Serotonin (5-HT) plays a major role at the spinal level by modulating most spinal functions through several receptor subtypes including the 5-HT2A receptor. To gain further insight into the cellular role of this receptor, we performed an immunocytochemical study of 5-HT2A receptors in the rat spinal cord, at light and electron microscope levels. The results showed that 5-HT2A receptors were widely distributed in the spinal cord at all segmental levels. Immunolabeling was particularly dense in lamina IX and in the dorsal horn lamina IIi. Immunoreactive cell bodies were numerous in lamina IX, where many but not all motoneurons were labeled, as shown by double labeling with choline acetyltransferase antibodies. Stained cell bodies were also observed in the gray matter. The study at the ultrastructural level focused on the lumbar dorsal horn (laminae I-II) and ventral horn (lamina IX). At both levels, 5-HT2A immunoreactivity was mainly postsynaptic on dendrites and cell bodies. However, a little presynaptic labeling was also observed in axon and axon terminals, some of them containing large granular vesicles attesting to their peptidergic nature. The main result of our study was the "nonsynaptic" plasma membrane localization of 5-HT2A receptors covering a large surface of cell bodies and dendrites, suggesting a paracrine form of action of serotonin. These observations are consistent with a double role (pre- and postsynaptic) for serotonin on these receptors on various cellular targets.

The distribution of 5-hydroxytryptamine1A and 5-hydroxytryptamine1B receptors in the visual cortex was studied by quantitative autoradiography during postnatal development. Overall, receptor densities increased throughout development, but exhibited regional rearrangements, particularly in the case of 5-hydroxytryptamine1B receptors. Neonatal treatment with 5,7-dihydroxytryptamine, which causes selective degeneration of serotoninergic neurons, had no effect on the density of 5-hydroxytryptamine1A receptors in the visual cortex. However, a transient increase in 5-hydroxytryptamine1B at postnatal days 10-12 was observed after this treatment, suggesting a regulation of postsynaptic receptors. Neonatal enucleation resulted in a marked increase in 5-hydroxytryptamine1B binding sites in all layers of the visual cortex by P16, whereas it had no effect upon 5-hydroxytryptamine1A binding sites. These results show that both receptor subtypes do not exhibit striking transient features in the visual cortex during postnatal development, but rather undergo discrete reorganizations. 5-Hydroxytryptamine1B receptors show changes in density after either neonatal degeneration of serotoninergic neurons or enucleation, indicating that the serotoninergic system involving this receptor subtype can exhibit some postnatal plasticity in the visual cortex.

The possible colocalization of 5-hydroxytryptamine1A receptors and choline acetyltransferase in the same neurons of the medial septum and diagonal band of Broca was investigated using double immunocytochemical techniques, either on the same section or on adjacent thin sections of the rat brain. The presence of both antigens in the same neurons was demonstrated at the light and electron microscopic levels. The proportion of cholinergic neurons that express 5-hydroxytryptamine1A receptors was similar in the different parts of the septal complex (around 25%). By contrast, the proportion of 5-hydroxytryptamine1A receptor-positive neurons also exhibiting choline acetyltransferase immunoreactivity was much higher (40–44%) in the dorsal and ventral groups of cholinergic cells, than in the intermediate group (18%).

Specific anti-rat 5-hydroxytryptamine1A (serotonin1A) receptor antibodies raised in a rabbit injected with a synthetic peptide corresponding to a highly selective portion of the third intracellular loop of the receptor protein (El Mestikawy et al. [1990] Neurosci. Lett. 118:189-192) were used for immunohistochemical mapping of serotonin1A receptors in the brain and spinal cord of adult rats. The highest density of immunostaining was found in limbic areas (lateral septum, CA1 area of Ammon's horn and dentate gyrus in the hippocampus, and frontal and entorhinal cortices), in the anterior raphe nuclei, and in the interpeduncular nucleus, in agreement with previous autoradiographic studies with selective radioligands showing the enrichment of these regions in serotonin1A receptor binding sites. Serotonin1A receptor-like immunoreactivity was also present, but at a moderate level, in the neocortex, in some thalamic and hypothalamic nuclei, in the nucleus of the solitary tract, in the dorsal tegmentum, in the nucleus of the spinal tract of the trigeminal nerve, and in the superficial layers of the dorsal horn in the spinal cord. In contrast, extrapyramidal areas, including the caudate putamen, the globus pallidus, and the substantia nigra as well as the cerebellum, exhibited very low to no immunostaining by antiserotonin1A receptor antibodies. At the cellular level, both the plasma membrane of neuronal perikarya and fine neuronal processes probably corresponding to dendritic fields were found to bind antiserotonin1A receptor antibodies. Regional differences were noted regarding these two types of immunostaining, because only dendrites bound antibodies within the hippocampus and the lateral septum, whereas both dendrites and neuronal cell bodies were immunoreactive in the medial septum, in the diagonal band of Broca, and in the dorsal and median raphe nuclei. Therefore, differential addressing of serotonin1A receptors could occur from one neuron to another. In general, the distribution and density of serotonin1A receptor-like immunoreactivity in the whole brain and in spinal cord were consistent with the mapping of serotonin1A receptor binding sites and serotonin1A receptor mRNA previously established by immunoautoradiographic and in situ hybridization procedures.

Modifications in serotonin (5-HT) neurotransmission have been associated with the physiopathology of anxiety and depression. Among the numerous 5-HT receptor subtypes, several (5-HT1A, 5-HT1B, 5-HT2 and 5-HT3) could be involved in these etiologies. By using a murine genetic model, we attempted to correlate variations in the density of receptor subtypes with modifications of anxiety-related behaviors. From a classic inbred strain (C57BL/6ByJ) and a linkage-testing inbred strain (ABP/Le), segregated F-2 populations for 3 loci located in the 4th, 7th and 9th chromosomes have been selected for their different responses in anxiety-related behavioral tests. The regional density of 5-HT1A, 5-HT1B, 5-HT2A and 5-HT2C receptors has been measured in the brains of parental strains, F-1s and F-2 populations by quantitative autoradiography. The results suggest that chromosomal fragments containing the brown, pink-eyed dilution and the short-ear loci, previously shown to be involved in anxiogenic processes, are mainly associated with a variation in the density of the 5-HT1B receptors.

The localization of the rat brain 5-HTIA receptor mRNA was analyzed by RNAse mapping and in situ hybridization during postnatal development, particularly in the cerebellum. The regional distribution of 5-HTIA mRNA during the first 2 postnatal weeks was different from that found in adults. In some areas of the immature brain (hippocampus, cerebral cortex), 5-HTIA mRNA was found in lower density than in the adult brain. In contrast, high concentrations of the transcript were present in other brain structures only during the first days after birth. Thus, in the cerebellum, the density of 5-HTIA mRNA decreased markedly from day 2 to day 9 after birth and could hardly be detected in the adult animal. The localization of the mRNA in the molecular/Purkinje cell layer of the immature cerebellum agreed with that of the 5-HTIA receptor protein visualized by immmunocytochemistry and was consistent with the hypothesis that Purkinje cells express this receptor.

The serotonin 5-HT1A receptors were visualized in the cerebellar vermis of 8-day-old and adult rats by immunocytochemistry using anti-5-HT1A receptor antibodies raised against a synthetic peptide corresponding to a highly selective portion of the receptor amino acid sequence (El Mestikawy et al, Neurosci Lett 118, 189-192, 1990). The 5-HT1A receptor-like immunoreactivity was particularly abundant in the posterior lobules (IXB-X) of the immature vermis where it was found in the molecular-Purkinje cell layers. Immunostaining was confined to the plasmic membrane of the Purkinje cell somas, dendrites and perhaps axons (at their somatic emergence) suggesting that 5-HT receptors might participate in non-junctional 5-HT neurotransmission in the immature cerebellum.

5-Hydroxytryptamine1A receptors were studied in rats during the first postnatal month in the normal cerebellum and in the granule cell-deprived cerebellum produced by X-irradiation at postnatal day 5. Quantitive autoradiographic studies on sagittal sections of cerebellar vermis, using [I-125]BH-8-MeO-N-PAT as radioligand or specific anti-receptor antibodies, revealed that 5-hydroxytryptamine1A receptors existed in the molecular/Purkinje cell layer but at variable density from one lobule to another. Thus, in both normal and X-irradiated rats, the posterior lobules were more heavily labelled than the anterior ones, and the density of 5-hydroxytryptamine1A sites decreased progressively in all the cerebellar folia down to hardly detectable levels at postnatal day 21. However, the intensity of labelling remained higher at postnatal day 8 and postnatal day 12 in X-irradiated rats than in age-paired controls. Measurements of [H-3]8-OH-DPAT specific binding to membranes from whole cerebellum confirmed that the density of 5-hydroxytryptamine1A sites per mg membrane protein (B(max)) was higher in X-irradiated animals than in age-paired controls. However, on a "per cerebellum" basis, no significant difference could be detected between the total number of 5-hydroxytryptamine1A sites, which progressively increased in both control and X-irradiated animals during the first postnatal month.These results therefore show that 5-hydroxytryptamine1A receptors are not located on developing granule cells. The progressive decrease in 5-hydroxytryptamine1A receptor density during the first postnatal month did not reflect a transient expression of 5-hydroxytryptamine1A receptors in the cerebellum of newborn rats, but resulted from the progressive "dilution" of these sites in this growing structure. The higher density of 5-hydroxytryptamine1A sites in X-irradiated rats simply reflected a lower "dilution" due to the delayed growth of the cerebellum in these animals.

Abstract: Numerous data suggested that the pharmacological and biochemical properties of 5‐hydroxytryptaminelA (5‐HTIA) receptors exhibit some regional differences in the CNS, notably within the raphe nuclei compared with various fore‐brain areas (such as the hippocampus). This possibility has been further investigated in the dorsal raphe nucleus and two areas within the hippocampus, the dentate gyrus and the CA1 area, using the quantitative autoradiographic technique. The potencies of 5′‐guanylylimidodiphosphate to inhibit the specific binding of 125I‐Bolton‐Hunter‐8‐methoxy‐2‐(N‐propyl‐N‐propylamino)tetralin (125I‐BH‐8‐MeO‐N‐PAT) to 5‐HT1A sites and of N‐ethylmaleimide to block these sites irreversibly were identical in the dorsal raphe nucleus and the hippocampal areas in rat brain sections. In contrast, slight but significant differences were noted in the pH dependence and pharmacological properties of 5‐HT1A sites labeled by 125I‐BH‐8‐MeO‐N‐PAT in these three regions. Similarly, heat denaturation experiments and tissue exposure to either phospholipase A2 or the alkylating agent 8‐methoxy‐2‐(N‐2′‐chloropropyl,N‐propyl)aminotetraline revealed regional differences in the properties of 5‐HT1A sites. However, in most cases, the observed variations were of greater amplitude between the CA1 area and the dentate gyrus, where 5‐HT1A sites are located postsynaptically, than between any one of these areas and the dorsal raphe nucleus where they act as (presynaptic) so‐matodendritic autoreceptors. These data further support that subtypes of 5‐HT1A receptors probably exist in the rat brain, but this heterogeneity seems unrelated to the pre‐ or postsynaptic location of these receptors.

The autoradiographic technique is the most relevant approach for the visualization at the light microscope level of the different classes (5-HT1A, 5-HT1B, 5-HT1C, 5-HT1D, 5-HT2 and 5-HT3) of receptors for the monoamine neurotransmitter serotonin (5-HT) in the central nervous system of mammals, including man. The only exception is the 5-HT4 subtype for which no satisfactory radioligand has been developed to date. Quantitative estimates of receptor labelling can be achieved by measurements of optical density on autoradiographic films of brain sections incubated with specific radioligands. This review summarizes the most significant contributions of quantitative autoradiography to the current knowledge of the respective regional distributions and differential regulations of the various classes of central 5-HT receptors.

The regional distribution of the mRNA encoding the 5-HT1A serotonin receptor (whose selective agonists are potential anxiolytic and antidepressant drugs) was investigated in rat brain sections by in situ hybridization histochemistry using two sets of [32P]labelled nucleoprobes, a riboprobe of 156 bases and oligoprobes of 30 bases corresponding to highly selective portions within the third intracellular loop and the N terminus domain of the amino acid sequence. These probes allowed the visualization of the 5-HT1A mRNA mainly in the limbic regions: dentate gyrus and area CA1 of the hippocampus, amygdala, entorhinal cortex, lateral septum and the dorsal raphe nucleus. These structures were also those which could be labelled by the specific 5-HT1A radioligand [125I]BH-8-MeO-N-PAT and antibodies raised against a synthetic 26 amino acid peptide whose sequence was taken from the most selective portion of the rat 5-HT1A receptor protein. These data suggest that the 5-HT1A receptors are not transported to a long distance from their site of synthesis, as it has been already reported for the somato-dendritic 5-HT1A autoreceptors in the dorsal raphe nucleus. Combined autoradiographic quantification of the 5-HT1A binding sites (labelled by a selective radioligand such as [125I]BH-8-MeO-N-PAT, the 5-HT1A receptor binding subunit (by radioimmunohistochemistry) and the 5-HT1A mRNA on adjacent brain sections should be a relevant approach for assessing the molecular mechanisms responsible for the functional alterations of these receptors under various pathological and pharmacological conditions.

Previous studies have shown that functional 5-HT1A receptors are present in the cerebellum only for the early postnatal period in rats. In order to investigate further the possible physiological significance of such a transient expression of 5-HT1A receptors during maturation of the cerebellum, anatomical studies were performed for identifying which cell type(s) are endowed with these receptors in 8-day-old rats. Autoradiography (using [125I]BH-8-MeO-N-PAT) with dry films and emulsion-coated coverslips, and radioimmunohistochemistry (using specific polyclonal anti-5-HT1A receptor antibodies) of vermis sections revealed that 5-HT1A receptors were mainly concentrated in the molecular layer of the anterior part of the lobule X and the posterior part of the lobule IXB. X-Irradiation on the 5th postnatal day yielded an agranular cerebellum whose density of 5-HT1A sites was higher than that in age-paired control animals. These data indicate that 5-HT1A receptors are not located on granule cells, but probably on glial cells in the molecular layer of the immature cerebellum. This location further supports the possible implication of glial 5-HT1A receptors in some trophic action of 5-HT during CNS maturation.

Measurements of endogenous levels of serotonin (5-HT), 5-hydroxyindole acetic acid (5-HIAA), dopamine (DA) and dihydroxyphenyl acetic acid (DOPAC), and biochemical and autoradiographic investigations on 5-HT and DA receptors were made in various brain regions in male rats at three different ages: 3 months, 10 months and 22 months. Age-dependent decreases in 5-HT levels associated with parallel increases in 5-HIAA/5-HT ratio were observed in the hypothalamus, striatum, hippocampus and cerebral cortex, suggesting an accelerated 5-HT turnover in aged rats. Similarly, DA levels were lower, and DOPAC/DA ratio was higher in the striatum of 22-month-old compared to 3-month-old or 10-month-old rats. Of the three different classes of 5-HT receptors which were examined, 5-HT1B sites exhibited the largest age-dependent decrease in density, followed by 5-HT2 sites, while 5-HT1A sites remained practically unchanged during aging. By comparison, the loss of striatal D2 receptors in 22-month-old rats compared to young adults was much greater than that of any 5-HT receptor subtype. Such differential age-dependent alterations of the various classes of 5-HT receptors and of dopaminergic versus serotoninergic synaptic markers might be responsible for at least some of the functional deficits in aged animals.

Polyclonal antibodies were raised by the repeated injection of rabbits with a synthetic peptide corresponding to a highly selective portion (amino acid residues 243 to 268) of the amino acid sequence of the rat 5-HT1A receptor. The anti-peptide antiserum allowed the immunoprecipitation of 5-HT1A receptors but not of other 5-HT1 sites solubilized from rat hippocampal membranes. Immunoautoradiographic labelling of rat brain sections with the anti-peptide antiserum was superimposed with the autoradiographic distribution of 5-HT1A sites labelled by the selective radioligand [3H]8-OH-DPAT.

The possible irreversible blockade of 5-hydroxytryptamine1 receptor subtypes 5-hydroxytryptamine1A, 5-hydroxytryptamine1B/5-hydroxytryptamine1D and 5-hydroxytryptamine1C by the chloramine 8-methoxy-2-(N-2'-chloropropyl,N-propyl)aminotetralin (8-MeO-2'-chloro-PAT) was investigated in rat brain sections by quantitative autoradiography using [3H]8-hydroxy-2-(di-n-propylamino)tetralin [( 3H]8-OH-DPAT), [3H]5-hydroxytryptamine, [125I]BH-8-MeO-N-PAT and [125I]cyanopindolol as radio-ligands. A marked reduction (-50% to -75%) of [3H]8-OH-DPAT and [125I]BH-8-MeO-N-PAT specific binding to 5-hydroxytryptamine1A sites in the hippocampus (CA1 area) and the dorsal raphe nucleus, and of [3H]5-hydroxytryptamine specific binding to 5-hydroxytryptamine1C sites in the choroid plexus was found in sections exposed to 1 microM 8-MeO-2'-chloro-PAT and then washed extensively. In contrast the specific binding of [3H]5-hydroxytryptamine to 5-hydroxytryptamine1B/5-hydroxytryptamine1D sites and of [125I]cyanopindolol to 5-hydroxytryptamine1B sites in the substantia nigra and dorsal subiculum remained unaltered by this treatment. Similarly [125I]cyanopindolol binding to beta-adrenergic receptors was not affected by 8-MeO-2'-chloro-PAT. Prior occupancy of 5-hydroxytryptamine1A sites by 10 microM 5-hydroxytryptamine or 8-OH-DPAT, and of 5-hydroxytryptamine1C sites by 10 microM 5-hydroxytryptamine prevented any subsequent blockade by 8-MeO-2'-chloro-PAT. These data indicate that 8-MeO-2'-chloro-PAT should be a useful alkylating agent for achieving selective irreversible blockade of central 5-hydroxytryptamine1A and 5-hydroxytryptamine1C receptors in vivo in the rat.