If serotonin does not exhaust you, it makes you stronger

Serotonin (5‐HT), a major modulator of motor activity, is synthesized and released by neurons from the raphe nuclei in the brainstem. These neurons project to most regions of the central nervous system including on the somatodendritic compartments of motoneurons. Contrasting physiological roles have been ascribed to serotonergic modulation of motoneurons. According to scientists interested in motor fatigue, serotonin inhibits motor activity by a phenomenon known as central fatigue, i.e. the inability of the central nervous system to maintain a given motor output during a sustained effort. By contrast, researchers studying the intrinsic properties of motoneurons are convinced that serotonin helps movement by increasing the excitability of motoneurons. Before deciding what is right and wrong, let us briefly review the evidence for the arguments.

Serotonin is synthesized from the amino acid tryptophan, which crosses the blood–brain barrier via the same transporter as branched‐chained amino acids. The activity of raphe spinal neurons follows the level of motor activity, which suggests that the release of serotonin is higher during intensive or prolonged physical efforts, and indeed, the amount of serotonin in the brain increases during fatiguing exercise. In agreement, reduced tryptophan intake improves long distance running in humans. Conversely, tryptophan‐infused horses are exhausted faster than control animals. These data, which suggest a link of causality between serotonin and fatigue, are supported by the fact that cyclists receiving buspirone – an agonist for 5‐HT_1A receptors – or a selective serotonin reuptake inhibitor (SSRI) are exhausted faster than control groups (reviewed by Meeusen et al. 2006).

On the other hand, experiments performed in vivo in decerebrated cats and in slice preparations from the spinal cord of adult turtles have shown that serotonin promotes the excitability of motoneurons by activating a 5‐HT₂ receptor, which facilitates a persistent inward current mediated by L‐type Ca²⁺ channels and Na⁺ channels (Hounsgaard et al. 1988; Perrier & Delgado‐Lezama, 2005).

Most studies demonstrating the induction of fatigue or the increase in motoneuron excitability are supported by convincing data. The apparent opposite effects of serotonin remained a mystery until experiments performed in slices from the turtle spinal cord provided an explanation for the serotonin paradox (Cotel et al. 2013). Unlike most in vitro studies, the authors investigated the effects of serotonin released from synapses. In agreement with the ‘motoneuron intrinsic property community’, they found that during moderate activity of the raphe spinal pathway, the release of serotonin promotes the excitability of motoneurons by activating 5‐HT₂ receptors positively coupled to L‐type Ca²⁺ ion channels (Perrier & Delgado‐Lezama, 2005). However, after prolonged raphe spinal activity, serotonin spills out from the synapses located on the somatodendritic compartments of motoneurons. By diffusion, serotonin reaches the axon initial segment of motoneurons, which is devoid of serotonergic synaptic boutons but nevertheless expresses 5‐HT_1A receptors. As a result, Na⁺ ion channels responsible for action potential genesis are inhibited and the gain of motoneurons reduced (Cotel et al. 2013). This study showed that depending on the level of motor activity, serotonin acts as an accelerator or a brake on motoneurons and reconciled both views on serotonin. However, do these findings in slices from reptiles, also apply in vivo in mammals? In this issue of the Journal of Physiology, Kavanagh et al. (2019) tested if serotonin also has a dual effect on human motoneurons of healthy subjects. To increase the level of serotonin in the central nervous system, they used the SSRI paroxetine. Since the release of serotonin increases with the level of motor activity, they assumed that paroxetine induced a moderate rise after brief motor activity and a stronger increase following intense effort. The SSRI intake resulted in a stronger force produced during a single brief maximal voluntary contraction. However, during repeated maximal voluntary contractions, the force produced and the time to task failure were reduced compared to when the same individuals ingested a placebo. To figure out if the fatigue observed after subjects took paroxetine had a central or a peripheral origin, the authors used the superimposed twitch method. This technique consists of measuring the extra force produced by an electric chock applied on the motor nerve during maximal voluntary contractions. The amplitude of the superimposed twitch reflects the inability of the central nervous system to activate the muscle. Before fatiguing contractions, the superimposed twitch was smaller for subjects who took paroxetine. However, it became bigger after repeated maximal voluntary contractions. Since paroxetine did not change the contractile properties of the muscle, the authors concluded that the additional fatigue occurring after SSRI intake was purely central. Finally, they tested if the increase in central fatigue by serotonin occurred at the level of motoneurons. To estimate the variations in motoneuron excitability, they measured the area and the persistence of F‐waves evoked by antidromic activation of motoneurons. After a prolonged maximal voluntary contraction, both parameters were more reduced for subjects who took paroxetine. Altogether, these results, which are in remarkable agreement with the mechanism uncovered in the spinal cord of the turtle, suggest that the dual effect of serotonin on motor control occurs at the level of motoneurons.

Additional information

Competing interests

None declared.

Author contributions

Sole author.

Funding

The author's work is funded by the Owensenske Foundation.