Seiichiro Amemiya

Effects of acute treadmill running at different intensities on activities of serotonin and corticotropin-releasing factor neurons, and anxiety- and depressive-like behaviors in rats

Accumulating evidence suggests that physical exercise can reduce and prevent the incidence of stress-related psychiatric disorders, including depression and anxiety. Activation of serotonin (5-HT) neurons in the dorsal raphe nucleus (DRN) is implicated in antidepressant/anxiolytic properties. In addition, the incidence and symptoms of these disorders may involve dysregulation of the hypothalamic-pituitary-adrenal axis that is initiated by corticotropin-releasing factor (CRF) neurons in the hypothalamic paraventricular nucleus (PVN). Thus, it is possible that physical exercise produces its antidepressant/anxiolytic effects by affecting these neuronal activities. However, the effects of acute physical exercise at different intensities on these neuronal activation and behavioral changes are still unclear. Here, we examined the activities of 5-HT neurons in the DRN and CRF neurons in the PVN during 30 min of treadmill running at different speeds (high speed, 25 m/min; low speed, 15m/min; control, only sitting on the treadmill) in male Wistar rats, using c-Fos/5-HT or CRF immunohistochemistry. We also performed the elevated plus maze test and the forced swim test to assess anxiety- and depressive-like behaviors, respectively. Acute treadmill running at low speed, but not high speed, significantly increased c-Fos expression in 5-HT neurons in the DRN compared to the control, whereas high-speed running significantly enhanced c-Fos expression in CRF neurons in the PVN compared with the control and low-speed running. Furthermore, low-speed running resulted in decreased anxiety- and depressive-like behaviors compared with high-speed running. These results suggest that acute physical exercise with mild and low stress can efficiently induce optimal neuronal activation that is involved in the antidepressant/anxiolytic effects.

Corticotropin-releasing factor antagonist reduces activation of noradrenalin and serotonin neurons in the locus coeruleus and dorsal raphe in the arousal response accompanied by yawning behavior in rats.

We previously reported that intracerebroventricular (icv) administration of corticotropin-releasing factor (CRF) antagonist attenuates the arousal response during yawning behavior in rats. However, the CRF-related pathway involved in the arousal response during yawning is still unclear. In the present study, we assessed the involvement of the CRF-containing pathway from the hypothalamic paraventricular nucleus (PVN) to the locus coeruleus (LC) and the dorsal raphe nucleus (DRN) in the arousal response during frequent spontaneous yawning, which was induced by several microinjections of l-glutamate into the PVN in anesthetized rats, using c-Fos immunohistochemistry. The PVN stimulation showed significant increases in activation of PVN CRF neurons, LC noradrenalin (NA) neurons and DRN serotonin (5-HT) neurons as well as arousal response during yawning. But icv administration of a CRF receptor antagonist, α-helical CRF (9-41), significantly inhibited the activation of both LC NA neurons and DRN 5-HT neurons except the activation of CRF neurons in the PVN, and significantly suppressed the arousal response during yawning. These results suggest that the CRF-containing pathway from PVN CRF neurons to LC NA neurons and DRN 5-HT neurons can be involved in the arousal response during yawning behavior.

Differential effects of background noise of various intensities on neuronal activation associated with arousal and stress response in a maze task.

Background noise (BGN) can affect performance of various tasks as a function of its intensity. Such effects may involve modulation of arousal level during task performance, though the neural mechanisms responsible for the intensity-dependence of effects of BGN are still unclear in detail. We examined the effects of BGN (white noise) of various intensities (control, <40 dB without BGN; 70 dB; 100 dB) during maze task on neuronal activity related to arousal and stress responses using c-Fos immunohistochemistry in rats. Performance (number of errors, time to goal, and number of rearings) during the maze task under 70 dB-BGN, but not 100 dB-BGN, was improved compared with the control condition. In addition, 70 dB-BGN increased c-Fos expression in brain areas responsible for arousal, including mesopontine tegmentum, basal forebrain (BF), locus coeruleus (LC), and cortex, whereas 100 dB-BGN markedly activated neurons in stress-related nuclei, such as the hypothalamic paraventricular nucleus, central nucleus and basolateral nucleus of the amygdala, as well as BF cholinergic neurons, LC neurons, and cortex. These findings suggest that BGN during maze task can induce differential neuronal activation depending on the intensity of BGN in the brain areas relating to arousal and stress responses, which might be involved in maze performance.

Emotional stress evoked by classical fear conditioning induces yawning behavior in rats.

Yawning is often observed not only in a state of boredom or drowsiness but also in stressful emotional situations, suggesting that yawning is an emotional behavior. However, the neural mechanisms for yawning during stressful emotional situations have not been fully determined, though previous studies have suggested that both parvocellular oxytocin (OT) and corticotropin-releasing factor (CRF) neurons in the hypothalamic paraventricular nucleus (PVN) are responsible for induction of yawning. Thus, using ethological observations and c-Fos immunohistochemistry, we examined whether emotional stress evoked by classical fear conditioning is involved in induction of yawning behavior in freely moving rats. Emotional stress induced yawning behavior that was accompanied by anxiety-related behavior, and caused neuronal activation of the central nucleus of the amygdala (CeA), as well as increases in activity of both OT and CRF neurons in the PVN. These results suggest that emotional stress may induce yawning behavior, in which the neuronal activation of the CeA may have a key role.

Effects of negative air ions on activity of neural substrates involved in autonomic regulation in rats

The neural mechanism by which negative air ions (NAI) mediate the regulation of autonomic nervous system activity is still unknown. We examined the effects of NAI on physiological responses, such as blood pressure (BP), heart rate (HR), and heart rate variability (HRV) as well as neuronal activity, in the paraventricular nucleus of the hypothalamus (PVN), locus coeruleus (LC), nucleus ambiguus (NA), and nucleus of the solitary tract (NTS) with c-Fos immunohistochemistry in anesthetized, spontaneously breathing rats. In addition, we performed cervical vagotomy to reveal the afferent pathway involved in mediating the effects of NAI on autonomic regulation. NAI significantly decreased BP and HR, and increased HF power of the HRV spectrum. Significant decreases in c-Fos positive nuclei in the PVN and LC, and enhancement of c-Fos expression in the NA and NTS were induced by NAI. After vagotomy, these physiological and neuronal responses to NAI were not observed. These findings suggest that NAI can modulate autonomic regulation through inhibition of neuronal activity in PVN and LC as well as activation of NA neurons, and that these effects of NAI might be mediated via the vagus nerves.

Negative rebound in hippocampal neurogenesis following exercise cessation.

Physical exercise can improve brain function, but the effects of exercise cessation are largely unknown. This study examined the time-course profile of hippocampal neurogenesis following exercise cessation. Male C57BL/6 mice were randomly assigned to either a control (Con) or an exercise cessation (ExC) group. Mice in the ExC group were reared in a cage with a running wheel for 8 wk and subsequently placed in a standard cage to cease the exercise. Exercise resulted in a significant increase in the density of doublecortin (DCX)-positive immature neurons in the dentate gyrus (at week 0). Following exercise cessation, the density of DCX-positive neurons gradually decreased and was significantly lower than that in the Con group at 5 and 8 wk after cessation, indicating that exercise cessation leads to a negative rebound in hippocampal neurogenesis. Immunohistochemistry analysis suggests that the negative rebound in neurogenesis is caused by diminished cell survival, not by suppression of cell proliferation and neural maturation. Neither elevated expression of ΔFosB, a transcription factor involved in neurogenesis regulation, nor increased plasma corticosterone, were involved in the negative neurogenesis rebound. Importantly, exercise cessation suppressed ambulatory activity, and a significant correlation between change in activity and DCX-positive neuron density suggested that the decrease in activity is involved in neurogenesis impairment. Forced treadmill running following exercise cessation failed to prevent the negative neurogenesis rebound. This study indicates that cessation of exercise or a decrease in physical activity is associated with an increased risk for impaired hippocampal function, which might increase vulnerability to stress-induced mood disorders.

Noradrenergic signaling in the medial prefrontal cortex and amygdala differentially regulates vicarious trial-and-error in a spatial decision-making task.

In uncertain choice situations, we deliberately search and evaluate possible options before taking an action. Once we form a preference regarding the current situation, we take an action more automatically and with less deliberation. In rats, the deliberation process can be seen in vicarious trial-and-error behavior (VTE), which is a head-orienting behavior toward options at a choice point. Recent neurophysiological findings suggest that VTE reflects the rats thinking about future options as deliberation, expectation, and planning when rats feel conflict. VTE occurs depending on the demand: an increase occurs during initial learning, and a decrease occurs with progression in learning. However, the brain circuit underlying the regulation of VTE has not been thoroughly examined. In situations in which VTE often appears, the medial prefrontal cortex (mPFC) and the amygdala (AMY) are crucial for learning and decision making. Our previous study reported that noradrenaline regulates VTE. Here, to investigate whether the mPFC and AMY are involved in regulation of VTE, we examined the effects of local injection of clonidine, an alpha2 adrenergic autoreceptor agonist, into either region in rats during VTE and choice behavior during a T-maze choice task. Injection of clonidine into either region impaired selection of the advantageous choice in the task. Furthermore, clonidine injection into the mPFC suppressed occurrence of VTE in the early phase of the task, whereas injection into the AMY inhibited the decrease in VTE in the later phase and thus maintained a high level of VTE throughout the task. These results suggest that the mPFC and AMY play a role in the increase and decrease in VTE, respectively, and that noradrenergic mechanisms mediate the dynamic regulation of VTE over experiences.

Adaptive Changes in the Sensitivity of the Dorsal Raphe and Hypothalamic Paraventricular Nuclei to Acute Exercise, and Hippocampal Neurogenesis May Contribute to the Antidepressant Effect of Regular Treadmill Running in Rats

Increasing clinical evidence suggests that regular physical exercise can prevent or reduce the incidence of stress-related psychiatric disorders including depressive symptoms. Antidepressant effect of regular exercise may be implicated in monoaminergic transmission including serotonergic transmission, activation of the hypothalamic-pituitary-adrenal (HPA) axis, and hippocampal neurogenesis, but few general concepts regarding the optimal exercise regimen for stimulating neural mechanisms involved in antidepressant properties have been developed. Here, we examined how 4 weeks of treadmill running at different intensities (0, 15, 25 m/min, 60 min/day, 5 times/week) alters neuronal activity in the dorsal raphe nucleus (DRN), which is the major source of serotonin (5-HT) neurons in the central nervous system, and the hypothalamic paraventricular nucleus (PVN), in which corticotropin-releasing factor (CRF) neurons initiate the activation of the HPA axis, during one session of acute treadmill running at different speeds (0, 15, 25 m/min, 30 min) in male Wistar rats, using c-Fos immunohistochemistry. We also examined neurogenesis in the hippocampus using immunohistochemistry for doublecortin (DCX) and assessed depressive-like behavior using the forced swim test after regular exercise for 4 weeks. In the pre-training period, acute treadmill running at low speed, but not at high speed, increased c-Fos positive nuclei in the DRN compared with the sedentary control. The number of c-Fos positive nuclei in the PVN during acute treadmill running was increased in a running speed-dependent manner. Regular exercise for 4 weeks, regardless of the training intensity, induced an enhancement of c-Fos expression in the DRN during not only low-speed but also high-speed acute running, and generally reduced c-Fos expression in the PVN during acute running compared with pre-training. Furthermore, regular treadmill running for 4 weeks enhanced DCX immunoreactivity in the hippocampal dentate gyrus (DG), and resulted in decreased depressive-like behavior, regardless of the training intensity. These results suggest that long-term repeated exercise, regardless of the training intensity, improves depressive-like behavior through adaptive changes in the sensitivity of DRN and PVN neurons to acute exercise, and hippocampal neurogenesis.

Noradrenergic neurons are activated depending on the difficulty of decision-making in rat

Neuronal activity in the dorsal raphé nucleus (DRN), a major source of serotonin, is modulated by the received reward size. To investigate whether DRN neurons code rewarding or aversive stimuli and/or positive or negative prediction error, we recorded single-unit activity in the DRN of two monkeys performing the trace conditioning task. This task consisted of two blocks with distinct contexts. In the appetitive block, liquid reward was used as an unconditioned stimulus (US). In the aversive block, air-puff directed at the monkey face was used as a negative US. In both blocks, three visual stimuli (conditioned stimuli: CSs) were paired with the US, with probabilities of 100, 50 and 0%, respectively. To confirm that monkeys learned the association of each specific CS with the US, we monitored licking behavior and anticipatory eye blinking. In 50 and 0% trials, tone was presented as a neutral stimulus in the absence of reward or air-puff. We recorded 211 task-related neurons. It was found that DRN neurons responded to the CSs in the appetitive block more often than in the aversive block; 38% (n = 81) responded to the rewarding CS with 100% probability, while 9% (n = 19) responded to the aversive CS with 100% probability. Among them, 13 neurons responded to both rewarding and aversive CSs. Many DRN neurons also responded to the USs (n = 176); either to reward only (n = 35), to air-puff only (n = 57) or to both (n = 84). In the appetitive block, rewardrelated activity was modulated by its probability with stronger response to the unpredicted than the predicted US. In the aversive block, short-latency response to air-puff delivery was frequently observed regardless of the CS-US predictability. These results suggest that the primate DRN codes information about both rewarding and aversive stimuli, and some DRN neurons exhibited activity similar to reward prediction error reported for dopamine neurons.

Involvements of amygdala on yawning responses induced by fear conditioning in rats

Emotional disorders are quite often accompanied with various psychiatric diseases and stress disorders, and it is an urgent problem to overcome these disorders for many peoples. Mice and rats exhibit profound emotional disorders such as elevated anxiety after having a status epilepticus (SE) by pilocarpine injection, however, its precise mechanism is not yet clarified. In order to know the pathogenic origin of the emotional disorders, we studied the changes in behavior and brain structures of mice after inducing SE with various conditions. In the present study, we used well-established behavior tests for anxiety, including a light and dark preference tests, an open field test, and an elevated plus maze test. Marked facilitation of anxiety-like behavior were observed in the mice which induced SE for 4.5 h as early as 4-6 days after SE. This emotional change was very stable and still observed 8 months after SE. It is known that SE induced mossy fiber sprouting of granule cells in dentate gyrus as well as a loss of parvalbumin-positive inter neurons in the cerebral cortex and hippocampus. Since mossy fiber sprouting was not observed on 6 days after SE by immunohistochemistry with antineuropeptide Y antibody, the sprouting was unlikely to be a pathogenic change for the emotional disorder. The mice in which SE was terminated at 1 hour by a phenobarbital administration also exhibited marked facilitation of anxiety-like behaviors 2 days after SE, however, this change was reversible and the mice showed a normal behavior 6 days after SE. These results suggested that there were at least two different mechanisms for the appearance of emotional disorder after pilocarpine-induced SE, one was reversible and another was irreversible, and the duration of SE was critical for the selection of mechanisms.