[Development of the habenulointerpeduncular tract in rats]

Development of the habenulointerpeduncular tract has been carried out on fixed brain preparations obtained from 21 day rat embryos and from neonatal animals on the 0 and 9 days of postnatal development by diffusion oflipophilic fluorescent carbocyanine dye DiI (1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate) through neuron membranes. The marker was introduced into the nuclei of the habenula, the interpeduncular nucleus, and into the area of raphe nuclei. Neurons and fibers that contained Dil were identified on vibratome sections by fluorescent and confocal microscopy. we have found that reciprocal links between the lateral habenula nucleus and raphe nuclei are formed in the prenatal period by stage E21. Raphe nuclei innervating neurons were located in dorso- and ventrocaudal parts of the lateral habenula nucleus. Projections of the medial habenula nucleus onto interpeduncular nucleus were found only in the postnatal P2 period. Neurons that provide a source of these projections form characteristic assemblies inside the medial habenula nucleus. Therefore, the present study for the first time describes heterogenic formation of different projection systems that are involved in the habenulointerpeduncular tract of rats at perinatal ontogenesis.

[Changes of the structural organization of the nucleus raphe pallidus after the decrease of endogenous serotonin level in the prenatal period of development in rats]

The role of serotonin in the nucleus raphe pallidus (NRP) development and the dynamics of its serotonin-producing neurons were studied during various time points of the postnatal period in normal Wistar rats and in animals developing prenatally under the conditions of serotonin deficiency. It was shown that NRP contained 2 populations of serotoninergic neurons with different morphological characteristics. At the initial stages of postnatal development (Day 5) serotonin-producing neurons included only large neurons, while the synthetic activity of small neurons appeared later (by Day 10). With age, under normal conditions,the size of large neurons and their number were increased which is indicative of continuing process of differentiation and/or functional load augmentation. The size and number of small neurons were practically unchanged with age. Serotonin deficiency during prenatal development lead to the disturbance of NRP structural organization. In comparison with the control animals, the size and the number of serotonin-producing neurons of both populations was decreased, their size remained unchanged with the age. Part of the neurons underwent degeneration, resulting in the reduction of their numbers. The damage observed may change the serotoninergic innervation of the medullary nuclei, responsible for the cardiorespiratory the control, thus causing the disturbances of cardio-vascular and respiratory systems.

Single-prolonged stress induces apoptosis in dorsal raphe nucleus in the rat model of posttraumatic stress disorder

INTRODUCTION

Post-traumatic stress disorder (PTSD) is an anxiety disorder that develops after exposure to a life-threatening traumatic experience. Meta-analyses of the brainstem showed that midsagittal area of the pons was significantly reduced in patients with PTSD, suggesting a potential apoptosis in dorsal raphe nucleus after single-prolonged stress (SPS). The aim of this study is to investigate whether SPS induces apoptosis in dorsal raphe nucleus in PTSD rats, which may be a possible mechanism of reduced volume of pons and density of gray matter.

METHODS

In this study, rats were randomly divided into 1d, 7d and 14d groups after SPS along with the control group. The apoptosis rate was determined using annexin V-FITC/PI double-labeled flow cytometry (FCM). Levels of Cytochrome c (Cyt-C) was examined by Western blotting. Expression of Cyt-C on mitochondria in the dorsal raphe nucleus neuron was determined by enzymohistochemistry under transmission electron microscopy (TEM). The change of thiamine monophosphatase (TMP) levels was assessed by enzymohistochemistry under light microscope and TEM. Morphological changes of the ultrastructure of the dorsal raphe nucleus neuron were determined by TEM.

RESULTS

Apoptotic morphological alterations were observed in dorsal raphe nucleus neuron for all SPS-stimulate groups of rats. The apoptosis rates were significantly increased in dorsal raphe nucleus neuron of SPS rats, along with increased release of cytochrome c from the mitochondria into the cytoplasm, increased expression of Cyt-C and TMP levels in the cytoplasm, which reached to the peak of increase 7 days of SPS.

CONCLUSIONS

The results indicate that SPS induced Cyt-C released from mitochondria into cytosol and apoptosis in dorsal raphe nucleus neuron of rats. Increased TMP in cytoplasm facilitated the clearance of apoptotic cells. We propose that this presents one of the mechanisms that lead to reduced volume of pons and gray matter associated with PTSD.

The dorsal raphe nucleus—From silver stainings to a role in depression

Over a hundred years ago, Santiago Ramón y Cajal used a new staining method developed by Camillo Golgi to visualize, among many other structures, what we today call the dorsal raphe nucleus (DRN) of the midbrain. Over the years, the DRN has emerged as a multifunctional and multitransmitter nucleus, which modulates or influences many CNS processes. It is a phylogenetically old brain area, whose projections reach out to a large number of regions and nuclei of the CNS, particularly in the forebrain. Several DRN-related discoveries are tightly connected with important events in the history of neuroscience, for example the invention of new histological methods, the discovery of new neurotransmitter systems and the link between neurotransmitter function and mood disorders. One of the main reasons for the wide current interest in the DRN is the nucleus' involvement in depression. This involvement is particularly attributable to the main transmitter of the DRN, serotonin. Starting with a historical perspective, this essay describes the morphology, ascending projections and multitransmitter nature of the DRN, and stresses its role as a key target for depression research.

Projection patterns from the amygdaloid nuclear complex to subdivisions of the dorsal raphe nucleus in the rat

The goal of the present study was to identify the projection from the subdivisions of the amygdaloid nuclear complex to specified subregions of the dorsal raphe (DR) nucleus and to attempt to compare the density of amygdaloid input to the DR with that of inputs from other limbic structures. Use of a retrograde tracer, gold-conjugated and inactivated wheatgerm agglutinin-horseradish peroxidase (WGA-apo-HRP-gold), demonstrated that amygdaloid input to midline DR subdivision originates mainly from the medial portion of the medial amygdaloid nucleus, whereas that to lateral wing subdivision derives from the region extending from the lateral portion of the medial amygdaloid nucleus to the commissural stria terminalis. Use of the retrograde tracer Fluorogold (FG) produced relatively large but circumscribed injection sites comprising midline DR as well as portions of lateral wing subdivision and confirmed that the medial amygdaloid nucleus provides the major input to the DR. We also demonstrated that although amygdaloid input was not as extensive as inputs from other limbic structures such as the medial prefrontal cortex or the lateral habenular nucleus, it was comparable to input from the lateral septal nucleus. Based on these observations, we suggest that the medial amygdaloid nucleus provides substantial input to the DR and may contribute an emotional influence on sleep-wakefulness cycle or pain-stress modulation. Furthermore, it seems that the medial amygdaloid-DR projection might be anatomically and functionally distinct from the well-characterized central amygdaloid-periaqeductal gray (PAG) circuit which is essential for conditioned fear.

Evidence that serotonin reuptake modulators increase the density of serotonin innervation in the forebrain

The mechanism of action of commonly used antidepressants remains an issue of debate. In the experiments reported here we studied the effects of three representative compounds, the selective serotonin reuptake inhibitor fluoxetine, the selective serotonin reuptake enhancer tianeptine and the selective norepinephrine reuptake inhibitor desipramine on the structure of central serotonin pathways after a 4-week administration. We found that the serotonin modulators fluoxetine and tianeptine, but not desipramine, increase the density of 5-HT and serotonin transporter (SERT)-immunoreactive axons in the neocortical layer IV and certain forebrain limbic areas, such as piriform cortex and the shell region of nucleus accumbens. These changes were noted in the absence of a significant effect of serotonin antidepressants on the expression of tryptophan hydroxylase (TPH-2), i.e. the rate-limiting enzyme for 5-HT biosynthesis and of SERT at the mRNA level. In addition, we found that anterogradely filled terminal axons from injections of biotinylated dextran amine into the dorsal raphe showed significantly more branching in animals treated with fluoxetine compared with animals treated with liposyn vehicle. Our findings suggest that antidepressants may exert very selective structural effects on their cognate monoamine systems in normal animals and raise the possibility that neurotrophic mechanisms may play a role in their clinical efficacy.

Two populations of glutamatergic axons in the rat dorsal raphe nucleus defined by the vesicular glutamate transporters 1 and 2

Most glutamatergic neurons in the brain express one of two vesicular glutamate transporters, vGlut1 or vGlut2. Cortical glutamatergic neurons highly express vGlut1, whereas vGlut2 predominates in subcortical areas. In this study immunohistochemical detection of vGlut1 or vGlut2 was used in combination with tryptophan hydroxylase (TPH) to characterize glutamatergic innervation of the dorsal raphe nucleus (DRN) of the rat. Immunofluorescence labeling of both vGlut1 and vGlut2 was punctate and homogenously distributed throughout the DRN. Puncta labeled for vGlut2 appeared more numerous then those labeled for vGlut1. Ultrastructural analysis revealed axon terminals containing vGlut1 and vGlut2 formed asymmetric-type synapses 80% and 95% of the time, respectively. Postsynaptic targets of vGlut1- and vGlut2-containing axons differed in morphology. vGlut1-labeled axon terminals synapsed predominantly on small-caliber (distal) dendrites (42%, 46/110) or dendritic spines (46%, 50/110). In contrast, vGlut2-containing axons synapsed on larger caliber (proximal) dendritic shafts (> 0.5 microm diameter; 48%, 78/161). A fraction of both vGlut1- or vGlut2-labeled axons synapsed onto TPH-containing dendrites (14% and 34%, respectively). These observations reveal that different populations of glutamate-containing axons innervate selective dendritic domains of serotonergic and non-serotonergic neurons, suggesting they play different functional roles in modulating excitation within the DRN.

Ultrastructural evidence for a role of gamma-aminobutyric acid in mediating the effects of corticotropin-releasing factor on the rat dorsal raphe serotonin system

The dorsal raphe nucleus (DRN) serotonin (5-HT) system has been implicated in acute responses to stress and in stress-related psychiatric disorders such as anxiety and depression. Substantial findings suggest that the neuropeptide corticotropin-releasing factor (CRF) is instrumental in modulating the activity of this system during stress. Because the DRN is neurochemically heterogeneous, dual immunoelectron microscopy was used to examine cellular substrates for interactions between CRF and either 5-HT or gamma-aminobutyric acid (GABA) in the dorsolateral and ventromedial DRN. CRF immunoreactivity was identified primarily within axon terminals, where immunolabeling was particularly enriched in dense-core vesicles. Although CRF terminals targeted 5-HT-containing dendrites in the dorsolateral DRN (16%; n = 251 terminals), synaptic contacts with dendrites that lacked detectable 5-HT immunolabeling were more numerous (48%). In contrast, dual labeling for CRF and GABA (n = 240 terminals) in the dorsolateral DRN revealed that substantially more CRF terminals contacted GABA dendrites (42%) as opposed to unlabeled dendrites (29%). In the ventromedial DRN, contacts between CRF axon terminals and either 5-HT-labeled dendrites or GABA-containing dendrites were fewer than in the dorsolateral DRN. As in the dorsolateral DRN, CRF terminals more frequently contacted GABA dendrites than 5-HT dendrites (30% vs. 8%, respectively). The findings support physiological studies suggesting that CRF has both direct and indirect effects on DRN-5-HT neurons and further implicate GABA as a primary mediator by which CRF and stressors alter the activity of the DRN-5-HT system.

The orexinergic synaptic innervation of serotonin- and orexin 1-receptor-containing neurons in the dorsal raphe nucleus

Orexin/hypocretin has been well demonstrated to excite the serotonergic neurons in the dorsal raphe nucleus (DRN). We studied the morphological relationships between orexin-containing axon terminals and serotonin- as well as orexin-receptor-containing neurons in the dorsal raphe nucleus. Using immunohistochemical techniques at the light microscopic level, orexin A (OXA)-like immunoreactive neuronal fibers in the DRN were found to make close contact with serotonergic neurons, while some of the serotonergic neurons also expressed the orexin 1 receptor (OX1R). At the electron microscopic level, double-immunostaining experiments showed that the orexin A-like immunoreactive fibers were present mostly as axon terminals that made synapses on the serotonin- and orexin 1-receptor-containing neurons. While only axodendritic synapses between orexin A-containing axon terminals and serotonergic neurons were detected, the synapses made by orexin A-containing axon terminals on the orexin 1-receptor-containing neurons were both axodendritic and axosomatic. The present study suggests that excitation effect of orexin A on dorsal raphe serotonergic neurons is via synaptic communication through orexin 1 receptor.

Prefrontal cortical projections to the rat dorsal raphe nucleus: ultrastructural features and associations with serotonin and gamma-aminobutyric acid neurons

Studies of human brain indicate that both the ventromedial prefrontal cortex (PFC) and the dorsal raphe nucleus (DRN) may be dysfunctional in major depressive illness, making it important to understand the functional interactions between these brain regions. Anatomical studies have shown that the PFC projects to the DRN, although the synaptic targets of this excitatory pathway have not yet been identified. Electrophysiological investigations in the rat DRN report that most serotonin neurons are inhibited by electrical stimulation of the PFC, suggesting that this pathway is more likely to synapse onto neighboring gamma-aminobutyric acid (GABA) neurons than onto serotonin cells. We tested this hypothesis by electron microscopic examination of DRN sections dually labeled for biotin dextran amine anterogradely transported from the PFC and immunogold-silver labeling for tryptophan hydroxylase (TrH) or for GABA. In the DRN, the majority of PFC axons either synapsed onto unlabeled dendrites or failed to form detectable synapses in single sections. Other PFC axons synapsed onto either TrH- or GABA-immunolabeled processes. Considerably more tissue sampling was necessary to detect PFC synapses onto TrH- than onto GABA-labeled dendrites, suggesting that the latter connections are more common. In other cases, PFC terminals and TrH- or GABA-immunoreactive dendrites either were closely apposed, without forming detectable synapses, or were separated by glial processes. These results provide potential anatomical substrates whereby the PFC can both directly and indirectly regulate the activity of serotonin neurons in the DRN and possibly contribute to the pathophysiology of depression.

Median raphe mediates estrogenic effects to the hippocampus in female rats

Subcortical regions such as the medial septum-diagonal band of Broca and supramammillary area have been shown to mediate indirect oestrogenic effects on hippocampal morphology and function. Here, the role of the median raphe (MR), a serotonergic subcortical structure, is studied. To this end, 17beta-estradiol-filled 30-gauge cannulae were implanted into the MR of female ovariectomized rats; cholesterol-filled cannulae served as controls. After seven days, using unbiased electron microscopic stereological calculations and semiquantitative analysis, the spine synapse density and surface density of glial fibrillary acidic protein-positive astrocyte processes, respectively, were determined in the stratum radiatum of the CA1 region of the hippocampus. Changes in the serotonergic innervation of the hippocampal CA1 region were determined by immunohistochemistry and subsequent morphometric analysis. In the stratum radiatum of the CA1 region, local estradiol application into the MR resulted in a 47% increase in spine synapse density. Simultaneously, the density of glial fibrillary acidic protein-positive fibers decreased by 16%. The density of serotonin (5-HT) innervation of the strata lacunosum moleculare and radiatum of the CA1 region of the hippocampus was reduced in response to estradiol, as shown by a decrease in the length of fibers (27.6 and 48.3% decrease, respectively) and the number of large varicosities (32.5 and 38.8% decrease, respectively). These observations suggest a major role of the MR in mediating oestrogenic effects on the hippocampus and an involvement of the serotonergic system.

Regional, cellular, and subcellular localization of RGS10 in rodent brain

The regulator of G protein signaling type 10 (RGS10) modulates Galphai/o signaling by means of its GTPase accelerating activity and is abundantly expressed in brain and in immune tissues. To elucidate RGS10 function in the nervous system, we mapped RGS10 protein in rat and mouse brain using light microscopic (LM) and electron microscopic (EM) immunohistochemical techniques. The LM showed that RGS10-like immunoreactivity (LIR) labels all cellular subcompartments of neurons and microglia, including their nuclei. There were several differences between RGS10-LIR distributions in rat and mouse, the most striking of which were the far denser immunoreactivity in rat dentate gyrus and dorsal raphe. The EM analysis corroborated and extended our findings from LM. Thus, EM confirmed the presence of dense RGS10-LIR in the euchromatin compartment of nuclei. The EM analysis also resolved dense staining on terminals at symmetric synapses onto pyramidal cell somata. Dual immunofluorescence showed that forebrain interneurons densely labeled with RGS10-LIR partially colocalized with parvalbumin-LIR. Dual-labeling histochemistry in caudoputamen demonstrated that densely labeled striatal cells were biased to the indirect-projecting output pathway. Dual-labeling immunofluorescence also showed that densely labeled RGS10-LIR cells in the dentate gyrus subgranular zone were not proliferating but that newly born cells could differentiate to express RGS10-LIR. Taken together, these data support a role for RGS10 in diverse processes that include modulation of pre- and postsynaptic G-protein signaling. Moreover, enrichment of RGS10 in transcriptionally active regions of the nucleus suggests an unforeseen role of RGS10 in modulating gene expression.

Electron microscopic examination of the orexin immunoreactivity in the dorsal raphe nucleus

The ultrastructure and the synaptic relationships of the orexin-A-like immunoreactive fibers in the dorsal raphe nucleus were examined with an immunoelectron microscopic method. At the electron microscopic level, most of the immunoreactive fibers, a varicosity appearance at the light microscopic level, were found as axon terminals. The large dense-cored vesicles contained in the immunoreactive axon terminals were the most intensely immunostained organellae. These axon terminals were often found to make synapses. While the axo-dendritic synapses were usually asymmetric in appearance, the axo-somatic synapses were symmetric. Orexin-A-like immunoreactive processes with no synaptic vesicles were also found. These processes often received asymmetric synapses. With less frequency, the synapses were found between the orexin-like immunoreactive processes. The results suggest that the orexin peptides are stored in the large dense-cored vesicles; the orexin-containing fibers may have influences on the physiological activities of the dorsal raphe nucleus through direct synaptic relationships.

Chronic excess of corticosterone increases serotonergic fibre degeneration in aged rats

Evidence is presented for the potentiating role of corticosterone on axonal degeneration of serotonergic neurones during ageing. Aged rats, 24 months old, were implanted subcutaneously with 2 x 100 mg pellets of corticosterone. Serotonergic and cholinergic (ChAT- and NADPHd-positive) fibre degenerations in the anteroventral thalamic nucleus (AVT) were measured 2 months after corticosterone implantation. Numbers of immunoreactive serotonergic raphe and mesolimbic cholinergic neurones were also quantified. Basal plasma corticosterone and adrenocorticotropin (ACTH) concentrations were assayed at 2, 4, 6, and 8 weeks after implantation in the plasma and at 1, 2, 4 and 6 weeks in urine. The degree of serotonergic fibre aberrations in the AVT increased significantly after corticosterone exposure, while that of ChAT-positive and NADPHd-stained axon aberrations showed a modest but nonsignificant increase. A positive correlation between the magnitudes of serotonergic and cholinergic fibre aberrations appeared in the AVT, but only in the corticosterone-treated rats. The number of serotonin immunopositive neurones in the raphe nuclei after corticosterone decreased marginally, while that of mesopontine ChAT-positive neurones was not influenced. Measurements of basal plasma corticosterone and ACTH, as well as urine corticosterone, revealed that the steroid implantation increased the plasma corticosterone level for at least 4 weeks and decreased ACTH level for at least 6 weeks. By the week 8, the pituitary-adrenal function was apparently restored. However, at sacrifice, both the weight of adrenal glands and that of thymus remained reduced, indicating the long-lasting effects of corticosterone on target tissues. It is concluded that the raphe serotonergic neurones and their projecting fibres are sensitive to corticosterone excess in aged rats and become more vulnerable to degeneration processes than under normal ageing conditions. Cholinergic neurones of brainstem origin, which also express massive NADPHd activity, are more resistant against corticosterone, but their axon degeneration correlates to serotonergic fibre degeneration.

Serotonin immunoreactivity in the retinal projecting isthmo-optic nucleus and evidence of brainstem raphe connections in the pigeon

Serotonin (5-HT) immunoreactive (-ir) profiles within the isthmo-optic nucleus (ION) of the centrifugal visual system (CVS) were studied in the pigeon using light microscopic immunohistofluorescent and electron microscopic immunocytochemical pre-embedding techniques. The brainstem origin of the 5-HT input upon the ION was determined by combining 5-HT immunohistofluorescence (FITC) and retrograde transneuronal tracing after intraocular injection of Rhodamine beta-isothiocyanate. The light microscopic results showed that 5-HT endings were mainly localised within the neuropillar zones of the ventral ION. The 5-HT-ir cell bodies, belonging to a lateral extension of the dorsal raphe system, were observed within the same region as the centrifugal ectopic neurons (EN) underlying the ION and some displayed dendritic processes which penetrated the nucleus. Double-labeled neurons, representing 5-HT-ir afferents to the ION, were identified only within the n. linearis caudalis region of the ventral raphe. The electron microscopic results confirmed the presence of 5-HT-ir dendritic processes within the ventral part of the nucleus and showed that they were contacted by axon terminals belonging to intrinsic interneurons. The functional organisation of the ION and the possible contribution of serotonergic raphe afferents and efferents are discussed in relation to present hypotheses linking the avian CVS to mechanisms of visual attention.

Cellular basis for the effects of substance P in the periaqueductal gray and dorsal raphe nucleus

Substance P (SP) is known to act at supraspinal sites to influence pain sensitivity as well as to promote anxiety. The effects of SP could be mediated in part by actions in the periaqueductal gray (PAG) and the dorsal raphe nucleus (DRN), adjoining mesencephalic cell groups that are strategically positioned to influence both nociception and mood. Previous studies have indicated that SP regulates both enkephalin and serotonin neurotransmission in these brain regions. To determine the mechanism underlying the effects of SP in the PAG and DRN, the distribution of the principal receptor for SP, the neurokinin 1 (NK1) receptor, was examined with respect to other neurotransmitter markers. PAG neurons that had NK1 receptor immunolabeling were interdigitated with and received contacts from enkephalin-containing neurons. However, only a few (16/144; 11%) neurons with NK1 receptor also contained enkephalin immunoreactivity after colchicine treatment. In the DRN, dendrites containing NK1 receptor were selectively distributed in the dorsomedial subdivision. The majority (132/137; 96%) of these dendrites did not contain immunoreactivity for the serotonin-synthesizing enzyme tryptophan hydroxylase. In contrast, neuronal profiles with NK1 receptor in both the PAG and the DRN often contained immunolabeling for glutamate. Light and electron microscopic examination revealed that 48-65% of cell bodies and dendrites with NK1 receptor were dually immunolabeled for glutamate. These data suggest that SP directly acts primarily on glutamatergic neurons in the PAG and DRN. To a lesser extent, enkephalin-containing neurons may be targeted. Through these actions, it may subsequently influence activity of larger populations of neurons containing enkephalin as well as serotonin. This circuitry could contribute to, as well as coordinate, effects of SP on pain perception and mood.

Altered presynaptic function in monoaminergic neurons of monoamine oxidase-A knockout mice

Monoamine oxidase-A knockout (MAO-A KO) mice have elevated brain serotonin (5-HT) and noradrenaline (NA) levels, and one would therefore anticipate increased monoamine release and compensatory changes in other aspects of presynaptic monoamine function. In this study we used voltammetry in brain slices from the locus coeruleus (LC), dorsal raphe (DRN) and striatum (CPu) in 7-week-old MAO-A KO and C3H control mice to measure stimulated monoamine efflux and its control by amine transporters and autoreceptors. In LC, peak NA efflux on stimulation (99 pulses, 100 Hz) was higher in MAO-A KO than C3H mice (938 +/- 58 nm cf. 511 +/- 42 nm; P < 0.001). The NA uptake half time (t(1/2)) was longer in MAO-A KO than in C3H mice (6.0 +/- 0.9 s cf. 1.9 +/- 0.3 s; P < 0.001) and the selective NA reuptake inhibitor desipramine (50 nm) had a smaller effect in MAO-A KO mice. NA transporter binding was significantly lower in the LC of MAO-A KO mice compared to C3H controls (P < 0.01) but not in the DRN. The alpha 2 agonist dexmedetomidine (10 nm) decreased stimulated NA efflux more in C3H than in MAO-A KO mice (73.3% cf. 29.6% inhibition, P < 0.001). In DRN, peak 5-HT efflux on stimulation (99 pulses, 100 Hz) was greater (P < 0.01) in MAO-A KO (262 +/- 44 nm) than C3H mice (157 +/- 16 nm). Moreover, 5-HT uptake t(1/2) was longer (P < 0.05) in MAO-A KO than in C3H mice (8.8 +/- 1.1 s cf. 4.9 +/- 0.6 s, P < 0.05) and the effect of citalopram (75 nm) was attenuated in MAO-A KOs. Serotonin transporter binding was also lower in both the DRN and LC of MAO-A KO mice. The 5-HT(1A) agonist 8-OH-DPAT (1 microm) decreased 5-HT efflux more in C3H than in MAO-A KO mice (38.3% inhibition cf. 21.6%, P < 0.001). In contrast, there were no significant differences between MAO-A KO and C3H mice in CPu dopamine efflux and uptake and the effect of the D(2/3) agonist quinpirole was similar in the two strains. In summary, MAO-A KO mice show major dysregulation of monoaminergic presynaptic mechanisms such as autoreceptor control and transporter kinetics.

GABA(B) receptors in the median raphe nucleus: distribution and role in the serotonergic control of hippocampal activity

Previous studies have shown that serotonergic neurons of the median raphe nucleus have a suppressive effect on theta synchronization in the hippocampus. Median raphe lesion, suppression of 5-HT neuronal activity by administration of GABA(A) receptor antagonist or by glutamate blockade or depletion produced long-lasting non-interrupted hippocampal theta in freely behaving rats independent of behavior and in rats anesthetized with urethane. Serotonergic neurons show a characteristic sleep-wake pattern of activity and there is evidence that GABAergic mechanisms play an important role in their regulation. In this study we analyzed the distribution and subcellular localization of GABA(B) receptors in the midbrain raphe complex using combined 5-HT/GABA(B) receptor immunohistochemistry at the light and electron microscopic levels and studied the effects of their pharmacological manipulation on hippocampal electroencephalographic activity in urethane-anesthetized rats. We found that sustained infusion of the GABA(B) receptor agonist baclofen into the median raphe nucleus, using the microdialysis technique, elicited lasting theta activity in the hippocampus. The effect was antagonized by selective GABA(B) receptor antagonists. The predominant localization of GABA(B) receptors in the median, as well as in dorsal raphe was found on serotonergic neurons which strongly indicates that the increase in theta occurrence after baclofen injection resulted from suppression of the serotonergic output originating from the median raphe. On the electron microscopic level, we found GABA(B) receptors located extrasynaptically indicating that these receptors are preferentially activated by strong inputs, i.e. when GABA released from the synaptic terminals is sufficient to spill over from the synaptic cleft. Such conditions might be satisfied during rapid eye movement sleep when GABAergic neurons in the raphe are firing at their highest rate and in rhythmic synchronized bursts. Our data indicate that midbrain raphe GABA(B) mechanisms play an important role in behavioral state control and in hippocampal activity, in particular.

In utero and lactational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin decreases serotonin-immunoreactive neurons in raphe nuclei of male mouse offspring

Female ddY mice were administered 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) by gavage for 8 weeks prior to pregnancy. In the male breast-fed offspring born to the TCDD-exposed mice, serotonergic neurons in the brainstem were examined using an immunocytochemical method at 42 days of age. In all offspring, a marked decrease in the intensity of immunostaining occurred in all raphe nuclei compared with the control offspring. The number of serotonin-immunoreactive neurons in each raphe nucleus was measured by computer-assisted analysis. Approximately a quarter to half of immunoreactive neurons were detected in the TCDD-exposed offspring raphe nuclei compared with the control offspring. The present findings suggest that in utero and/or lactational TCDD exposure cause a long-lasting change in the serotonergic system in the raphe nuclei of offspring.

Evidence for a role of GABA interneurones in the cortical modulation of midbrain 5-hydroxytryptamine neurones

Recent electrophysiological studies demonstrate that the ventral medial prefrontal cortex has a powerful inhibitory influence on 5-hydroxytryptamine (5-HT) neurones in the dorsal raphe nucleus. Here we utilised a combination of anatomical and electrophysiological methods to characterise the cellular substrate underlying this effect.Anterograde tracing (Phaseolus vulgaris leucoagglutinin) using electron microscopy demonstrated a pathway from the ventral medial prefrontal cortex that makes neuronal contacts throughout the dorsal raphe nucleus. These contacts were predominantly asymmetrical synapses adjoining GABA immunoreactive dendrites and spines. In vivo extracellular recordings were made in the dorsal raphe nucleus of the anaesthetised rat from a subpopulation of non-5-HT neurones. These neurones were fast-firing, irregular and with short spike width, properties strongly reminiscent of immunochemically identified GABA interneurones in other brain regions. Recordings of classical 5-HT neurones were also included. Electrical stimulation of the ventral medial prefrontal cortex elicited a rapid onset (16 ms latency), orthodromic excitation of the non-5-HT neurones (13/25 neurones). This stimulation also caused a pronounced inhibition of most 5-HT neurones tested, with a longer latency (30 ms), and this was partially blocked by locally applied bicuculline. These data provide the first evidence that the ventral medial prefrontal cortex influences the activity of large numbers of raphe 5-HT neurones by targeting a local network of GABA neurones. This circuitry predicts that physiological and pathological changes in the ventral medial prefrontal cortex will impact on significant parts of the forebrain 5-HT system.

Serotonergic innervation of the isthmo-optic nucleus of the pigeon centrifugal visual system. An immunocytochemical electron microscopic study

The ultrastructural features of serotonergic fibers, terminals and synaptic contacts were studied with the pre-embedding immunocytochemical method in the isthmo-optic nucleus of the pigeon centrifugal visual system. The 5-HT immunoreactive (-ir) profiles were diffusely distributed and their density was low. The labeled axons were thin and unmyelinated (mean diameter=0.21+/-0.03 microm) though a few larger myelinated axons were observed (mean diameter=0.51+/-0.07 microm). The 5-HT-ir terminals or varicosities were small (diameter=0.71+/-0.54 microm) and contained small agranular synaptic vesicles (diameter=28.5+/-6.9 nm) and large granular vesicles (diameter=102.2+/-19.5 nm). The latter only constituted approximately 1% of the total profiles containing synaptic vesicles in the isthmo-optic nucleus. In single thin sections, only 5% of the 5-HT-ir varicosities exhibited an active asymmetrical zone synapsing upon dendritic profiles of centrifugal visual neurons. Calculations indicated that 17% of these 5-HT-ir varicosities were actually engaged in junctional synaptic relationships, whereas the remaining (83%) were nonjunctional. The data suggest that, within the isthmo-optic nucleus, 5-HT acts both at synaptic junctions (wiring transmission) and at a distance via the extracellular space (volume transmission). These 5-HT afferents could thus modulate the activity of the retinopetal neurons and visual information processing.

Morphological study of 5-HT neurons and astroglial cells on brain of adult rats perinatal or chronically exposed to 2,4-dichlorophenoxyacetic acid

2,4-D is a chlorophenoxyherbicide used worldwide. We have studied the morphological alterations of 5-HT neurons and glial cells in the mesencephalic nuclei of adult rats exposed to 2,4-D both perinatally (during pregnancy and lactation) and chronically (during pregnancy, lactation and after weaning) with quantitative methods. Pregnant rats were daily exposed to 70 mg/kg of 2,4-D from gestation day (GD) 16 to post-natal day (PND) 23 through diet. After weaning, pups were assigned to one of two sub-groups: T1 (fed with untreated diet until PND 90) and T2 (maintained with 2,4-D diet until PND 90). Brain sections were immunocytochemically stained using polyclonal anti-5-HT, anti-GFAP and anti-S-100 protein antibodies as cells markers. 2,4-D exposure during pregnancy and lactancy (T1 group) produced an increase in 5-HT neuronal area and immunoreactivity (IR) in the mesencephalic nuclei studied. However, with the chronical 2,4-D exposure (T2 group) only the 5-HT neuronal area from the dorsal raphe nucleus (DRN) was increased, suggesting an adaptable response of 5-HT neurons in median raphe nucleus (MRN). The presence of reactive astrocytes in mesencephalic nuclei and in hippocampus were also different for the two 2,4-D exposure designs, showing the existence of a correspondence between neuronal changes and astrogliosis. Results support evidences that 2,4-D alters the serotoninergic system and that 5-HT neurons of each mesencephalic nuclei show different responses to the 2,4-D exposure designs which are parallel to astrogliosis.

Agonist-induced internalization of serotonin-1a receptors in the dorsal raphe nucleus (autoreceptors) but not hippocampus (heteroreceptors)

Serotonin-1A (5-HT(1A)) receptors in the CNS are a major target for psychotropic drugs. In nucleus raphe dorsalis (NRD) and hippocampus (CA3), the selective 5-HT(1A) agonist (+)-8-hydroxy-2-(di-N-propylamino) tetralin (8-OH-DPAT) reduces the firing activity of serotoninergic (5-HT) and pyramidal neurons, respectively. When located on 5-HT (autoreceptors), but not on non-5-HT (heteroreceptors) neurons, 5-HT(1A) receptors are known to be subject to desensitization. Using quantitative electron microscopy after pre-embedding immunogold labeling with specific antibodies, we examined the subcellular distribution of these receptors after acute administration of 8-OH-DPAT (0.5 mg/kg, i.v.). Silver-intensified immunogold particles associated with the plasma membrane or the cytoplasm were counted in somata and dendrites within the NRD, 15 min, 1 hr and 24 hr after 8-OH-DPAT injection, and in hippocampal dendrites 1 hr after the same treatment. Significant decrease in the density of membrane labeling and concomitant increase of cytoplasmic labeling were demonstrated in the NRD, 15 min and 1 hr after 8-OH-DPAT administration, with a return to baseline level at 24 hr. Internalization was blocked by previous administration of the 5-HT(1A) antagonist N-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-N-(2-pyridinyl) cyclohexane-carboxamide (WAY 100635), which, by itself, was without apparent effect. In hippocampus (CA3), there were no apparent changes in the distribution of the receptor after 8-OH-DPAT administration. These findings are in line with earlier results showing a desensitization of 5-HT(1A) autoreceptors but not heteroreceptors after treatment with 5-HT(1A) receptor agonist. They suggest that this desensitization is the result of autoreceptor internalization.

Synaptic contacts between serotonergic and cholinergic neurons in the rat dorsal raphe nucleus and laterodorsal tegmental nucleus

We examined synaptic connectivity between cholinergic and serotonergic neurons in the dorsal raphe nucleus and the laterodorsal tegmental nucleus of the rat. To this purpose we employed two variations (the combination of pre-embedding immunogold-silver intensification with avidin-biotin-peroxidase complex technique and the combination of avidin-biotin-peroxidase/3, 3'-diaminobenzidine/silver-gold intensification with avidin-biotin-peroxidase/3,3'-diaminobenzidine reaction) of a double pre-embedding immunoelectron procedure, using primary antibodies against vesicular acetylcholine transporter and serotonin. At the light-microscopic level, serotonin-like immunoreactive neurons in the dorsal raphe nucleus appeared as reddish black and vesicular acetylcholine transporter-like immunoreactive axon terminals were brown colored using a combination of pre-embedding immunogold-silver technique and avidin-biotin-peroxidase complex technique. Serotonin-like immunoreactive fibers projected to the laterodorsal tegmental nucleus. At the electron microscopy level, with both methods we observed in the dorsal raphe nucleus vesicular acetylcholine transporter-immunopositive axon terminals in synaptic contact with serotonin-like immunoreactive dendrites and, to a lesser degree, with serotonin-like immunoreactive cell bodies. These synapses usually were of the symmetrical type. Occasionally we noted, next to vesicular acetylcholine transporter-immunopositive axon terminals, also immunonegative terminals synapsing with the serotonin-like immunoreactive dendrites. In the laterodorsal tegmental nucleus we found serotonin-like immunoreactive axon terminals and immunonegative terminals forming synapses with vesicular acetylcholine transporter-immunoreactive dendrites. Most synapses formed by the serotonin-like immunopositive terminals were of the asymmetrical type. Our results suggest that serotonergic neurons in the dorsal raphe nucleus and cholinergic neurons in the laterodorsal tegmental nucleus may reciprocally influence each other by means of synaptic connectivity. Such connectivity may serve to regulate pain sensation, or be involved in the regulation of the sleeping-waking cycle.

Estrogen receptor-alpha in the raphe serotonergic and supramammillary area calretinin-containing neurons of the female rat

It is well established that estrogen affects hippocampal long-term potentiation and hippocampus-related memory processes. Furthermore, theta rhythm, in conjunction with long-term potentiation, influences memory and is regulated by subcortical structures, including the median raphe and supramammillary area. To test the validity of the hypothesis that the effects of estrogen on the hippocampus are mediated, at least partly, via these subcortical structures, it must first be determined whether the neurons of the median raphe and supramammillary area contain estrogen receptors. Light and electron microscopic double immunostaining for estrogen receptor-alpha plus serotonin and estrogen receptor-alpha plus calretinin on vibratome sections of the median raphe and supramammillary area, respectively, demonstrated that large populations of the median raphe serotonin and supramammillary area calretinin neurons exhibit estrogen receptor-immunoreactive nuclei. These observations indicate that circulating gonadal hormones can affect hippocampal electric activity indirectly, via those subcortical structures that are involved in theta rhythm regulation.

Distribution of bulbospinal gamma-aminobutyric acid-synthesizing neurons of the ventral respiratory group of the rat

Spinal respiratory motoneuron activity is controlled primarily by excitatory and inhibitory neurons in the medulla oblongata. To identify bulbospinal inhibitory neurons, immunohistochemistry for glutamic acid decarboxylase (GAD) was combined with retrograde labeling of projections to the C(4) ventral horn with Fluoro-Gold. GAD-immunoreactive bulbospinal neurons were located in the ventrolateral portion of the intermediate reticular nucleus, the ventral portion of the medial reticular nuclei, and the raphe and spinal vestibular nuclei. Small numbers of bulbospinal ventral respiratory group neurons were GAD immunoreactive. These neurons were distributed throughout the rostral ventral respiratory group and the Bötzinger complex. Surprisingly, low numbers of Bötzinger neurons, a population thought to be exclusively inhibitory, were GAD immunoreactive. These results suggest that the rostral ventral respiratory group and the Bötzinger complex both contain heterogeneous bulbospinal neuron populations, only some of which have gamma-aminobutyric acid (GABA)-mediated inhibitory control over phrenic motoneurons. Furthermore, the ventral respiratory group contained many GABAergic neurons that lacked bulbospinal projections.

Morphometry of the dorsal raphe nucleus serotonergic neurons in suicide victims

BACKGROUND

The serotonin deficiency hypothesis of suicide has been important heuristically. Few studies have directly examined the brainstem dorsal raphe nucleus (DRN) serotonin neurons. We determined the number and morphometry of DRN serotonergic neurons in suicide victims (n = 7) compared to controls (n = 6).

METHODS

Brainstems were collected at autopsy, fixed and cryoprotected. Tissue was sectioned, stained for Nissl and processed with an antiserum that cross-reacts with tryptophan hydroxylase. All DRN neurons were identified, counted and analyzed every 1000 microns. Neuron morphometry was characterized by soma area (micron 2), sphericity, perimeter, length and density (neurons per mm3).

RESULTS

Neuron number and density was higher in suicide victims (1,780 +/- 127 neurons/mm3) than controls (1,349 +/- 68). The DRN volume did not differ between groups (66 +/- 9 mm3 for controls vs. 67 +/- 5 mm3 for suicides). Mean neuronal area and sphericity did not differ between suicides and controls. The total number and the density of DRN neurons did not correlate with age.

CONCLUSIONS

The finding of an increased number of neurons indicates that impaired serotonergic transmission found in association with serious suicide attempts is not due to fewer neurons.

Medial septal and median raphe innervation of vasoactive intestinal polypeptide-containing interneurons in the hippocampus

Vasoactive intestinal polypeptide-immunoreactive interneurons are known to form three anatomically and neurochemically well-characterized neuron populations in the hippocampus. Two of these establish synaptic contacts selectively with other GABAergic cells (interneuron-selective cells), whereas the third type innervates pyramidal cell bodies and proximal dendrites like a conventional basket cell. Our aim was to examine which of the vasoactive intestinal polypeptide-containing interneuron populations are among the targets of GABAergic septohippocampal and serotonergic raphe-hippocampal pathways. Anterograde tracing with Phaseolus vulgaris leucoagglutinin combined with double immunocytochemistry for vasoactive intestinal polypeptide was used at the light and electron microscopic levels. Our results show that both interneuron-selective cells and vasoactive intestinal polypeptide-containing basket cells receive synaptic input from the medial septum and median raphe nucleus. The GABAergic component of the septohippocampal pathway establishes multiple contacts on both cell types. In the case of the raphe-hippocampal projection, single or double contacts were more frequent on vasoactive intestinal polypeptide-positive interneuron selective cells (76%), whereas multiple contacts predominated on basket cells (83%). The extrinsic GABAergic innervation of interneuron-selective cells in the hippocampus indicates a complex interaction among GABAergic systems, which might ensure the timing and rhythmic synchronization of inhibitory processes in the hippocampus. On the other hand, our results suggest that the serotonergic effect on perisomatic inhibition is exerted via vasoactive intestinal polypeptide-containing basket cells that are functionally distinct from their parvalbumin-positive relatives, which appear to escape control of serotonergic as well as local interneuron-selective cells.

Electrophysiological comparison of 5-Hydroxytryptamine1A receptor antagonists on dorsal raphe cell firing

Single-unit recording studies were undertaken in chloral hydrate-anesthetized rats to compare the effects on dorsal raphe cell firing of several putative 5-hydroxytryptamine (HT)1A receptor antagonists, including WAY 100635 (N-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-N-(2-pyridinyl) cyclohexanecarboxamide), p-MPPI (4-(2-methoxyphenyl)1-[2'-[N-(2"-pyridinyl)-p-iodobenzamido]ethyl] pip erazine), and two newly described 5-HT1A receptor antagonists, NDL-249 [(R)-3-(N-propylamino)-8-fluoro-3, 4-dihydro-2H-1-benzopyran-5-carboxamide] and NAD-299 [(R)-3-N, N-dicyclobutylamino-8-fluoro-3, 4-dihydro-2H-1-benzopyran-5-carboxamide]. Consistent with a 5-HT1A receptor antagonist profile, pretreatment with an approximately equimolar (0.02-0.03 micromol/kg) i.v. dose of each compound caused a significant rightward shift in the dose-response curve for 8-OH-DPAT [8-hydroxy-2-(di-n-propylamino)tetralin]. Antagonist potency was clearly highest for NAD-299 and WAY 100635, which caused shifts roughly 3 times greater than those for either p-MPPI or NDL-249 (ED50 for 8-OH-DPAT, 1.3 +/- 0.3 microg/kg; after NAD-299, 18.2 +/- 1.0 microg/kg; after WAY 100635, 16.9 +/- 2.9 microg/kg; after NDL-249, 6.0 +/- 1.2 microg/kg; after p-MPPI, 4.7 +/- 1.1 microg/kg). In separate studies, each of the antagonists was administered alone in increasing cumulative doses to evaluate whether they possessed intrinsic agonist activity in this system. At doses below 0.01 micromol/kg, none of the drugs altered firing by more than +/-20% basal rates. At higher doses (>0.1 micromol/kg), WAY 100635, NDL-249, and NAD-299 caused a dose-dependent suppression of dorsal raphe cell firing (ED50 = 0.6 +/- 0.2, 0.7 +/- 0.3, and 0. 9 +/- 0.4 micromol/kg, respectively). However, the ED50 values for inhibition by these drugs were roughly 30 times higher than the doses that antagonized effects of 8-OH-DPAT. Moreover, the inhibition by all three antagonists (but not 8-OH-DPAT) was readily reversed by d-amphetamine (3.2 mg/kg i.v.), a releaser of norepinephrine, suggesting that these effects were likely due to alpha adrenergic receptor blockade rather than to 5-HT1A receptor agonism. Thus, it was concluded that WAY 100635, NAD-299, NDL-249, and p-MPPI all fulfill criteria as 5-HT1A receptor antagonists lacking intrinsic efficacy in the dorsal raphe system. The newly described compound NAD-299 exhibits antagonist potency comparable to that of WAY 100635 in this electrophysiological assay.

A microdialysis study of the in vivo release of 5-HT in the median raphe nucleus of the rat

The present study has examined several characteristics of the release of 5-HT in the median raphe nucleus in terms of its dependence of nerve impulse, provenance of a vesicular storage fraction as well as the regulatory role played by 5-HT1A receptors. Tetrodotoxin (1 microM) and reserpine (5 mg kg(-1), i.p.) virtually suppressed the output of 5-HT. The administration of EEDQ (10 mg kg(-1), i.p.) did not alter the basal release of 5-HT but abolished the reduction of 5-HT release induced by 8-OH-DPAT (0.1 mg kg(-1), s.c.). The perfusion of 1-100 microM of 8-OH-DPAT or the novel 5-HT1A agonist BAY x 3702 decreased the efflux of 5-HT, whereas the perfusion of the 5-HT1A antagonist WAY-100635 failed to alter 5-HT release. The decrease in dialysate 5-HT induced by 100 microM 8-OH-DPAT was reversed by the concurrent perfusion of 100 microM WAY-100635. Also, the perfusion of 100 microM WAY-100635 for 2 h inhibited partly the reduction of 5-HT release evoked by the systemic administration of 8-OH-DPAT (0.1 mg kg(-1)). These results indicate that extracellular 5-HT in the median raphe nucleus is stored in vesicles and released in an impulse-dependent manner. Also, the basal release of 5-HT in the median raphe nucleus does not appear to be under the tonic control of somatodendritic 5-HT1A receptors by endogenous 5-HT. Instead, this feedback mechanism seems to be triggered when an excess of the transmitter or a 5-HT1A agonist is present in the extracellular space of the median raphe nucleus.

Ultrastructural localization of mu-1 opioid receptor in the dorsal raphe nucleus of the rat

A simple pre-embedding avidin-biotin-peroxidase complex technique was used to study the ultrastructural localization of mu-1 opioid receptor in the rat dorsal raphe nucleus. Using low concentrations of the first antiserum for incubation with a short reaction time to 3,3'-diaminobenzidine, the immunostaining was faint at the light microscopic level. However, at the electron microscopic level strong immunoreaction was observed. Mu-1 opioid receptors were found to be localized on the postsynaptic membrane of dendrites, extra-synaptic plasma membrane, and the surface of the small, clear vesicles in axon terminals. Of the total 407 immunopositive profiles observed, 76.4% (311/407) were dendrites and 18.9% (77/407) were axon terminals. The immunostained myelinated axons and perikarya were relatively rare, with frequencies of 1.0% (4/407) and 3.7% (15/407), respectively. About 50.8% of the immunopositive dendrites (158/311) were immunostained having their MOR-LI results beneath the postsynaptic membrane, although about 19.6% of them (31/158) also exhibited MOR-LI on other components, including the extrasynaptic plasma membrane. Other immunopositive dendrites showed staining in some other contents, including extrasynaptic plasma membrane (82/311, 26.4%) or not on the plasma membranes (71/311, 22.8%). Less than half of the immunopositive axon terminals (35/77, 45.5%) were found to make synapses with nonimmunoreactive dendrites (31/77, 40.3%) or immunopositive dendrites (4/77, 5.2%); none were found to make synapses with immunoreactive perikarya. The present study shows that mu-1 opioid receptor in the dorsal raphe nucleus plays a role at both synapse or not.

Neuropeptide FF receptors control morphine-induced analgesia in the parafascicular nucleus and the dorsal raphe nucleus

The ability of (1DMe)Y8Fa (D.Tyr-Leu-(NMe)Phe-Gln-Pro-Gln-Arg-Phe-NH2), a selective neuropeptide FF analog resistant to enzymatic degradation, to control morphine-induced analgesia was investigated in rat after microinfusion into the dorsal raphe nucleus and the nucleus parafascicularis of the thalamus. Infusion of (1DMe)Y8Fa (2.5 nmol) in the nucleus raphe dorsalis did not modify the animal response in the tail-immersion test but significantly reversed analgesia induced by coinjected morphine (27 nmol). Similarly, (1DMe)Y8Fa (5 nmol) inhibited morphine effects in the hot-plate test after co-injection into the parafascicular nucleus. Furthermore, (1DMe)Y8Fa injected into the parafascicular nucleus attenuated analgesia induced by morphine injected into the nucleus raphe dorsalis and similarly, the neuropeptide FF analog in the nucleus raphe dorsalis decreased the effects of 27 nmol morphine injected in the parafascicular nucleus. The density of neuropeptide FF receptors did not decrease in the nucleus raphe dorsalis after lesion of serotonergic neurons by 5,7-dihydroxytryptamine. However, after this lesion, (1DMe)Y8Fa injected in the nucleus raphe dorsalis was no longer able to modify analgesic effects of morphine in hot-plate and tail-immersion tests. Similarly, the serotonin (5-HT) depletion induced by a systemic administration of para-chlorophenylalanine did not modify morphine analgesia microinjected into the nucleus raphe dorsalis and the parafascicular nucleus but blocked the ability of (1DMe)Y8Fa to reverse morphine effects in both nuclei. These data show that neuropeptide FF exerts anti-opioid effects directly into both the nucleus raphe dorsalis and the parafascicular nucleus and acts also at distance on opioid functions. Furthermore, anti-opioid effects of neuropeptide FF require functional serotonergic neurons although neuropeptide FF receptors are not carried on these neurons.

Ultrastructural localization of delta-1 opioid receptor in the dorsal raphe nucleus of the rat

The ultrastructural localization of delta-1 opioid receptor in the rat dorsal raphe nucleus was studied by the preembedding avidin-biotin-peroxidase complex technique. With application of a low concentration of the first antiserum in incubation and control of short-time reaction to 3,3'-diaminobenzidine, the immunoreaction seemed to be faint at the light microscopic level. At the electron microscopic level, however, delta-1 opioid receptor immunoreaction products were found to be localized specifically on the postsynaptic membrane of dendrites, dense-cored vesicles, and the surface of the small, clear vesicles in axon terminals with strong immunoreactivity. Of the total 659 immunopositive profiles observed, up to 62.4% (411/659) were dendrites, whereas 33.8% (223/659) were axon terminals. The immunostained myelinated axons and perikarya were relatively rare, with the frequencies 0.8% (5/659) and 3.0% (20/659), respectively. Most of the immunopositive dendrites (338/411, 82.2%) were immunostained only at the postsynaptic membranes. Other immunoreactive dendrites showed their immunoreaction products also in some other contents besides the postsynaptic membranes (44/411, 10.7%) or only in those contents but not the postsynaptic membranes (25/411, 6.1%). Only four dendrites showed their immunoreactive results only at the membrane not related to synapse (4/267, 1.0%). No dendrite was found immunostained in all the contents. About half of the immunopositive axon terminals (125/223, 56.1%) were found to make synapse with nonimmunoreactive dendrites (76/223, 34.1%) or immunoreactive dendrites (49/223, 22.0%), while only one was found to make contact with immunoreactive perikarya. The present study showed that delta-1 opioid receptor in the dorsal raphe nucleus is mostly localized on postsynaptic membrane; the main function of the delta-1 receptor in the dorsal raphe nucleus is to receive signals from the opioid-containing axon terminals through synapses.

Ultrastructural relationships of the pro-opiomelanocortin axons with the serotoninergic neurons in the dorsal raphe nucleus of the rat

The relationships of the corticotropin-like intermediate lobe peptide (CLIP)/ACTH-immunoreactive axons with the serotoninergic and non-serotoninergic neurons in the dorsal raphe nucleus of the rat were examined by means of a double label immunocytochemical method. It is suggested that the rare contacts established by the CLIP/ACTH-immunoreactive fibers with serotoninergic neurons (cell bodies and dendrites) are not under a synaptic from. In contrast, the contacts with non-serotoninergic neurons were predominantly formed with dendrites and showed a substantial number of synapses.

Noradrenergic input to nociceptive modulatory neurons in the rat rostral ventromedial medulla

Within the rostral ventromedial medulla (RVM), there are two classes of putative pain modulation neurons: ON cells and OFF cells, which respectively burst or pause prior to withdrawal reflexes elicited by noxious stimulation. Alpha-adrenergic agonists injected into the RVM produce changes in the latency of spinal nocifensive reflexes and, when iontophoretically applied, alter the firing of RVM ON but not OFF cells. To provide further information about the contribution of norepinephrine to RVM neuron function, we analyzed the distribution of tyrosine hydroxylase immunoreactive (TH-ir) appositions upon RVM ON and OFF cells. In the lightly anesthetized rat, seven ON and five OFF cells were identified by changes in their discharge rate in relation to nociceptive withdrawal reflexes and were labeled by intracellular injection of neurobiotin. Sections containing labeled cells were visualized by using avidin conjugated to a Texas Red fluorophore. Tissue with labeled cells was subsequently processed for TH-ir by using a Bodipy fluorophore conjugated secondary antibody. The distribution of the Bodipy-labeled fibers and terminals upon the Texas Red-labeled neurons was mapped using a confocal laser-scanning microscope. All the labeled neurons exhibited close TH-ir appositions. Appositions were of two types: swellings and fibers. Although the numbers and density of appositions varied among the cells, there were no consistent differences that correlated with physiological properties. Thus the overall density of appositions for ON cells (29.0 +/- 22.2 x 10(4) microns2) did not differ significantly from that for OFF cells (25.4 +/- 22.2 x 10(4) microns2). Tyrosine hydroxylase immunoreactive (TH-ir) appositions upon ON and OFF cells varied with their location along the dorso-ventral axis with more ventral neurons having a greater density of TH-ir swelling-type appositions. In a separate study, TH-ir and dopamine-beta-hydroxylase-like immunoreactivity (DBH-ir) were mapped in the same sections by using confocal microscopy. Nearly 97% of the TH-ir profiles co-localized with DBH-ir. These observations provide evidence that both ON and OFF cells in the RVM are targeted by noradrenergic inputs.

Electron microscopic study of GABAergic synaptic innervation of nitric oxide synthase immunoreactive neurons in the dorsal raphe nucleus in the rat

A double immunocytochemical method combining the preembedding avidin biotin peroxidase complex technique and the postembedding immunogold technique was used to examine synaptic interactions between GABAergic and nitric oxide synthase containing neurons in the same tissue sections of the dorsal raphe nucleus of the Wistar white rat. Although a large number of immunogold stained GABAergic axon terminals were found to be presynaptic to dendrites containing nitric oxide synthase-like immunoreaction product, synapses between GABA-like immunoreactive axon terminals and nitric oxide synthase-like immunoreactive perikarya were rare. The labeled boutons were found to make symmetrical and asymmetrical synapses. No axo-axonic synapse was found. These results suggest that GABAergic neurons could modulate nitric oxide producing neurons in the dorsal raphe nucleus through direct synaptic relations.

Raphe 5-HT1A autoreceptors, but not postsynaptic 5-HT1A receptors or beta-adrenoceptors, restrain the citalopram-induced increase in extracellular 5-hydroxytryptamine in vivo

In vivo microdialysis in rat ventral hippocampus was used (i) to verify the importance of 5-HT1A autoreceptors in the raphe as targets for drugs that enhance the citalopram-induced elevation of forebrain 5-hydroxytryptamine (5-HT), and (ii) to further examine the specificity of (-)-penbutolol in this regard. The selective 5-HT1A receptor antagonist WAY100635 (s.c., or intra-raphe) or the mixed 5-HT1A/1B/beta-adrenoceptor antagonist (-)-penbutolol (s.c.), potentiated the citalopram-induced 5-HT rise, whereas local "reverse' dialysis of WAY100635 into the ventral hippocampus did not. Furthermore, the (-)-penbutolol-induced augmentation proved stereoselective and not mediated by beta-adrenoceptors (no effect of s.c. (+)-penbutolol, or beta 1- and beta 2-adrenoceptor blockers (betaxolol, ICI118.551)). These data provide direct evidence that increased stimulation of 5-HT1A autoreceptors in the midbrain raphe impedes the effect of citalopram on forebrain extracellular 5-HT, whereas neither postsynaptic 5-HT1A receptors nor beta-adrenoceptors appear to be involved.

Electron microscopic study of GABAergic synaptic innervation of neurotensin-immunoreactive neurons in the dorsal raphe nucleus

A double immunocytochemical method combining the preembedding avidin-biotin-peroxidase-complex technique and the postembedding immunogold technique was used to examine synaptic interactions between GABAergic and neurotensin-containing neurons in the same tissue sections of the dorsal raphe nucleus of the rat. Whereas the neurotensin-like immunoreactive perikarya rarely received synapses from GABA-like immunostaining axon terminals, the neurotensin-like immunoreactive dendrites frequently received synapses from GABA-like immunoreactive neurons. These results suggest that GABAergic neurons could modulate neurotensinergic neurons in the dorsal raphe nucleus through synaptic relations. The immunocytochemically identified local synaptic circuit in the dorsal raphe was discussed.

The role of various calcium and potassium channels in the regulation of somatodendritic serotonin release

We prepared slices from midbrain containing the raphe nuclei and from hippocampus of rats. The brain slices were loaded with [3H]serotonin and superfused in order to measure the release of radioactivity at rest and in response to electrical stimulation. No difference was observed in the resting and stimulated fractional release of tritium in the somatodendritic and axon terminal parts of serotonergic neurons. The selective 5-HT1A receptor agonist 8-OH-DPAT decreased the electrically induced tritium efflux from raphe nuclei slices preloaded with [3H]serotonin, and this inhibition was reversed by the 5-HT1A receptor antagonist (+)WAY-100135. The 5-HT1B receptor agonist CGS-12066B but not 8-OH-DPAT, inhibited the stimulation-evoked tritium efflux from hippocampal slices after labeling with [3H]serotonin. The electrical stimulation-evoked tritium efflux in raphe nuclei slices incubated with [3H]serotonin was completely external Ca(2+)-dependent, and omega-conotoxin GVIA and Cd2+, but not diltiazem, inhibited the tritium overflow. In raphe nuclei slices 4-aminopyridine enhanced the electrical stimulation-induced tritium release in a concentration-dependent manner. The inhibition of tritium efflux by 8-OH-DPAT was abolished with 4-aminopyridine. Glibenclamide or tolbutamide proved to be ineffective. These data indicate that (1) different 5-HT receptor subtypes (5-HT1A and 5-HT1B) regulate dendritic and axon terminal 5-HT release; (2) serotonin release from the dendrites may be regulated by the voltage-sensitive N-type Ca2+ channels; (3) the 5-HT1A receptor-mediated inhibition of serotonin release may be due to opening of voltage-sensitive K+ channels.

Development of the raphe-hippocampal projection in vitro

In this study we examined whether the serotonergic raphe-hippocampal projection preserves its characteristic target selectivity for GABAergic interneurons when developing in vitro, in organotypic cultures. Hippocampal slices from one- to three-day-old rats were co-cultured with slices derived from the raphe nuclei of the same animals. After several weeks of in vitro incubation, a large number of raphe fibres--visualized by immunostaining for serotonin--were found to innervate the hippocampal tissue. In our random sample of over 250 serially sectioned boutons--52 of which were completely reconstructed from serial sections--only two were found to form conventional synapses in the electron microscope, and contacted dendritic spines. These results demonstrate that raphe-hippocampal serotonergic afferent are unable to form synaptic contacts with their normal targets in vitro, if explanted one to three days postnatally. Neurons in the afferent and/or target area may have passed a critical age when selective synaptic contacts can be formed, or unknown chemical or electrical signals may be missing under these conditions, which should serve to guide subcortical afferents to their synaptic target elements.

Changes in serotonin-immunoreactivity in the dorsal and median raphe nuclei of rats exposed to 2,4-dichlorophenoxyacetic acid through lactation

Comparison of serotonin-immunoreactive (SER-IR) neurons in nucleus raphe dorsalis (NRD) and median raphe nucleus (MRN) of 25-d-old rat pups exposed to 70 mg/kg/d 2,4-dichloro-phenoxyacetic acid through mothers milk and control pups was made using an immunohistochemical analysis. Significant 2,4-D-treatment-related increase in size and density of SER-IR neuronal somata as well as in fiber length were observed. We postulate that exposure to 2,4-dichlorophenoxyacetic acid on the first day of life would modify the synthesis of 5-HT or the maturation of the brain serotonergic system.

Neuropsychiatric features of Lewy body disease

Although traditionally associated with Parkinson's disease, the eosinophilic intracytoplasmic neuronal inclusion known as the Lewy body has recently been regarded as the primary neuropathologic finding in a variety of conditions affecting the aging brain. The term Lewy Body Disease (LBD) will be used in this review to refer to a spectrum of clinical states varying from those due to incidental or mildly symptomatic histopathologic changes to progressive dementia and psychosis. Many unanswered questions remain about the neurobehavioral and neuropathological implications of Lewy bodies, but it is useful to consider the LBD spectrum in terms of the variable effects on neuropsychiatric function that can be observed clinically.

Methylenedioxymethamphetamine induces opposite changes in central pre- and postsynaptic 5-HT1A receptors in rats

The present study examined the short- and long-term effects of single and repeated administration of 3,4-methylenedioxy-methamphetamine (MDMA, 'ecstasy') on somatodendritic and postsynaptic 5-HT1A receptors of the rat brain. [3H]8-Hydroxy-2-(di-n-propylamino)tetralin ([3H]8-OH-DPAT) was used to label 5-HT1A receptors in the brain stem region containing the dorsal raphe nucleus and in the frontal cortex. As expected, both schedules of treatment reduced the serotonin (5-hydroxytryptamine, 5-HT) content and [3H]paroxetine binding in the frontal cortex but not in the brain stem. Multiple but not single MDMA administration significantly reduced 5-HT1A receptor density in the selected brain stem region. In the frontal cortex, both MDMA treatments increased or tended to increase 5-HT1A receptor number, the effect being more marked after repeated drug administration.

Pattern and time course of immediate early gene expression in rat brain following acute stress

The pattern and time course of brain activation in response to acute swim and restraint stress were examined in the rat by in situ hybridization using complementary RNA probes specific for transcripts encoding the products of the immediate early genes c-fos, c-jun and zif/268. A widespread pattern of c-fos messenger RNA expression was detected in response to these stressors; surprisingly, the expression patterns were substantially similar following both swim and restraint stress. A dramatic induction of c-fos messenger RNA was observed in numerous neo- and allocortical regions, the lateral septal nucleus, the hypothalamic paraventricular and dorsomedial nuclei, the anterior hypothalamic area, the lateral portion of the retrochiasmatic area, the medial and cortical amygdaloid nuclei, the periaqueductal gray, and the locus coeruleus; however, a prominent induction of c-fos was also seen in numerous additional subcortical and brainstem regions. Although not as widely expressed in response to stress as c-fos, induction of zif/268 messenger RNA was also detected throughout many brain areas; these regions were largely similar to those in which c-fos was induced, although in a number of regions zif/268 was expressed in regions devoid of c-fos messenger RNA. Few brain areas showed increased expression of c-jun following stress; these regions also showed induction of c-fos and/or zif/268. The time courses of expression of all three immediate early genes were similar, with peak levels observed at the 30 or 60 min time point, and a markedly reduced signal evident at 120 min post-stress. However, in a number of cases a delayed and/or prolonged induction was noted that may be indicative of secondary neuronal activation. A number of recent studies have attempted to define neural pathways which convey stress-related information to the hypothalamic-pituitary-adrenal axis. The present results reveal a widespread pattern of neuronal activation in response to acute swim or restraint stress. These findings may aid in the identification of stress-specific neural circuits and are thus likely to have important implications for our understanding of neuronal regulation of the stress response.

Serotonin-containing projections to the mesencephalic trigeminal nucleus of the cat

BACKGROUND

It is well known that the mesencephalic trigeminal nucleus (MTN) neurons transmit somatosensory information from proprioceptors in the oral-facial region. Several mechanisms of sensory transduction in these specialized receptors have been proposed, but the neurotransmitters which are responsible for mediating proprioceptive information are still unknown. The current study concentrates on the distribution of one putative neurotransmitter system, serotonin (SER), in the cat MTN. A second objective was to clarify the location and sources of serotoninergic projections on the MTN neurons.

METHODS

To determine whether SER was localized in the MTN, the peroxidase-antiperoxidase (PAP) immunocytochemical technique was applied at light and electron microscopic levels in colchicine-treated animals. The origin of SER-containing fibers in the MTN was studied using a double-labeling method combining retrograde transport with wheat germ agglutinin conjugated to horseradish peroxidase (WGA-HRP) and SER immunocytochemistry.

RESULTS

There were no SER-containing neurons in the MTN. The cell bodies of immunonegative MTN neurons were closely surrounded by fine SER-positive fibers and terminals. The labeled fibers were in most cases very thin and sometimes varicose. Ultrastructurally, direct synaptic contacts between SER-containing terminals and perikarya of MTN neurons of all sizes could be seen. The majority of SER-labeled structures were synaptic terminals in which the immunoreactive material was located within the small round clear as well as the small granular vesicles (diameter 50-80 nm) and a few large dense-cored vesicles (up to 150 nm). Retrograde tracing demonstrated that most of cells in the nuclei raphe dorsalis, pontis and magnus were WGA-HRP-labeled.

CONCLUSIONS

These results indicated that MTN neurons received serotoninergic projections from the raphe nuclei of the brainstem. In light of these morphological data, it is concluded that the MTN of the cat is under the influence of SER-containing axons and this serotoninergic input may modulate MTN neuronal activity at the first synaptic relay.

Synaptic relations of neurotensinergic neurons in the dorsal raphe nucleus

The ultrastructure and synaptic relations of neurotensinergic neurons in the rat dorsal raphe nucleus (DRN) were examined. The neurotensin-like immunoreactive (NT-L1) neurons in the DRN were fusiform or spherical. The NT-LI perikarya could only be detected in colchicine-treated animals whereas the immunoreactive axon terminals could only be found in the animals not treated with colchicine. Although many NT-LI dendrites received synapses from nonimmunoreactive axon terminals, the NT-LI perikarya received few synapses. NT-LI axon terminals also made synapses on nonimmunoreactive dendrites. Occasionally, synapses were found between the NT-LI axon terminals and NT-LI dendrites in the cases in which the animals were not treated with colchicine.

Developmental changes in the brain-stem serotonergic nuclei of teleost fish and neural plasticity

1. During early ontogeny, the serotonergic neurons in the brain stem of the three-spined stickleback shows a temporal and spatial developmental pattern that closely resembles that of amniotes. 2. However, in the adult fish, only the midline nuclei of the rostral group (dorsal and median raphe nuclei) and the dorsal lateral tegmental nucleus are consistently serotonin-immunoreactive (5-HTir), whereas the groups of the upper and lower rhombencephalon (raphe pontis, raphe magnus, and raphe pallidus/obscurus nuclei) are variable and, when present, contain relatively small numbers of 5-HTir neurons. 3. Using specific antisera against tryptophan 5-hydroxylase and aromatic L-amino acid decarboxylase, we have shown that the lateral B9 group and the groups of the upper and lower rhombencephalon are consistently present in adult sticklebacks. The results are discussed in relation to other known instances of neurotransmitter plasticity or transient neurotransmitter expression in teleost fish. 4. While there are several instances of transient expression of neurotransmitter markers by discrete neuronal populations, there is so far no evidence of changes from one neurotransmitter phenotype to another in the brain of teleost fish. However, there are indications of plasticity of expression of catecholamines and indoleamines, and their respective synthesizing enzymes, as reflected in age-dependent changes and variation between individuals of different physiological status. 5. As the brain grows continuously in teleost fish, and new neurons are added from proliferative regions, synaptic connections may be expected to undergo remodeling in all brain regions throughout life. Thus, the teleostean brain may be considered a suitable model for experimental studies of different aspects of neural plasticity.

Chronic treatment with citalopram facilitates the effect of a challenge dose on cortical serotonin output: role of presynaptic 5-HT1A receptors

In animals given citalopram (10 mg/kg) twice daily for 14 days a further dose of 1 mg/kg, administered 24 h after the last dose, markedly increased cortical dialysate serotonin (5-hydroxytryptamine, 5-HT), but had no effect in control animals. The effect on dialysate 5-HT in the dorsal raphe was not increased by the chronic treatment. At 25 micrograms/kg, 8-hydroxy-2-(di-n-propylamino)tetralin, an agonist at 5-HT1A receptors, reduced cortical 5-HT output in controls but not in animals treated chronically with citalopram whereas 50 micrograms/kg reduced 5-HT output in both groups. These findings suggest that somatodendritic 5-HT1A receptors are desensitized after chronic treatment with citalopram and this results in facilitation of its effect on cortical dialysate 5-HT. These results also agree with the concept that the effect of 5-HT re-uptake inhibitors on increasing 5-HT output in the frontal cortex is attenuated by their simultaneous ability to activate somatodendritic 5-HT1A receptors via an increase of endogenous 5-HT in the raphe region.

Immunocytochemical study of the CLIP/ACTH-immunoreactive nerve fibres in the dorsal raphe nucleus of the rat

The injection of corticotrophin-like intermediate lobe peptide (CLIP) in the lateral ventricle or the dorsal raphe nucleus (DRN) of the rat is followed by a significant increase in the amount of paradoxical sleep. In the DRN, CLIP would act through a somatic and/or dendritic release of serotonin. To establish the anatomical basis of these effects, the nerve fibres immunoreactive for CLIP/ACTH were labelled and their fine anatomical relationships with the neuronal elements of the DRN were studied at the ultrastructural level. Half of the profiles of labelled varicosities were 'free' in the neuropile, the other half was in close contact with dendrites, either shafts or spines. It was interesting to note that few contacts with dendritic shafts were under the form of a synapse, whereas a large number of the contacts with spines were synaptic. The chemical nature of these dendritic targets remains to be determined but preliminary results indicate that they are partly serotoninergic.