Time-dependent modulation of serotonin and its receptors 1A and 2A expression in allergic contact dermatitis.

BACKGROUND: Serotonin implication in allergic contact dermatitis was earlier suggested. OBJECTIVE: To study the expression of serotonin (5-HT) and its receptors 1A and 2A during the kinetics of allergic contact dermatitis (ACD). METHOD: Biopsies from 10 female nickel-allergic patients were obtained at 0, 6, 24, 48 and 72 h after nickel application using a patch testing procedure and then analysed by immunohistochemistry. RESULTS: Higher 5-HT epidermal immunoreactivity was seen at 0 and 6 h compared with 24, 48 and 72 h. The number of 5-HT labelled platelets was higher at 48 and 72 h compared with 0, 6 and 24 h. There was a decrease in the number of 5-HT1AR positive cells at 24, 48 and 72 h compared with 0 h. Furthermore, epidermal 5-HT1AR immunoreactivity was stronger at 48 and 72 h compared with 0 h. Total (dermis plus epidermis) number of 5-HT2AR positive cells was gradually increased in a time-dependent manner at 6, 24, 48 and 72 h compared with 0 h. CONCLUSIONS: Our results suggest the implication of 5-HT, 5-HT1AR and 5-HT2AR in the development of human ACD and the possibility to target those receptors at an early stage of this inflammatory condition.

Corticosterone regulation of tryptophan hydroxylase in midbrain of the rat.

The tryptophan hydroxylase activity in the rat midbrain decreases after adrenalectomy and is restored by treatment with corticosterone. Cycloheximide, adminiistered intracisternally, prevents the restoration of the enzyme activity by corticosterone. Cycloheximide administration to adrenalectomized rats resutlts in a further decrease in the enzyme activity. an indication that the enzyme has a rapid turnover even in the absence of corticosterone.

In vivo conversion of 3H-L-tryptophan into 3H-serotonin in brain areas of adrenalectomized rats.

Rats were adrenalectomized 10 days before we estimated in vivo the conversion index of (3)H-tryptophan into radioactive serotonin in brainstem and telediencephalon. We found that the conversion index in the brainstem of adrenalectomized rats is smaller than in the same area of sham-operated rats. Conversely, the conversion index in the telediencephalon was similar in the two groups of rats. The serotonin concentrations were unchanged by adrenalectomy, which suggests that in brainstem the decrease of tryptophan hydroxylase is reflected by the conversion index estimation and not by measurement of serotonin steady-state concentrations.

Amnesia produced by electroconvulsive shock or cycloheximide: conditions for recovery.

Retrograde amnesia for a passive avoidance response was produced in rats by electroconvulsive shock and in mice by cycloheximide, an inhibitor of protein synthesis. One day after training the memory could be restored if a "reminder" of the original foot shock was given after the retention test on which the amnesia was demonstrated. Memory did not return if the reminder was given without the prior retention test or if the reminder and the test were separated by 23 hours.

Facilitated sexual behavior reversed and serotonin restored by raphe nuclei transplanted into denervated hypothalamus.

Fetal raphe cells transplanted into the hypothalamus reversed facilitation of feminine sexual behavior in rats with brain lesions induced by 5,7-dihydroxytryptamine. Immunocytochemical and chemical analyses of serotonin indicate that reinnervation of the ventromedial nucleus of the hypothalamus by the transplants is associated with behavioral recovery. The findings suggest that transplanted fetal tissue can exert functional regulation over an innate, complex, hormone-dependent behavior in adult rats.

S100beta protein expression: gender- and age-related daily changes.

S100beta is a soluble protein released by glial cells mainly under the activation of the 5-HT1A receptor. It has been reported as a neuro-trophic and -tropic factor that promotes neurite maturation and outgrowth during development. This protein also plays a role in axonal stability and the plasticity underlying long-term potentiation in adult brains. The ability of S100beta to rapidly regulate neuronal morphology raises the interesting point of whether there are daily rhythm or gender differences in S100beta level in the brain. To answer this question, the S100beta expression in adult female and male rats, as well as in adult female CD-21 and S100beta -/- female mice, were investigated. Scintillation counting and morphometric analysis of the immunoreactivity of S100beta, showed rhythmic daily expression. The female and male rats showed opposite cycles. Females presented the highest value at the beginning of the rest phase (5:00 h), while in males the maximum value appeared in the beginning of the motor activity period (21:00 h). These results confirm previous S100beta evaluations in human serum and cerebrospinal fluid reporting the protein's function as a biomarker for brain damage (Gazzolo et al. in Clin Chem 49:967-970, 2003; Clin Chim Acta 330:131-133, 2003; Pediatr Res 58:1170-1174, 2005), similar behavior was also observed for GFAP in relation to Alzheimer Disease (Fukuyama et al. in Eur Neurol 46:35-38, 2001). The data should be taken into account when considering S100beta as a biomarker of health condition. In addition, the results raise questions on which structure or condition imposes these rhythms as well as on the physiological meaning of the observed gender differences.

Neuroimmune mechanisms in patients with atopic dermatitis during chronic stress.

OBJECTIVE: To identify pathoaetiological neuroimmune mechanisms in patients with atopic dermatitis (AD) and chronic stress, focusing at nerve density, sensory neuropeptides, and the serotonergic system. METHODS: Eleven patients with AD with histories of stress worsening were included. Biopsies from involved and non-involved skin were processed for immunohistochemistry. Salivary cortisol test was done as a marker for chronic stress. RESULTS: There were more acanthosis and fewer nerve fibres in epidermis and papillary dermis of involved compared with non-involved skin. Whereas there was no significant change in the number of substance P and calcitonin gene-related peptide-positive nerve fibres between the involved and non-involved skin, there was an increase in the epidermal fraction of 5-hydroxtrytamine 1A (5-HT1A) receptor and serotonin transporter protein (SERT) immunoreactivity in the involved skin. The number of 5-HT2AR, CD3-positive cells, and SERT-positive cells, most of them being CD3 positive, was increased in involved skin. There was an increase in mast cells in the involved skin, and these cells were often located close to the basement membrane. There was a strong tendency to a correlation between 5-HT2AR positive cells in the papillary dermis of involved skin and low cortisol ratios, being an indicator of chronic stress. CONCLUSION: A changed innervation and modulation of the serotonergic system are indicated in chronic atopic eczema also during chronic stress.

Persistent Angiogenesis in the Autism Brain: An Immunocytochemical Study of Postmortem Cortex, Brainstem and Cerebellum.

In the current work, we conducted an immunocytochemical search for markers of ongoing neurogenesis (e.g. nestin) in auditory cortex from postmortem sections of autism spectrum disorder (ASD) and age-matched control donors. We found nestin labeling in cells of the vascular system, indicating blood vessels plasticity. Evidence of angiogenesis was seen throughout superior temporal cortex (primary auditory cortex), fusiform cortex (face recognition center), pons/midbrain and cerebellum in postmortem brains from ASD patients but not control brains. We found significant increases in both nestin and CD34, which are markers of angiogenesis localized to pericyte cells and endothelial cells, respectively. This labeling profile is indicative of splitting (intussusceptive), rather than sprouting, angiogenesis indicating the blood vessels are in constant flux rather than continually expanding.

Specific in vitro uptake of serotonin by cells in the anterior pituitary of the rat.

In vivo studies have suggested that serotonin (5HT) influences anterior pituitary function at the hypothalamic level. The present in vitro study investigated the possibility that 5HT may act directly on the anterior pituitary. The high affinity uptake of [3H]5HT into adult rat anterior pituitary tissue was examined in two types of experiments. 1) To test the specificity and saturability of uptake of 5HT in the anterior pituitary, pituitary tissue was incubated (37 C) with [3H]5HT (10(-8)-10(-6) M) in the presence and absence of excess (10(-5) M) unlabeled 5HT, norepinephrine, fluoxetine (FLUOX), metergoline, or cyproheptadine. A Hofstee analysis of the specific uptake of [3H]5HT gave an apparent Km value of 4.23 x 10(-7) M and a Vmax of 1576 pmol/g/10 min [3H]5HT. The total uptake of [3H]5HT was not altered by norepinephrine or metergoline, but was significantly reduced (P less than 0.01-0.001) by FLUOX and cyproheptadine. Uptake was shown to be temperature and sodium dependent and not directly dependent on energy derived from glycolysis or aerobic metabolism. 2) To study the site of uptake of 5 HT in the anterior pituitary, in concomitant radioautographic experiments, tissue was incubated with [3H]5HT with and without excess 5HT or FLUOX. Three patterns of silver grain distribution were observed: 1) nonrandom concentrations over select anterior pituitary cells near blood vessels, 2) heavy aggregates of silver grains usually associated with blood vessels, and 3) a seemingly random dispersal of grains over pituitary tissue. Tissue incubated with [3H]5HT alone contained 10% heavily labeled cells, 32% moderately labeled cells, and 58% weakly labeled cells. In contrast, no heavily labeled cells were seen when tissue was incubated with either excess 5HT or FLUOX in addition to [3H]5HT. Our findings of saturable and specific high affinity uptake of [3H]5HT into a subgroup of anterior pituitary cells suggest a direct pituitary action of 5HT.

Use of tissue culture models to study neuronal regulatory trophic and toxic factors in the aged brain.

Dementia is believed to result from the loss of selective neurons within the brain, but approaches for systematic study of that degenerative process are hampered by the complexity of the neuronal milieu. Tissue culture models provide a means to reduce dramatically the variables inherent in the study of neuronal plasticity. Three levels of complexity can be described: cellular and molecular diversity; primary and secondary interconnections; and finally, the dynamics influenced by age. The following review discusses the advantages and disadvantages of tissue culture models for the detailed study of neuronal trophic and toxic factors. Our selection of factors is broadened to include ions, intermediate metabolites, antioxidants, steroids, neuropeptides, gangliosides, metals, neurotransmitters, brain extracts, and protein molecules. Most of these factors have been shown to be altered in the aged brain, to have a significant effect on cultured neurons, or both. This multilevel analysis provides the reader with an overview of the events regulating neuronal survival, differentiation and death. An understanding of these basic questions is necessary to sequence the molecular events resulting in neuronal death.

Bilateral serotonergic projections to the dorsal hippocampus of the rat: simultaneous localization of 3H-5HT and HRP after retrograde transport.

Horseradish peroxidase (HRP, Sigma VI, 30-70 nl of a 10-15% solution in saline) or 3H-5HT (30 Ci/mmole, 2.5 X 10 -3 M containing 3.3 X 10(-3) M norepinephrine in saline, 50-100 nl) was injected unilaterally into the dorsal hippocampus in separate groups of rats. HRP-labeled cells were seen in the hippocampus, medial septal nucleus, nucleus of the diagonal band, supramammilary nucleus, median raphe nucleus, interfascicular portion of the dorsal raphe nucleus, and the locus coeruleus. In contrast, 3H-5HT-labeled cells were largely restricted to the raphe nuclei. In this nucleus an equal number of ipsilateral and bilateral cells were found. Occasionally, these labeled cells stretched across the midline (bridge pattern). In another series, the 3H-5HT and HRP were injected into the same hippocampus either as a mixture or sequentially. This resulted in double labeling of the median and dorsal raphe neurons. A final group of rats received injections of 3H-5HT and HRP into opposing hippocampi. Double-labeled cells accounted for 10% of the neurons labeled. In addition, closely paired neurons composed of an HRP- and 3H-5HT-containing cell were found. In summary, the serotonergic fibers may play a key role in harmonizing the electrical activity of the hippocampi by use of bilateral projections, paired neurons with differential projections, and bridging neurons stretching across the midline but with unilateral projections.

5-HT1A receptor localization on the axon hillock of cervical spinal motoneurons in primates.

Serotonin (5-HT) has direct and specific effects on the activity of spinal cord motoneurons. The 5-HT1A receptor has been shown to mediate motoneuron responses in spinal reflex pathways using the highly selective 5-HT1A receptor agonist 8-OH-DPAT. We have developed an antipeptide antibody that recognizes a specific region (the second external loop) of the 5-HT1A receptor. This 5-HT1A receptor antibody labels populations of neurons and glia in the primate cervical spinal cord. The highest receptor density is present in the superficial lamina of the dorsal horn, around the central canal, and on the axon hillock of large ventral horn motoneurons. The cellular labeling pattern on motoneurons shows a single, densely stained, tapering process emanating from the perikaryon. A more diffuse label is also present throughout the soma. Dendritic labeling was not apparent. These results suggest that post-synaptic 5-HT1A receptors may be involved in modulating spinal motoneuron activity at the key site of action potential initiation, the axon hillock.

An immunocytochemical study of the serotonergic innervation of the thalamus of the rat.

The serotonergic innervation of the rat thalamus was studied with an indirect immunocytochemical technique using an antiserum raised against a serotonin-hemocyanin conjugate in animals pretreated with L-tryptophan and a monoamine oxidase (MAO) inhibitor. When pretreatment was not used there was a decrease in the number of immunoreactive fibers observed in the thalamic region. Innervation was greatest in nucleus ventralis corporis geniculati lateralis, and in the following midline nuclei: nucleus periventricularis, nucleus rhomboideus, and nucleus reuniens. Also well labeled were nucleus anterior ventralis, the intralaminar nuclei, and nucleus lateralis dorsalis. Moderate innervation was found in nucleus reticularis, nucleus anterior dorsalis, nucleus ventralis medialis, nucleus lateralis pars posterior, the posterior complex, and in nucleus dorsalis corporis geniculati lateralis. Very few serotonergic fibers were observed in nucleus paratenialis, nucleus gelatinosus, nucleus anterior medialis, nucleus medialis dorsalis, nucleus ventralis, nucleus ventralis pars dorsomedialis, or in nucleus corporis geniculati medialis. Serotonin immunoreactivity was also noted in a number of fiber bundles in the thalamic region. These include the fasciculus retroflexus, the fasciculus mamillothalamicus, the stria medullaris, and the stria terminalis. These results differ from those of previous descriptions of the serotonergic innervation of thalamic nuclei most notably in the midline nuclei and in the posterior complex. In this study the midline nuclei, nucleus rhomboideus, and nucleus reuniens were more densely innervated than had been described, and in the posterior complex a moderate, rather than sparse, innervation was observed. The more densely innervated nuclei of the anterior and lateral nuclear groups, nucleus lateralis dorsalis and nucleus anterior ventralis, also contained a greater number of labeled fibers than had been indicated.

Pyramidal cell axons show a local specialization for GABA and 5-HT inputs in monkey and human cerebral cortex.

Various mechanisms are thought to control excitation of pyramidal cells of the cerebral cortex. With immunocytochemical methods, we found that the proximal portions of numerous pyramidal cell axons (Pyr-axons) in the human and monkey neocortex are immunoreactive for the serotonin (5-HT) receptor 5-HT-(1A). With double-labeling experiments and confocal laser microscopy, we found that most (93.4%) of the 5-HT(1A)-immunoreactive Pyr-axons present in layers II and III were innervated by parvalbumin-immunoreactive chandelier cell axon terminals. In addition, Pyr-axons were compartmentalized: 5-HT-(1A) receptors were found proximal to inputs from chandelier cells. Although we found close appositions between GABAergic chandelier cell axon terminals and Pyr-axons, suggesting synaptic connections, we did not observe 5-HT-immunoreactive fibers in close proximity to the Pyr-axons. These results suggested that Pyr-axons are under the influence of 5-HT in a paracrine manner (via 5-HT-(1A) receptors) and, more distally, are under the influence of gamma-aminobutyric acid (GABA) in a synaptic manner (through the axons of chandelier cells). The local axonal specialization might represent a powerful inhibitory mechanism by which the responses of large populations of pyramidal cells can be globally controlled by subcortical serotonin afferents, in addition to local inputs from GABAergic interneurons.

Serotonin neurons, neuroplasticity, and homeostasis of neural tissue.

Homeostasis is the process by which the internal milieu of the body is able to maintain equilibrium in the face of constant insults from the external world. Endocrine, immune, and vascular systems play pivotal roles in adjusting internal biochemical reactions to counteract assaults from the outside. Despite the vast accumulation of data over the last 50 years, a role for serotonin in brain homeostasis has not been proposed. In this chapter I will review the plasticity and anatomy of serotonergic neurons in integrating external sensory and motor systems as well as internal endocrine, glial and vascular signals with the various cellular elements comprising neural tissue. Steroids and neuropeptides have both been shown to alter the morphology of serotonergic neurons. In turn, alterations in serotonin levels in the adult brain can change the morphology of its target cells. A pivotal role for serotonin in the homeostasis of neural tissue is consistent with the function of serotonin throughout evolution and explains the large number of biological systems, behavioral activities, and clinical diseases associated with serotonergic neurons.

MDMA (ecstasy) inhibition of MAO type A and type B: comparisons with fenfluramine and fluoxetine (Prozac).

3,4-Methylenedioxymethamphetamine (MDMA), a serotonin (5-HT) neurotoxin, has been shown to promote the release of serotonin (5-HT) and block its reuptake. The increased buildup of extracellular 5-HT should normally be degraded by monoamine oxidase (MAO). The effects of both enantiomers of MDMA were examined on MAO-A and monoamine oxidase-B (MAO-B) activity in rat brain homogenates. Both enantiomers competitively inhibited 5-HT catabolism by rat brain MAO-A. The Ki of MDMA for MAO-A was 22 mumol/L. A mixed type of inhibition by MDMA was observed for phenethylamine catabolism by MAO-B for both optical antipodes. Logistical analysis of concentration response curves for MDMA inhibition of MAO-A and MAO-B show an IC50 of 44 mumol/L for inhibition of MAO-A by MDMA. The IC50 value of MDMA inhibition of MAO-B was 370 mumol/L, showing a selective potency for MAO-A inhibition. The MAO inhibitory properties of fenfluramine (FEN) and fluoxetine (FLUOX) were compared to those of MDMA. The rank order potency of these drugs for MAO-A inhibition was MDMA > FLUOX > FEN, whereas for MAO-B inhibition, FLUOX > MDMA > FEN. A combination of FLUOX and MDMA at their respective IC50 did not inhibit MAO activity more than either drug alone at equivalent concentrations. These results indicate that the actions of FEN do not appear to involve MAO inhibition. MDMA (ecstasy) produced a preferential inhibition of MAO-A (IC50 = 44 mumol/L), which should increase extracellular 5-HT.(ABSTRACT TRUNCATED AT 250 WORDS)

Activation of protein kinase C (PKC) by 3,4-methylenedioxymethamphetamine (MDMA) occurs through the stimulation of serotonin receptors and transporter.

This report further characterizes the intermediate metabolic effects of the psychotropic amphetamine derivative, 3,4-methylenedioxymethamphetamine (MDMA or "ecstasy"), on the activity of second messenger-dependent kinases. Previous work has demonstrated that two injections of MDMA (20 mg/kg) elicits a prolonged translocation of the calcium and phospholipid-dependent enzyme, protein kinase C (PKC) in rats. However, because MDMA has actions at the 5-HT transporter and 5-HT2A/2C receptors, our experiments were directed at uncovering which of these many sites may be involved in this second messenger dependent response. A single injection of MDMA produced a time- and dose-dependent increase in the density of cortical and hippocampal PKC (as measured by 3H-phorbol 12,13-dibutyrate (PDBu) binding sites. MDMA-mediated PKC translocation was long-lasting and remained above control (saline-treated rats) for up to 24 h after injection. This effect was mimicked by another substituted amphetamine, p-chloroamphetamine (pCA), but with a temporal-response curve that was to the left of MDMA's. However, pure uptake inhibitors like fluoxetine, cocaine, and the selective 5-HT2A/2C agonist, DOB, were unable to produce a long-lasting translocation of PKC binding sites in rat cortex. Fluoxetine, a selective serotonin uptake inhibitor (SSRI) and ketanserin a 5-HT2A antagonist, attenuated PKC translocation by MDMA with differing efficacies; however, both compounds completely prevented the loss of 5-HT uptake sties after multiple doses of MDMA. These results suggest that MDMA increases PKC translocation by two interrelated mechanisms that involve 5-HT2A/2C receptors and the 5-HT transporter. This pathway appears to include: (1) the drug binding to the 5-HT transporter, (2) the release of cytosolic 5-HT stores into the extracellular space, and (3) the activation of post-synaptic 5-HT2A/2C receptors linked to G-protein-mediated phospholipid hydrolysis.

Characterization of the translocation of protein kinase C (PKC) by 3,4-methylenedioxymethamphetamine (MDMA/ecstasy) in synaptosomes: evidence for a presynaptic localization involving the serotonin transporter (SERT).

3, 4-methylenedioxymethamphetamine (MDMA or Ecstasy) is a substituted amphetamine whose acute and long-term effects on the serotonin system are dependent on an interaction with the 5-HT uptake transporter (SERT). Although much of the work dedicated to the study of this compound has focused on its ability to release monoamines, this drug has many important metabolic consequences on neurons and glial cells. The identification of these physiological responses will help to bridge the gap that exists in the information between the acute and neurotoxic effects of amphetamines. Substituted amphetamines have the ability to produce a long-term translocation of protein kinase C (PKC) in vivo, and this action may be crucial to the development of serotonergic neurotoxicity. Our earlier results suggested that PKC activation occurred through pre- and postsynaptic mechanisms. Because the primary site of action of these drugs is the 5-HT transporter, we now expand on our previous results and attempt to characterize MDMA's ability to translocate PKC within cortical 5-HT nerve terminals. In synaptosomes, MDMA produced a concentration-dependent increase in membrane-bound PKC (as measured by 3H-phorbol 12, 13 dibutyrate, 3H-PDBu) bindings sites. This response was abolished by cotreatment with the specific serotonin reuptake inhibitor (SSRI), fluoxetine, but not by the 5-HT2A/2C antagonist, ketanserin. In contrast, full agonists to 5-HT1A and 5-HT2 receptors did not produce significant PKC translocation. MDMA-mediated PKC translocation also requires the presence of extracellular calcium ions. Using assay conditions where extracellular calcium was absent prevented in vitro activation of PKC by MDMA. Prolonged PKC translocation has been hypothesized to contribute to the calcium-dependent neurotoxicity produced by substituted amphetamines. In addition, many physiological processes within 5-HT nerve terminals, including 5-HT reuptake and vesicular serotonin release, are susceptible to modification by PKC-dependent protein phosphorylation. Our results suggest that prolonged activation of PKC within the 5-HT nerve terminal may contribute to lasting changes in the homeostatic function of 5-HT neurons, leading to the degeneration of specific cellular elements after repeated MDMA exposure.

Cellular localization of the 5-HT1A receptor in primate brain neurons and glial cells.

Activation of 5HT1A receptors produces many different physiologic responses, which may be due to their localization on diverse cells in the brain. A 5-HT1A receptor antipeptide (aa170-186) antibody was produced that showed both high titer for peptide binding and immunocytochemical staining. Studies performed in perfusion-fixed brain tissue showed immunoreactive neurons, glial, and ependymal cells in the rat, mouse, cat, and monkey. Results from our studies of Macaca fascicularis brains are presented. We observed two main neuronal labeling patterns in the primate brain: (1) A general, diffuse somatodendritic distribution of 5-HT1A receptor immunoreactivity is seen in the raphe nuclei where the dendritic shaft, its branches and spines, and the entire perikaryon are immunolabeled. This pattern is also observed in the nucleus locus coeruleus, in scattered large brainstem reticular neurons, and in dentate gyrus hilar interneurons. (2) A discrete localization of 5-HT1A receptor immunoreactivity on the initial axon segment (axon hillock) is noted in pyramidal neurons of layer III and V of cerebral cortex, Cornu Ammonus (1-4) of the hippocampus, and in most brainstem and cervical spinal cord motoneurons. In addition to neuronal labeling, 5-HT1A receptor immunoreactivity is seen in the cell body and processes of astrocytes, and other nonneuronal cells. This pattern is particularly evident in the white matter of cerebral cortex and spinal cord, the pontine nuclei, the brainstem tectum, and the hilus of the dentate gyrus. The clinical implications of 5-HT1A cellular localization are briefly discussed.

Target cell stimulation of dissociated serotonergic neurons in culture.

Dissociated mesencephalic raphe cells from fetal rats (14-18 days) were grown in culture in 96 well Linbro plates. The maturation of serotonergic cells was qualitatively studied using immunocytochemistry with a serotonin antibody and quantitatively by measuring the retention of radioactivity following incubation in the presence of a low concentration of [3H]5-hydroxytryptamine (6 X 10(-8) M). The 5-hydroxytryptamine immunoreactive neurons showed specific staining in the perikaryon, nucleus, dendrites, axons and growth cones. These neurons formed varicose fibers and growth cones after 18 h in culture and survived for up to 21 days in culture. Each serotonergic neuron concentrated approximately 1 fmol of serotonin after 20 min of incubation. Maturation of mesencephalic serotonergic neurons was increased in co-cultures of both normal (hippocampus, cerebral cortex, olfactory bulb and striatum) and abnormal (spinal cord) target neurons. The best stimulation was produced by dissociated hippocampal neurons (14-18 days of gestation) on mesencephalic raphe cells (14 days of gestation) after 4 days in culture. This stimulation was seen in culture conditions which favored neuronal but not glial survival. Our results obtained using cultures of dissociated serotonergic cells are consistent with an expansive network pattern developed by this chemical transmitter system in the adult brain.