during exercise in the rat
16. Chaouloff, F., D. Laude, and J. L. Elghozi. Physical exercise: evidence for differential consequences of tryptophan
on 5-HT synthesis and metabolism in central serotonergic cell bodies and terminals.J. Neural Transm.
17. Chaouloff, F., D. Laude, Y. Guezennec, and J. L. Elghozi. Motor activity increases tryptophan
, 5-hydroxyindoleacetic acid, and homovanillic acid in ventricular cerebrospinal fluid of the conscious rat
18. Chaouloff, F., D. Laude, D. Merino, B. Serrurier. Y. Guezennec, and J. L. Elghozi. Amphetamine, and α-methyl-p-tyrosine affect the exercise-induced imbalance between the availability of tryptophan
and synthesis of serotonin
in the brain of the rat
19. Chaouloff, F., D. Laude, B. Serrurier, D. Merino, Y. Guezennec, and J. L. Elghozi. Brain serotonin
response to exercise in the rat
: the influence of training duration. Biogenic Amines
20. Clement, H. W., F. Schafer, C. Ruwe, D. Gemsa, and W. Wesemann. Stress-induced changes of extracellular 5-hydroxyindoleacetic acid concentrations followed in the nucleus raphe dorsalis and the frontal cortex of the rat
. Brain Res.
21. Conlay, L. A., R. J. Wurtman, I. Lopez-Coviella, et al. Effects of running the Boston marathon on plasma concentrations of large neutral amino acids. J. Neural Transm.
22. Davis, J. M., S. P. Bailey, J. A. Woods, F. J. Galiano, M. T. Hamilton, and W. P. Bartoli. Effects of carbohydrate feedings on plasma free tryptophan
and branched-chain amino acids during prolonged cycling.Eur. J. Applied Physiol.
23. Decombaz, J., P. Reinhardt, K. Anantharaman, G. von Glutz, and J. R. Poortmans.Biochemical changes in a 100 km run: free amino acids, urea, and creatinine. Eur. J. Applied Physiol.
24. De Meirleir, K., M. L'hermite-Baleriaux, M. L'hermite, R. Rost, and W. Hollmann. Evidence for serotoninergic control of exercise-induced prolactin release. Horm. Metab. Res.
25. Felig, P. and J. Wahren. Amino acid metabolism in exercising man. J. Clin. Invest.
26. Fernstrom, J. D., and R. J. Wurtman. Brain serotonin
content: physiological regulation by plasma neutral amino acids.Science
27. Fischer, H. G., W. Hollmann, and K. De Meirleir. Exercise changes in plasma tryptophan
fractions and relationship with prolactin. Int. J. Sports Med.
28. Gerin, C., A. Legrand, and A. Privat. Study of 5-HT release with a chronically implanted microdialysis probe in the ventral horn of the spinal cord of unrestrained rats during exercise on a treadmill.J. Neurosci. Meth.
29. Handley, S. L., and J. W. McBlane, 5-HT drugs in animal models of anxiety. Psychopharmacology
30. Heyes, M. P., E. S. Garnett, and G. Coates. Nigrostriatal dopaminergic activity is increased during exhaustive exercise stress in rats. Life Sci.
31. Jacobs, B. L. and C. A. Fornal. 5-HT and motor control: a hypothesis. Trends Neurosci.
32. Ji, L. L., R. H. Miller, F. J. Nagle, H. A. Lardy, and F. W. Stratman. Amino acid metabolism during exercise in trained rats: the potential role of carnitine in the metabolic fate of branched-chain amino acids. Metabolism
33. Kurosawa, M., K. Okada, A. Sato, and S. Uchida. Extracellular release of acetylcholine, noradrenaline and serotonin
increases in the cerebral cortex during walking in conscious rats. Neurosci. Lett.
34. Martinsen, E. W. The role of aerobic exercise in the treatment of depression. Stress Med.
35. Mcmenamy, R. H. Binding of indole analogues to human serum albumin: effects of fatty acids. J. Biol. Chem.
36. Morgan, W. P. Reduction of state anxiety following acute physical activity. In: Exercise and Mental Health: The Series in Health Psychology and Behavioral Medicine
, W. P. Morgan and S. E. Goldston (Eds.). New York: Hemisphere Publishing Corp., 1987, pp. 105-109.
37. Morgan, W. P., D. L. Costill, M. G. Flynn, J. S. Raglin, and P. J. O'Connor. Mood disturbance following increased training in swimmers.Med. Sci. Sports Exerc.
38. Post, R. M., J. Kotin, F. K. Goodwin, and E. K. Gordon. Psychomotor activity and cerebrospinal fluid amine metabolites in affective illness. Am. J. Psychiat.
39. Wallis, D. I. 5-HT receptors
involved in initiation or modulation of motor patterns: opportunities for drug development.Trends. Pharmacol. Sci.