Cloning of two tryptophan hydroxylase genes expressed in the diencephalon of the developing zebrafish brain

The monoamine serotonin (5-HT) exerts key neuromodulatory activities in all animal phyla, but the development and function of the serotonergic system is still incompletely understood. The zebrafish Danio rerio is an excellent model to approach this question since it is amenable to a combination of genetic, molecular and embryological studies. In order to characterize the organization of serotonergic neurons in the zebrafish we cloned two cDNAs encoding distinct forms of tryptophan hydroxylase (Tph), the rate-limiting enzyme in serotonin synthesis. We report here the pattern of expression of these two genes in relation with immunoreactive TH and 5-HT nuclei in the developing zebrafish embryo and early larva. tphD1 expression starts at 22 h post-fertilization (hpf) in the epiphysis and in basal spinal cells. Expression persists in the epiphysis until at least 4 days (dpf). Between 48 hpf and 3 dpf, tphD1 expression is initiated in retinal amacrine cells and in restricted preoptic and posterior tubercular nuclei within the basal diencephalon. At 3 and 4 dpf, tphD1 expression is newly initiated in the caudal hypothalamus and in branchial arches-associated neurons. tphD2 mRNA is detected transiently (between 30 somites and 32 hpf) in a restricted preoptic nucleus. All sites of tphD1 or D2 expression within the anterior central nervous system are also immunoreactive for 5-HT, but are not positive for TH. However, neither tphD gene is expressed in raphe nuclei, suggesting that additional tph gene(s) exist in zebrafish to account for 5-HT synthesis in that location. The co-expression of tphD1, tphD2 and 5-HT in the zebrafish diencephalon appears in striking contrast to the situation in mammals, where diencephalic serotonin results from re-uptake rather than from local production.     

Cloning of two tryptophan hydroxylase genes expressed in the diencephalon of the developing zebrafish brain

The monoamine serotonin (5-HT) exerts key neuromodulatory activities in all animal phyla, but the development and function of the serotonergic system is still incompletely understood. The zebrafish Danio rerio is an excellent model to approach this question since it is amenable to a combination of genetic, molecular and embryological studies. In order to characterize the organization of serotonergic neurons in the zebrafish we cloned two cDNAs encoding distinct forms of tryptophan hydroxylase (Tph), the rate-limiting enzyme in serotonin synthesis. We report here the pattern of expression of these two genes in relation with immunoreactive TH and 5-HT nuclei in the developing zebrafish embryo and early larva. tphD1 expression starts at 22 h post-fertilization (hpf) in the epiphysis and in basal spinal cells. Expression persists in the epiphysis until at least 4 days (dpf). Between 48 hpf and 3 dpf, tphD1 expression is initiated in retinal amacrine cells and in restricted preoptic and posterior tubercular nuclei within the basal diencephalon. At 3 and 4 dpf, tphD1 expression is newly initiated in the caudal hypothalamus and in branchial arches-associated neurons. tphD2 mRNA is detected transiently (between 30 somites and 32 hpf) in a restricted preoptic nucleus. All sites of tphD1 or D2 expression within the anterior central nervous system are also immunoreactive for 5-HT, but are not positive for TH. However, neither tphD gene is expressed in raphe nuclei, suggesting that additional tph gene(s) exist in zebrafish to account for 5-HT synthesis in that location. The co-expression of tphD1, tphD2 and 5-HT in the zebrafish diencephalon appears in striking contrast to the situation in mammals, where diencephalic serotonin results from re-uptake rather than from local production.     

Optogenetic Recruitment of Dorsal Raphe Serotonergic Neurons Acutely Decreases Mechanosensory Responsivity in Behaving Mice

The inhibition of sensory responsivity is considered a core serotonin function, yet this hypothesis lacks direct support due to methodological obstacles. We adapted an optogenetic approach to induce acute, robust and specific firing of dorsal raphe serotonergic neurons. In vitro, the responsiveness of individual dorsal raphe serotonergic neurons to trains of light pulses varied with frequency and intensity as well as between cells, and the photostimulation protocol was therefore adjusted to maximize their overall output rate. In vivo, the photoactivation of dorsal raphe serotonergic neurons gave rise to a prominent light-evoked field response that displayed some sensitivity to a 5-HT1A agonist, consistent with autoreceptor inhibition of raphe neurons. In behaving mice, the photostimulation of dorsal raphe serotonergic neurons produced a rapid and reversible decrease in the animals'' responses to plantar stimulation, providing a new level of evidence that serotonin gates sensory-driven responses.     

Serotonin receptor activation reduces calcium current in an acutely dissociated adult central neuron

The release of serotonin (5-HT) from the terminals of serotonergic (raphe) neurons is under inhibitory feed-back control. 5-HT, acting on raphe cell body autoreceptors, also mediates inhibitory postsynaptic potentials as a result of release from collaterals from neighboring raphe neurons. This may involve a ligand (5-HT)-gated increase in the membrane potassium conductance, leading to a decrease in action potential frequency, which could indirectly reduce calcium influx into nerve terminals. In this report we demonstrate that 5-HT can also directly reduce calcium influx at potentials including and bracketing the peak of calcium current activation. Using acutely isolated, patch-clamped dorsal raphe neurons, we found that low concentrations of 5-HT and the 5-HT1A-selective agonist 8-OH-DPAT reversibly decrease whole-cell calcium current. This effect is antagonized by the putative 5-HT1A-selective antagonist NAN 190. Hence, the inhibition of calcium current may serve a physiological role in these cells and elsewhere in the brain.     

Severe deficits in 5-HT2A -mediated neurotransmission in BDNF conditional mutant mice

BDNF is thought to provide critical trophic support for serotonin neurons. In order to determine postnatal effects of BDNF on the serotonin system, we examined a line of conditional mutant mice that have normal brain content of BDNF during prenatal development but later depletion of this neurotrophin in the postnatal period. These mice show a behavioral phenotype that suggests serotonin dysregulation. However, as shown here, the presynaptic serotonin system in the adult conditional mutant mice appeared surprisingly normal from histological, biochemical, and electrophysiological perspectives. By contrast, a dramatic and unexpected postsynaptic 5-HT2A deficit in the mutant mice was found. Electrophysiologically, serotonin neurons appeared near normal except, most notably, for an almost complete absence of expected 5-HT2A -mediated glutamate and GABA postsynaptic potentials normally displayed by these neurons. Further analysis showed that BDNF mutants had much reduced 5-HT2A receptor protein in dorsal raphe nucleus and a similar deficit in prefrontal cortex, a region that normally shows a high level of 5-HT2A receptor expression. Recordings in prefrontal slice showed a marked deficit in 5-HT2A -mediated excitatory postsynaptic currents, similar to that seen in the dorsal raphe. These findings suggest that postnatal levels of BDNF play a relatively limited role in maintaining presynaptic aspects of the serotonin system and a much greater role in maintaining postsynaptic 5-HT2A and possibly other receptors than previously suspected.     

Comparative distribution of nitric oxide synthase- and serotonin-containing neurons in the raphe nuclei of four mammalian species

 Monoclonal antibodies were generated against serotonin (5-HT) and the C-terminal portion of the neuronal form of nitric oxide synthase (nNOS), the enzyme producing nitric oxide in neurons. These antibodies were used to compare the distribution of 5-HT- and nNOS-containing neurons in the raphe nuclei of four animal species (rat, mouse, guinea pig, and cat). It was found that the rat was the only species in which the raphe nuclei contain a substantial number of nNOS-immunoreactive (IR) cell bodies. In this species and as observed by other authors, all mesencephalic raphe nuclei contained nNOS-IR cells, the largest group being located in the nucleus raphe dorsalis. The coexistence of nNOS and 5-HT immunoreactivities in these nuclei was visualized by double labeling. In the medulla, the nuclei raphe magnus and obscurus displayed a rather low number of nNOS-IR neurons. In the other species, nNOS-IR cell bodies were found in very low numbers, whatever raphe nucleus was considered. The rostral pole of the nucleus raphe dorsalis and the nuclei raphe magnus and obscurus contained a few nNOS-IR neurons which did not show any coincidence with the 5-HT neurons. In addition, nNOS-IR axons were rare. It is concluded that in the mouse, guinea pig, and cat the involvement of nitric oxide in functions subserved by 5-HT within the raphe nuclei might be minimal. Accepted: 5 May 1998     

Lack of CB1 receptor activity impairs serotonergic negative feedback

Serotonergic and endocannabinoid systems are important substrates for the control of emotional behaviour and growing evidence show an involvement in the pathophysiology of mood disorders. In the present study, the absence of the activity of the CB₁ cannabinoid receptor impaired serotonergic negative feedback in mice. Thus, in vivo microdialysis experiments revealed increased basal 5-HT extracellular levels and attenuated fluoxetine-induced increase of 5-HT extracellular levels in the prefrontal cortex of CB₁ knockout compared with wild-type mice. These observations could be related to the significant reduction in the 5-HT transporter binding site density detected in frontal cortex and hippocampus of CB₁ knockout mice. The lack of CB₁ receptor also altered some 5-HT receptors related to the 5-HT feedback. Extracellular recordings in the dorsal raphe nucleus (DRN) revealed that the genetic and pharmacological blockade of CB₁ receptor induced a 5-HT_1A autoreceptor functional desensitization. In situ hybridization studies showed a reduction in the expression of the 5-HT_2C receptor within several brain areas related to the control of the emotional responses, such as the DRN, the nucleus accumbens and the paraventricular nucleus of the hypothalamus, whereas an over-expression was observed in the CA3 area of the ventral hippocampus. These results reveal that the lack of CB₁ receptor induces a facilitation of the activity of serotonergic neurons in the DRN by altering different components of the 5-HT feedback as well as an increase in 5-HT extracellular levels in the prefrontal cortex in mice.     

Distribution of serotonin immunoreactive neurons in the brainstem of the hamster, guinea pig, rabbit, and rat

Immunohistochemical techniques were employed to study the distribution of serotonin (5-HT) immunoreactive neurons in the brainstem of the hamster, guinea pig, rabbit and rat. 5-HT neurons were principally found to be concentrated in the midline raphe nuclei, particularly, the raphe pallidus, raphe obscurus, raphe magnus, raphe median, raphe pontis and raphe dorsalis nuclei. Characteristically, these cell bodies are displayed in bands or wing-like patterns which extend laterally from the raphe into reticular formations. The formations often appear to blend with the catecholamine system. They are particularly evident in the brainstems of the rabbit and hamster which contain wider and more lateral extensions of the serotonergic (5-HT) neurons than those observed in the brainstems of the rat and guinea pig. The widespread distribution of 5-HT immunoreacted cell bodies in the brainstem shows that there are significant prospects of 5-HT in neuronal activities.     

Tryptamine Concentrations in Areas of 5-Hydroxytryptamine Terminal Innervation After Electrolytic Lesions of Midbrain Raphe Nuclei

The possible existence of tryptamine-containing neurons originating in the midbrain raphe is suggested by several reports of tryptamine-mediated responses to electrical stimulation of the raphe nuclei. To assess this hypothesis, we have investigated the effects of electrolytic lesions of the median and dorsal raphe nuclei on striatal, hypothalamic, and hippocampal concentrations of tryptamine, 5-hydroxytryptamine (5-HT), and 5-hydroxyindoleacetic acid. In addition, the rat striatal tryptophan concentrations were also determined. No changes in the concentrations of tryptamine were observed at 1 or 2 weeks after lesioning the dorsal and median raphe nuclei, at which time the other 5-hydroxyindoles were markedly reduced; furthermore, no reductions were observed in tryptamine concentrations in the striatum, hypothalamus, or hippocampus of rats pretreated with a monoamine oxidase inhibitor. The only change observed in these rats was a limited increase in striatal tryptamine and tryptophan observed at 1 day after lesioning. The results indicate that tryptamine concentration is independent of the integrity of 5-HT-containing neurons of the midbrain raphe nuclei. Furthermore, if tryptamine-containing neurons that have terminal projections to the striatum, hypothalamus, and hippocampus exist, their cell bodies are located in regions outside the dorsal and median raphe nuclei. Another possibility could be that tryptamine is located in glial cells.     

Altered serotonin and dopamine metabolism in the CNS of serotonin 5-HT(1A) or 5-HT(1B) receptor knockout mice

Measurements of serotonin (5-HT), dopamine (DA), and noradrenaline, and of 5-HT and DA metabolites, were obtained by HPLC from 16 brain regions and the spinal cord of 5-HT(1A) or 5-HT(1B) knockout and wild-type mice of the 129/Sv strain. In 5-HT(1A) knockouts, 5-HT concentrations were unchanged throughout, but levels of 5-HT metabolites were higher than those of the wild type in dorsal/medial raphe nuclei, olfactory bulb, substantia nigra, and locus coeruleus. This was taken as an indication of increased 5-HT turnover, reflecting an augmented basal activity of midbrain raphe neurons and consequent increase in their somatodendritic and axon terminal release of 5-HT. It provided a likely explanation for the increased anxious-like behavior observed in 5-HT(1A) knockout mice. Concomitant increases in DA content and/or DA turnover were interpreted as the result of a disinhibition of DA, whereas increases in noradrenaline concentration in some territories of projection of the locus coeruleus could reflect a diminished activity of its neurons. In 5-HT(1B) knockouts, 5-HT concentrations were lower than those of the wild type in nucleus accumbens, locus coeruleus, spinal cord, and probably also several other territories of 5-HT innervation. A decrease in DA, associated with increased DA turnover, was measured in nucleus accumbens. These changes in 5-HT and DA metabolism were consistent with the increased aggressiveness and the supersensitivity to cocaine reported in 5-HT(1B) knockout mice. Thus, markedly different alterations in CNS monoamine metabolism may contribute to the opposite behavioral phenotypes of these two knockouts.     

Differential Effects of Ipsapirone on 5-Hydroxytryptamine Release in the Dorsal and Median Raphe Neuronal Pathways

Abstract: Serotonergic neurons of the dorsal and median raphe nuclei are morphologically dissimilar. Recent results challenge previous evidence indicating a greater inhibition of dorsal raphe neurons after 5-hydroxytryptamine_1A (5-HT_1A) autoreceptor activation. As both nuclei innervate different forebrain territories, this issue is critical to understanding the changes in brain function induced by anxiolytic and antidepressant drugs. Using microdialysis, we examined the modifications of 5-HT release induced by the selective 5-HT_1A agonist ipsapirone in both neuronal pathways. Maximal and minimal basal 5-HT values (in the presence of 1 µ M citalopram) were 45.0 ± 4.8 fmol/fraction in the median raphe nucleus and 8.4 ± 0.4 fmol/fraction in the dorsal hippocampus. Ipsapirone (0.3, 3, and 10 mg/kg s.c.) reduced dose-dependently 5-HT in the two raphe nuclei and four forebrain areas. Maximal reductions (to ∼25% of predrug values) were observed in cortex and striatum and in median raphe nucleus. The effects were more moderate in dorsal and ventral hippocampus (to 66 and 50% of baseline, respectively). These results are consistent with a higher sensitivity of dorsal raphe neurons to 5-HT_1A autoreceptor activation. Yet the differential reduction of 5-HT release in the median raphe nucleus and hippocampus suggests the presence of complex mechanisms of control of 5-HT release in these neurons.     

p-Chlorophenylalanine Changes Serotonin Transporter mRNA Levels and Expression of the Gene Product

Abstract: After a single intraperitoneal injection of the irreversible tryptophan hydroxylase inhibitor p -chlorophenylalanine (PCPA; 300 mg/kg), there was a rapid down-regulation of serotonin (5-HT) transporter mRNA levels in cell bodies. This change was significant at 1 and 2 days after PCPA administration within the ventromedial but not the dorsomedial portion of the dorsal raphe nucleus. Seven days after PCPA treatment, 5-HT transporter mRNA levels were significantly elevated compared with controls in both regions of the dorsal raphe nucleus. PCPA administration produced no change in the [³H]-citalopram binding and synaptosomal [³H]5-HT uptake in terminal regions at 2 and 7 days after treatment but significantly reduced both these parameters by ∼20% in the hippocampus and in cerebral cortex 14 days after PCPA administration. The striatum showed a lower sensitivity to this effect. No significant changes were observed in the levels of [³H]citalopram binding to 5-HT cell bodies in the dorsal raphe nucleus. In the same animals used for 5-HT transporter mRNA level measurements, levels of tryptophan hydroxylase mRNA in neurons of the ventromedial and dorsomedial portions of the dorsal raphe nucleus were increased 2 days after PCPA administration and fell to control levels 7 days after injection in the ventromedial region but not in the dorsomedial portion of the dorsal raphe nucleus, where they remained significantly higher than controls. Altogether, these results show that changes in 5-HT transporter mRNA are not temporally related to changes in 5-HT transporter protein levels. In addition, our results suggest that the 5-HT transporter and tryptophan hydroxylase genes are regulated by different mechanisms. We also provide further evidence that dorsal raphe 5-HT neurons are differentially regulated by drugs, depending on their location.     

Serotonergic Modulation of Photic Entrainment in the Syrian Hamster

In this review, we describe six lines of evidence that reveal a modulatory role for serotonin (5-HT) in the regulation of the response of suprachiasmatic nucleus (SCN) neurons to retinal illumination in the Syrian hamster. Electrical stimulation of the median raphe nucleus, sufficient to elicit the release of 5-HT in the SCN, inhibits light-induced phase shifts of the hamster circadian activity rhythm. Two 5-HT receptors capable of mediating the effects of 5-HT on photic responses, the 5-HT₇ receptor and the 5-HT_1B receptor, are present in the hamster SCN. Light-induced phase shifts are attenuated by systemic and local administration of two 5-HT receptor agonists, 8-OH-DPAT, and TFMPP, and these agents attenuate photic phase shifts by acting on pharmacologically distinct receptors. Furthermore, both compounds also attenuate light-induced Fos expression and photic suppression of pineal melatonin content, indicating that serotonergic modulation of photic signal transduction in the SCN is not limited to the regulation of circadian phase. Finally, both 8-OH-DPAT and TFMPP inhibit RHT neurotransmission in the hypothalamic slice preparation. Further, TFMPP fails to attenuate responses to exogenous glutamate on retinorecipient SCN neurons, consistent with a presynaptic site of action for the drug. Based on these data, we propose that 5-HT modulates RHT neurotransmission in the SCN through at least two distinct mechanisms: (1) via activation of 5-HT₇ receptors probably located on retinorecipient neurons; and (2) via activation of presynaptic 5-HT_1B receptors leading to reduced release of glutamate from RHT terminals in the SCN.     

Regulation of release processes in central serotoninergic neurons

Different technical, physiological and biochemical aspects concerning the study of the release of 5-HT are discussed herein. Isotopic methods are the most suitable techniques since these allow the release of 3H-5-HT to be measured after having determined the identity of the labelled compounds formed from 3H-tryptophan by co-chromatography. Under these conditions, the 3H-amine released in the superfusates comes from serotoninergic nerve endings, since tryptophan hydroxylase is exclusively localized in serotoninergic neurons. Moreover, it appears that newly synthesized 5-HT is preferentially released. The release of 5-HT is dependent on neuronal activity, but is not always linked to the synthesis of 5-HT. The increase in the firing rate of serotoninergic cell bodies by a local application of glutamate in the area of the nucleus raphe dorsalis induces a marked increase n the release of 5-HT in the caudate nucleus; an opposite effect is observed after cooling this region. The local depolarization of serotoninergic terminals located in the caudate nucleus increases the release of this amine. This effect is blocked by TTX. LSD reduces the stimulating effect of KCl, thus indicating that the release of 5-HT can be controlled at a presynaptic level. In addition, the release of the amine is dependent on the presence of calcium. Serotoninergic neuronal activity can be controlled at the preterminal or at the cell body levels by the activity of other neuronal systems. The effects of the release of dopamine from dendrites, and that of GABA in the substantia nigra are reported herein. Furthermore, changes in the activity of the dopaminergic, gabaergic and serotoninergic systems innervating the nucleus raphe dorsalis modulate the release of 5-HT, measured both in the caudate nucleus and in the nucleus raphe magnus. Finally, it has been reported that the release of 5-HT can be estimated in the raphe nuclei dorsalis and magnus. It has been shown that the amounts of 3H-5-HT continuously formed from 3H-TRP and released in the nucleus raphe dorsalis are much greater than those estimated in the caudate nucleus or in the substantia nigra. Although the quantities of endogenous 5-HT measured in the nucleus raphe dorsalis are the highest in the brain, this structure presents only a few serotoninergic nerve endings. This raises the question of the origin of the 5-HT released in serotoninergic nuclei. A possible dendritic release of 5-HT is discussed.     

Blockade of substance P (neurokinin 1) receptors enhances extracellular serotonin when combined with a selective serotonin reuptake inhibitor: an in vivo microdialysis study in mice

Abstract Substance P antagonists of the neurokinin-1 receptor type (NK1) are gaining growing interest as new antidepressant therapies. It has been postulated that these drugs exert this putative therapeutic effect without direct interactions with serotonin (5-HT) neurones. Our recent microdialysis experiment performed in NK1 receptor knockout mice suggested evidence of changes in 5-HT neuronal function (Froger et al. 2001). The aim of the present study was to evaluate the effects of coadministration of the selective 5-HT reuptake inhibitor (SSRI) paroxetine with a NK1 receptor antagonist (GR205171 or L733060), given either intraperitoneally (i.p.) or locally into the dorsal raphe nucleus, on extracellular levels of 5-HT ([5-HT]ext) in the frontal cortex and the dorsal raphe nucleus using in vivo microdialysis in awake, freely moving mice. The systemic or intraraphe administration of a NK1 receptor antagonist did not change basal cortical [5-HT]ext in mice. A single systemic dose of paroxetine (4 mg/kg; i.p.) resulted in a statistically significant increase in [5-HT]ext with a larger extent in the dorsal raphe nucleus (+ 138% over basal AUC values), than in the frontal cortex (+ 52% over basal AUC values). Co-administration of paroxetine (4 mg/kg; i.p.) with the NK1 receptor antagonists, GR205171 (30 mg/kg; i.p.) or L733060 (40 mg/kg; i.p.), potentiated the effects of paroxetine on cortical [5-HT]ext in wild-type mice, whereas GR205171 (30 mg/kg; i.p.) had no effect on paroxetine-induced increase in cortical [5-HT]ext in NK1 receptor knock-out mice. When GR205171 (300 micro mol/L) was perfused by 'reverse microdialysis' into the dorsal raphe nucleus, it potentiated the effects of paroxetine on cortical [5-HT]ext, and inhibited paroxetine-induced increase in [5-HT]ext in the dorsal raphe nucleus. Finally, in mice whose 5-HT transporters were first blocked by a local perfusion of 1 micro mol/L of citalopram into the frontal cortex, a single dose of paroxetine (4 mg/kg i.p.) decreased cortical 5-HT release, and GR205171 (30 mg/kg i.p.) reversed this effect. The present findings suggest that NK1 receptor antagonists, when combined with a SSRI, augment 5-HT release by modulating substance P/5-HT interactions in the dorsal raphe nucleus.     

A tryptophan hydroxlyase 1 reporter that directs Cre recombinase extinguishable placental alkaline phosphatase expression in serotonergic (5-HT) neurons and peripheral tissues

The serotonergic (5-HT) system modulates many behaviors and has been implicated in psychiatric disorders, but the density of 5-HT processes has complicated analyses. We have used regulatory regions from the Tryptophan hydroxylase 1 (Tph1) gene to drive expression of LoxP-flanked placental alkaline phosphatase (PLAP) to generate the Tph1-Lox-PLAP reporter mouse line. In these mice, PLAP is expressed in the hindbrain raphe nuclei and in peripheral tissues known to express Tph1. Tph1 is expressed at low levels in neurons. While, in Tph1-Lox-PLAP mice, most PLAP-expressing neurons are monoaminergic, PLAP was expressed in only 5-10% of neurons expressing the predominant neuronal 5-HT biosynthetic enzyme Tph2, serotonin transporter (SERT) or aromatic amino acid decarboxylase (AADC). To test this reporter further, we examined the brains of mice carrying the anorexia (anx) mutation, in which increased overall density of 5-HT immunoreactivity had been previously observed at P21. PLAP-labeling of processes in anx/anx and anx/+ mice was reduced at P0. By P10, distribution of PLAP-labeled processes in anx/+ and +/+ cortices was indistinguishable, but differed markedly from that seen in the cortical layers of anx/anx mice. Thus, the Tph1-LoxP-PLAP reporter revealed a dosage sensitive role of the anx mutation in the early 5-HT system and later cortical layer-specific differences in 5-HT process distribution in anx/anx mice. Thus, the Tph1-LoxP-PLAP reporter provides a sensitive indicator for analyses of serotonergic cells in the brain and periphery.