Activity anorexia is a biobehavioral condition in animals with the simultaneous (and paradoxical) occurrence of dietary restriction (DR) and excessive physical activity. These two factors may combine to produce starvation and death (21). It occurs in both rats and mice, especially when young. This is an unusual response, since energy expenditure is increasing at a time when food intake is limited. These observations have important clinical relevance, since some patients with anorexia nervosa (AN) often use excess exercising as a means of maintaining a reduced body weight by burning off “excess” calories. In these conditions, exercise may also provide a hedonic reward instead of that derived from food.
How then may physical activity decrease both food intake and body weight? The decrease in food consumption following activity could be the result of a number of explanations: changes in hypothalamic 5-hydroxytryptamine (5-HT) concentrations affecting satiety (13), elevation in endorphin levels during activity and weight loss (21), or increased motivation for exercise decreasing food intake as in the hypothesis of motivational interaction of eating and exercise (21).
Both catecholaminergic and serotonergic systems are involved in feeding behavior in rats (16,17). Alpha-adrenergic stimulation in the area of the paraventricular nucleus of the hypothalamus induces feeding, whereas beta-adrenergic stimulation is inhibitory (16). Dihydroxyphenylalanine (DOPA)-depleted animals may stop all operant behaviors and actually starve themselves to death (12). This may be analogous to the malnourished patient with AN in whom appetite may be suppressed if central noradrenergic pathways have insufficient norepinephrine (NE) and/or if DOPA levels are chronically depleted.
We have previously investigated in mice the effect of tyrosine administration on cognitive and neurochemical alterations caused by DR (1). Mice fed 60%DR showed improved maze performance, whereas that of 40%DR animals was significantly impaired but could be restored by tyrosine. Maze performance was related to increased beta/alpha tone in the hippocampus, whereas there were opposite changes (increase in alpha/beta tone) in the hypothalamus. Tyrosine treatment reversed these alterations. Tyrosine administration has been shown to decrease stress in humans (2). The purpose of this study is to report the effects in mice of tyrosine on the behavior and neurochemistry of anorexia caused by exercise. It is a completely different paradigm to the DR model (1) and important because activity may lead to anorexia, and some anorectic patients use activity to prevent weight gain. The results may also be relevant to the study of exercise physiology and sports nutrition.
In all experiments, all ACSM guidelines for animal care were followed and the local animal care facility board authorized all the protocols.
Our experiments are derived from the protocols described by Routtenberg in rats with modifications for mice (23). Two-month-old female Sabra mice with similar mean weight were divided into four groups of 10 mice (Control Saline or Tyrosine, Activity Saline or Tyrosine). In the Activity groups, mice were placed two in a cage with a fixed wheel attached to an electronic revolution counter. The controls received the same treatment while the wheels were inactive. The experiments were complete, with their own controls. Each experiment was calculated separately and then averaged. Each protocol was repeated three times on different mice to check the consistency of the results.
Cages containing wood-chip bedding were placed in a temperature-controlled room at 22°C on a 12-h light/dark cycle (lights on at 7:00 a.m.). Mice had access to water ad libitum. The food provided was Purina chow (Ralston Purina Company, St. Louis, MO). They were fed for 2 h·d−1 over 1 wk to adapt to the feeding schedule.
During feeding, the mice were moved to special cages where food was available and where they had no access to the wheel. Weight and food consumption were monitored over 2 wk using food cakes (26), which were weighed before and after the feeding period. After a week of conditioning, wheel activity was commenced and revolutions were monitored with an electronic revolutions counter every 21.5 h. Each group was injected daily with either saline or tyrosine (100 mg·kg−1·d−1) 1 h before the maze testing. The protocol was as follows: Administration of saline or tyrosine beginning at 8.00 a.m., maze performance at 9.00 a.m. (about 5–10 min for each mouse), and food consumption (2 h). Wheel running took place over approximately 21.5 h·d−1.
Brains were dissected and assayed for adrenergic and serotonergic activity in the hippocampus and hypothalamus. Menstrual cycle measurements showed arrest at the proestrus stage as a function of both semistarvation and hyperactivity (22).
Behavioral paradigm eight-arm spatial maze.
The radial eight-arm spatial maze was similar to that developed for rats by Olton and Samuelson, but has been scaled down for mice (1,14). The reward used was a small amount of food, since the two groups were food restricted. Water was not used because it would have demanded deprivation over 23.5 h·d−1, 6 d before and during the days of the maze, which could have caused exhaustion of the mice and affected exercise tolerance.
On the first day of testing, mice were introduced to the radial maze for 10 min, and the first 16 entries to the arms were recorded. On this day, no reward was offered. On the next day a small amount of food was presented in each arm. The animals were observed until they entered all eight arms or until they completed 16 entries (whichever came first). The animals were then taken out of the apparatus. The number of correct entries, once to each of the eight arms within the 16 trials (maximum possible, eight) was recorded and the test was repeated for 5 d. Tyrosine (100 mg·kg−1·d−1) or normal saline was administered to Control and Activity mice 1 h before the behavioral tests. The results were calculated according to the following formula for each mouse and then were averaged: number of entries: (day 2 + day 3 + day 4 + day 5) − 4 (day 1) (equivalent to the area under the curve). The number of correct entries needed to complete the maze was calculated for each mouse during days 2–5, minus four times baseline (day 1).
Tyrosine did not affect food consumption in controls and there was no significant difference in weight between these groups, Control Saline, Control Tyrosine, Activity Saline, and Activity Tyrosine (Fig. 1). The cognitive function in Activity Tyrosine was improved compared with Control Saline independent of food consumption, which was the same.
Catecholamines and 5-HT analysis.
Catecholamines, 5-HT, and their intermediates were separated and detected by HPLC/ECD, using dehydroxybenzylamine as an internal standard (1).
Tyrosine hydroxylase activity.
Tyrosine hydroxylase (TH) was evaluated by measuring the level of DOPA that was produced during a 30-min period as described by Nagatsu et al., with modifications (30). On the 15th day of the experiment, each group was divided into two: half of the mice were sacrificed at time zero, whereas the other half were sacrificed at 30 min; 100 mg·kg−1 benzyloxyamine (an inhibitor of DOPA decarboxylase) was administered intraperitoneally 10 min before the sacrifice of the first group and 40 min before the sacrifice of the second group. Hence, for example, if 0800 h was the time of sacrifice of the first group, the inhibitor was administered to both groups at 0750 h, whereas tyrosine or saline were administered half an hour (0730 h) before the sacrifice of the first group and an hour (0700) before the sacrifice of the second group. The level of DOPA was evaluated using HPLC/ECD as described above.
Analysis of data.
Data are given as means and standard errors. Differences between groups were examined by analysis of variance (ANOVA) with multiple levels. Homogeneity of variances of the different groups was then assessed by Bartlett’s test. Tukey post hoc test was used to detect specific differences. Differences within groups (e.g., Saline vs Tyrosine groups) were tested by Student’s t-test. A two-tailed P value < 0.05 was taken as the level of significance.
Figure 1 shows the weight changes in the Activity and Control groups. Mice in the Activity group lost weight significantly faster than the Control groups (27% and 15%, respectively, over 14 d). Tyrosine did not influence body weight in either group.
During the first week, there was a gradual increase in food consumption from about 1 g·d−1 per mouse on the first day until it reached 7 to 8 g·d−1 per mouse (data not shown). Figure 2 A shows the changes in food consumption during the second week of the experiment. The Activity Saline group consumed 22% less than the Control Saline group (a, P < 0.001). Tyrosine restored food consumption to normal (b, P < 0.001).
Figure 2 B describes the number of wheel revolutions after injections of either saline or tyrosine. Saline caused a significant decrease (−18%) in revolutions from 4750 revolutions·d−1 to 3900 revolutions·d−1 per mouse (a, P < 0.001) (probably because of the stress following injection), whereas tyrosine caused an increase (+18%) at the same level of significance from 4600 revolutions·d−1 to 5600 revolutions·d−1 per mouse (b, P < 0.001). Two days after cessation of the injections, activity returned to preinjection levels in both treatment groups.
Figure 3 shows the results of performance in the eight-arm maze. A decrease in the number of entries implies learning (14). Both activity and tyrosine administration improved cognitive performance. Activity was better than Controls (a, P < 0.05) and Activity Tyrosine was significantly better than Activity Saline (b, P < 0.001).
5-HT, catecholamines, and their intermediates.
Figure 4 describes the effect of activity and tyrosine on the levels of 5-HT and catecholamine intermediates in the hypothalamus. Activity significantly increased 5-HT and 5-HIAA (a, P < 0.001; b, P < 0.01, respectively), whereas tyrosine caused a significant decrease in 5-HT (b, P < 0.05) and increase in control 5-HIAA (a, P < 0.01) (Fig. 4 A and B). In addition, activity caused a significant increase in dopamine (Fig. 4 C) (a, P < 0.05).
With regard to hippocampal 5-HT and its metabolites, there was a significant increase in 5-HIAA in the Activity Saline (b, P < 0.05) and Control Tyrosine (a, P < 0.01) (Fig. 5 A) groups. Activity and Control Tyrosine showed significantly decreased levels of DOPAC (Fig. 5 B) (b and a, respectively, P < 0.001). Tyrosine had no effect in the Activity group. There were no significant changes in hippocampal 5-HT in Activity Saline in relation to Control Saline and in Control Tyrosine versus Control Saline, whereas there was decrease in the level of 5-HT in the Activity Tyrosine toward Activity Saline groups that did not reach significance (high SEM). There were no differences in hippocampal dopamine between Control and Activity Saline, Activity Saline and Activity Tyrosine, or Control Saline and Tyrosine.
In both brain areas, there were no significant changes in the levels of other catecholamines and their intermediates: NE, methoxyhydroxyphenylglycol, epinephrine (EPI), DOPA, and homovanillic acid (data not shown).
Figure 6 shows the TH activity in the hypothalamus (measured by changes in DOPA concentrations). Activity significantly increased the level of the enzyme (a, P < 0.05), which was reversed by tyrosine (b, P < 0.05). In the hippocampus there were no significant changes after activity and no effect of tyrosine was observed (data not shown).
Activity may lead to anorexia in some animal models, and some anorectic patients use activity in order to prevent weight gain (so-called anorexia athletica (6)). Both activity and appetite regulation are affected by sympathetic and serotonergic activity. Activity together with a restricted diet (2 h feeding per day) caused extreme weight loss and anorexia (21). When food restriction and running occur together, the animals increase exercising over time, which is “unusual,” since energy expenditure is increasing when food intake is limited. The reasons for these apparently paradoxical behaviors are probably multifactorial, involving different neurotransmitter and modulator responses (16,18,24) as well as complex motivational effects (21). This work describes the beneficial effects of tyrosine on food consumption, cognitive function, and exercise tolerance.
Food consumption was significantly reduced by activity and was associated with an increase in hypothalamic dopamine, 5-HT, and 5-HIAA. Tyrosine reversed these changes, in particular normalizing 5-HT concentrations. The increase in 5-HT and 5-HIAA in the hypothalamus and in the level of 5-HIAA in the hippocampus was also found in rats on an exercise schedule without DR (3). Decrease in food intake following increased brain 5-HT levels is well documented and is the basis of attempts to modify appetite in the treatment of obesity (31). Lipolysis may be the main determinant of these changes because free fatty acids displace tryptophan bound to albumin, allowing them to cross the blood-brain barrier (4). Our results suggest that tyrosine might be used in the treatment of activity anorexia by preventing exercise-induced increases in hypothalamic 5-HT concentrations.
Cognitive function in the eight-arm maze improved after both exercise and tyrosine administration (P < 0.05 and P < 0.001, respectively). In the hippocampus following exercise, there was a decrease in DOPAC and an increase in 5-HIAA. Tyrosine treatment caused a significant increase in both exercising and a further improvement in maze performance. The improved cognitive function might be explained by the increase in opiates/opiate peptides, which are known as regulators of learning and memory (19), and their concentration increases during both wheel running (21) and DR (24). Another possible mechanism is secondary to the increased concentration of dopamine that is involved in long-term potentiation (8). Tyrosine also improved cognitive function in DR mice (1) and in men under stress (25), perhaps by supplying substrates to noradrenergic neurons whose impaired activity may be caused by lack of tyrosine precursor for NE synthesis.
Exercise tolerance and fatigue.
In prolonged exercise, there is increased muscle utilization of branched chain amino acids (BCAA). This leads to a relative increase in the ratio of free tryptophan to the large neutral amino acids (LNAA) (including tyrosine), and preferential entry of tryptophan to the brain (29). The consequent elevated concentrations of 5-HT in specific areas of the brain may contribute to the development of central/mental fatigue during and after exercise (20). This is the rationale for the supplementation of BCAA to combat exercise fatigue (9). Recent work suggests that intake of BCAA may improve performance in slower runners in the marathon and decrease physical and mental exertion in laboratory animals, although the results are rather marginal (6). According to Davis and Bailey (5), increases and decreases in brain 5-HT synthesis during prolonged exercise hasten and delay, respectively, fatigue; and nutritional manipulations designed to attenuate brain 5-HT synthesis during exercise improve endurance performance.
We should like to suggest an alternative therapy for fatigue: decreasing tryptophan availability through supplementation with tyrosine to augment the catecholamine to 5-HT tone. No studies have yet been performed to test whether tyrosine supplements may improve exercise tolerance (27), although there is much circumstantial evidence supporting such an effect.
Physical exercise enhances the synthesis of catecholamines in various tissues: NE and EPI in the brain stem, heart, spleen, and adrenals (7), and dopamine in the CNS, where there is a large reserve of tyrosine (4). The activity of dopamine rather than NE is predominant in the central control of motor activity (11) and seems to be involved, together with 5-HT (see below), in the onset of fatigue. Since the level of brain tyrosine was unaltered by running (4), it suggests that exercise may accelerate the synthesis of dopamine by increasing the activity of tyrosine hydroxylase, which is what we found. In the hypothalamus, tyrosine hydroxylase activity was increased by exercise and decreased after tyrosine administration, probably because of feedback inhibition (28). The destruction of catecholaminergic neurons by 6-hydroxydopamine affects the ability of rats to exercise (11), which could be partially reversed by administration of the dopamine agonist apomorphine. We suggest the use of the natural amino acid tyrosine could serve the same purpose and combat central fatigue.
There are other mechanisms for such catecholamine/5-HT interactions. Chaouloff et al. (4) showed that the control of 5-HT synthesis in brain from tryptophan is altered during exercise by catecholaminergic activity (especially dopamine), which may delay exhaustion. In addition, pretreatment of exercising rats with amphetamine, which releases dopamine, extends the time to reach exhaustion, thus supporting the hypothesis (4). Numerous anatomical and pharmacological data indicate that dopaminergic activity can influence the metabolism of 5-HT (10). Dopaminergic neurons decrease the release of 5-HT in the substantia nigra and striatum (10). In addition, studies on synaptosomal preparations of forebrain have revealed that dopamine may inhibit tryptophan hydroxylase (4). Administration of tryptophan at the onset of running, or after 24 h of food deprivation (15), significantly increased its brain concentration. However, exercise did not modify the elevations of 5-HIAA in brain induced by either tryptophan or fasting. Administration of compounds known to affect the activity of dopamine, such as amphetamine, potentiate the inhibition of 5-HT synthesis induced by running, whereas alpha-methylparatyrosine prevented this effect of exercise. It was concluded that brain tryptophan regulation of 5-HT synthesis was altered during exercise depending on central catecholaminergic activity (17).
Activity with diet restriction may occur in anorexia nervosa, and people who start dieting and exercise. We found in our model that there is increase in 5-HT that causes fatigue, and also in dopamine production, which regulates the 5-HT levels. Tyrosine is beneficial and may decrease fatigue in normal people, whereas in anorectics there may be a decrease in 5-HT that could lead to increased food intake. The model of activity anorexia is well established (23). We are not suggesting that anorectic subjects should exercise more, rather that the effect of tyrosine might enhance appetite.
Additional research needs to be done in humans to vali-date these findings, which could also be relevant to the management of obesity. Obesity is treated on the basis of diet and exercise, and the development of fatigue in obese people occurs earlier than in normal subjects. Therefore, administration of tyrosine might improve their capacity for exercise.
In conclusion, our principal findings are that activity anorexia is associated with increased hypothalamic 5-HT concentrations. Tyrosine administration reverses this, and improves food consumption, cognitive behavior, and exercise performance. These results require confirmation in human subjects. If so, then activity with nutritional modulation of neurotransmitter concentrations may have implications for the treatment of psychobehavioral problems associated with prolonged dieting as in obesity and in the extreme case of anorexia nervosa. Furthermore, in normal subjects, tyrosine may improve exercise tolerance and delay fatigue.
This work was supported by United States–Israel Binational Science Foundation grant 9400140 and the C. E. Botnar Trust.
Address for correspondence: Elliot M. Berry, Department of Human Nutrition and Metabolism, Hebrew University-Hadassah Medical School, P.O. Box 12272, Jerusalem, Israel 91120; E-mail: firstname.lastname@example.org.
1. Avraham, Y., O. Bonne, and E. M. Berry. Behavioural and neurochemical alterations caused by diet restriction: the effect of tyrosine
administration in mice. Brain Res. 732: 133–144, 1996.
2. Banderet, L. E., and H. R. Lieberman. Treatment with tyrosine
, a neurotransmitter precursor, reduces environmental stress in humans. Brain Res. Bull. 22: 752–762, 1989.
3. Blomstrand, E., D. Perret, M. Parry-Billings, and E. A. Newshlome. Effect of sustained exercise on plasma amino acid concentrations and on 5-hydroxytryptamine metabolism in six different brain regions in the rat. Acta. Physiol. Scand. 136: 473–481, 1989.
4. Chaouloff, F., D. Laude, D. Merino, B. Serrurrier, Y. Guezennec, and J. L. Elghozi. Amphetamine and alpha-methyl-p-tyrosine
affect the exercise induced imbalance between the availability of tryptophan and synthesis of 5-HT in the brain of the rat. Neuropharmacology 26: 1099–1106, 1987.
5. Davis, J. M., and S. P. Bailey. Possible mechanisms of central nervous system fatigue during exercise. Med. Sci. Sports Exerc. 29: 45–57, 1997.
6. Garner, D. M., and L. W. Rosen. Eating disorders
among athletes: research and recommendations. J. Appl. Sport Sci. Res. 5: 100–107, 1991.
7. Gordon, R., S. Spector, A. Sjoerdsma, and S. Undenfriend. Increased synthesis of norepinephrine and epinephrine in the intact rat during exercise and exposure to cold. J. Pharmacol. Exp. Ther. 153: 440–447, 1966.
8. Gurden, H., J. P. Tassin, and T. M. Jay. Integrity of the mesocortical dopaminergic system is necessary for complete expression of in vivo hippocampal-prefrontal cortex long term potentiation. Neuroscience 94: 1019–1027, 1999.
9. Hassmen, P., E. Blomstrand, B. Ekblom, and E. A. Newshlome. Branched-chain amino acid supplementation during 30km competitive run: mood and cognitive performance. Nutrition 10: 405–410, 1994.
10. Hery, F., P. Soubrie, S. Bourgoin, J. L. Montastruc, F. Artaud, and J. Glowinski. Dopamine released from dendrites in the substantia nigra controls the nigral and striatal release of 5-HT. Brain Res. 193: 143–151, 1980.
11. Heyes, M. P., E. S. Garnett, and G. Coates. Central dopaminergic activity influences rats ability to exercise. Life Sci. 36: 671–677, 1985.
12. Hoebel, B. G. Brain neurotransmitters in food and drug reward. J. Clin. Nutr. 42: 1133–1150, 1985.
13. Kanarek, R. B., and G. H. Collier. Self starvation: a problem of over-riding the satiety signal? Physiol. Behav. 30: 307–311, 1983.
14. Olton, D. S., and R. J. Samuelson. Remembrances of places passed: spatial memory in rats. J. Exp. Psychol. Anim. Behav. Proc. 297: 97–116, 1976.
15. Knott, P. J. and G. Curzon. Free tryptophan in plasma and brain tryptophan metabolism. Nature 239: 452–453, 1972.
16. Leibowitz, S. F. Neurochemical systems of the hypothalamus: control of feeding and drinking behaviour and water-electrolyte excretion. In:Handbook of the Hypothalamus.
Vol. 3A. J. P., Morgane and J. Panksepp (Eds.). New York: Marcel Dekker, 1980, pp. 299–437.
17. Leibowitz, S. F., and J. T. Alexander. Hypothalamic 5-HT in control of eating behaviour: meal size, and body weight. Biol. Psychol. 44: 851–864, 1998.
18. Marrazzi, M. A., and E. D. Luby. An auto-addiction opioid model of chronic anorexia nervosa. Int. J. Eat. Disord. 5: 191–208, 1986.
19. Morris, B. J., and H. M. Johnston. A role of hippocampal opioids in long term functional plasticity. Trends Neurosci. 18: 350–355, 1995.
20. Newshlome, E. A., and E. Blomstrand. The plasma level of some amino acids and physical and mental fatigue. Experientia 52: 413–415, 1996.
21. Pierce, D. W., and W. F. Epling. Activity anorexia. In:Neuromethods, 18, Animal Models in Psychiatry, Vol. 1.
A. A. Boulton, G. B. Baker, and M. T.-M. Iverson. (Eds.). Totowa, NJ: Humana Press, 1991, pp. 267–311.
22. Pirke, K. M., A. Brooks, T. Wilckens, R. Marquard, and U. Schweiger. Starvation induced hyperactivity in the rat: the role of endocrine and neurotransmitter changes. Neurosci. Biobehav. Rev. 17: 287–294, 1993.
23. Routtenberg, A. “Self-starvation” of rats living on activity wheels: adaptation effects. J. Comp. Physiol. Psychol. 66: 234–238, 1968.
24. Shoam, S., E. L. Marcus, Y. Avraham, and E. M. Berry. The effect of diet restriction on the levels of enkephalin and dynorphin in the brain. Nutr. Neurosci. 3: 41–55, 2000.
25. Shurtleff, D., J. R. Thomas, J. Schrot, K. Kawalski, and R. Harford. Tyrosine
reverses cold induced working memory deficit in humans. Pharmacol. Biochem. Behav. 47: 935–941, 1994.
26. van Leeuwen, S. D., O. B. Bonne, Y. Avraham, and E. M. Berry. Separation as a new animal model for self-induced weight loss. Physiol. Behav. 62: 77–81, 1997.
27. Wagenmakers, A. J. Amino acid supplements to improve athletic performance. Curr. Opin. Clin. Nutr. Metab. Care 2: 539–544, 1999.
28. Westerink, B. H. C., and E. Wirix. On the significance of tyrosine
for the synthesis and catabolism of dopamine in rat brain: evaluation by HPLC with electrochemical detection. J. Neurochem. 40: 758–764, 1983.
29. Wurtman, R. J., and J. D. Fernstrom. Control of brain monoamine synthesis by diet and plasma amino acids. Am. J. Clin. Nutr. 28: 638–647, 1975.
30. Nagatsu, T., M. Levitt, and S. Udenfriend. Tyrosine
hydroxylase, the initial step in norepinephrine biosynthesis. J. Biol. Chem. 239: 2910–2917, 1964.
31. Simansky, K. J. Serotonergic control of the organization of feeding and satiety. Behav. Brain Res. 73: 37–42, 1996.