Contribution of the d-Serine-Dependent Pathway to the Cellular Mechanisms Underlying Cognitive Aging.

An association between age-related memory impairments and changes in functional plasticity in the aging brain has been under intense study within the last decade. In this article, we show that an impaired activation of the strychnine-insensitive glycine site of N-methyl-d-aspartate receptors (NMDA-R) by its agonist d-serine contributes to deficits of synaptic plasticity in the hippocampus of memory-impaired aged rats. Supplementation with exogenous d-serine prevents the age-related deficits of isolated NMDA-R-dependent synaptic potentials as well as those of theta-burst-induced long-term potentiation and synaptic depotentiation. Endogenous levels of d-serine are reduced in the hippocampus with aging, that correlates with a weaker expression of serine racemase synthesizing the amino acid. On the contrary, the affinity of d-serine binding to NMDA-R is not affected by aging. These results point to a critical role for the d-serine-dependent pathway in the functional alterations of the brain underlying memory impairment and provide key information in the search for new therapeutic strategies for the treatment of memory deficits in the elderly.

A critical role for the glial-derived neuromodulator D-serine in the age-related deficits of cellular mechanisms of learning and memory.

Age-associated deficits in learning and memory are closely correlated with impairments of synaptic plasticity. Analysis of N-methyl-D-aspartate receptor (NMDAr)-dependent long-term potentiation (LTP) in CA1 hippocampal slices indicates that the glial-derived neuromodulator D-serine is required for the induction of synaptic plasticity. During aging, the content of D-serine and the expression of its synthesizing enzyme serine racemase are significantly decreased in the hippocampus. Impaired LTP and NMDAr-mediated synaptic potentials in old rats are rescued by exogenous D-serine. These results highlight the critical role of glial cells and presumably astrocytes, through the availability of D-serine, in the deficits of synaptic mechanisms of learning and memory that occur in the course of aging.

Somatostatin receptors 2 and 5 are the major somatostatin receptors in insulinomas: an in vivo and in vitro study.

Somatostatin (SRIF) receptors (sst) are present on normal pancreatic endocrine beta-cells. However, the use of SRIF analogs in the scintigraphic imaging of insulinomas and in the medical management of these tumors seems to be restricted to a subgroup of patients. The aim of this study was to determine the prevalence of sst expression in vitro and characterize sst subtype binding in insulinomas and its correlation with in vivo sst receptor scintigraphy (SRS). In vitro studies were performed on 27 insulinomas from 25 patients: 22 with benign and three with malignant tumors. Semiquantitative RT-PCR of sst mRNAs was performed for 20 of these insulinomas. Sst2 and sst5 were expressed in 70%, sst1 in 50%, and sst3 and sst4 subtypes only in 15-20% of the tumors. (125)I-Tyr(0)DTrp(8)SRIF(14) binding was assessed by quantitative autoradiography in 18 insulinomas, and competition experiments were performed with SRIF(14) and L797-591, L779-976, L796-778, L803-087, L817-818, selective agonists of the five sst subtypes, and BIM23244, a selective agonist of sst2 and sst5. Significant specific binding was observed in 72% of the insulinomas. Displacement experiments with ligands of higher affinity for each of the sst receptors revealed significant binding with the sst2 and sst5 ligands in 72%, sst3 in 44%, sst1 in 44%, and sst4 in 28% of cases. All insulinomas displaying sst2 binding were also sst5 sensitive. However, the ratio of sst5/sst2 displacement was variable and only equal to that for SRIF(14) in experiments with the sst2/sst5 agonist BIM23244. SRS was performed 10 times in nine patients; it detected 60% of the tumors, including metastases of a malignant insulinoma. All the tumors detected by SRS displayed high levels of (125)I-Tyr(0)DTrp(8)SRIF(14) binding. The mechanisms underlying the loss of expression of sst2/sst5 in a third of insulinomas remains to be determined, but this loss of expression may be involved in beta-cell dysfunction.

Overexpression of the V3 vasopressin receptor in transgenic mice corticotropes leads to increased basal corticosterone.

The vasopressin V3 receptor (V3) is specifically expressed in pituitary corticotropes and mediates the stimulatory effect of vasopressin on adrenocorticotropic hormone (ACTH) release. The V3 gene is overexpressed in corticotrope pituitary tumours compared to normal pituitaries. We hypothesized that V3 overexpression might induce changes in corticotrope function and alter the regulation of the hypothalamic-pituitary-adrenal axis. Thus, we generated transgenic mice (POMV3) expressing the human V3 receptor in the pituitary under the control of rat pro-opiomelanocortin (POMC) promoter sequences. The transgene was efficiently transcribed and vasopressin binding was increased in both corticotropes and melanotropes. In-vitro ACTH release and inositol phosphate formation were unchanged in POMV3 pituitaries, but the responses to vasopressin were significatively increased. In vivo, basal circulating concentrations of ACTH in POMV3 mice were similar to those of controls but corticosterone concentrations were moderately increased. In addition, the levels of POMC mRNA in the transgenic pituitaries were comparable to those of control mice. Finally, POMV3 mice responded with a similar maximal increase of ACTH and corticosterone to a 20-min acute restraint stress. Together, these results show that hypophyseal V3 overexpression led to increased basal concentrations of corticosterone and suggest that the negative glucocorticoid feedback may be altered at the pituitary level.

Comparison of somatostatin receptor expression in human gliomas and medulloblastomas.

The expression of the five somatostatin receptor subtypes, sst1-5 was compared on tissue containing glial tumours (glioblastomas or oligodendrogliomas), medulloblastomas, and on normal human cortex. By semiquantitative reverse transcription coupled to polymerase chain reaction, the receptor expression profiles were high in cortex and in tissue containing oligodendrogliomas. It was moderate in medulloblastomas. Tissue containing glioblastomas displayed lower expression of somatostatin receptor subtypes, sst1 and sst3 being mostly expressed. By 125I-Tyr0DTrp8 somatostatin-14 or 125I-Leu8DTrp22 Tyr25 somatostatin-28 autoradiography combined with synaptophysin immunohistochemistry, it was possible to differentiate between isolated tumoral cell component infiltrating the cerebral parenchyma (cortex or white matter) and tumoral tissue (without residual parenchyma) in glioblastomas or oligodendrogliomas. Glial tumoral tissue per se presented few somatostatin receptors. By contrast, medulloblastoma tumoral cells exhibited numerous octreotide sensitive somatostatin receptors. sst2 immunocytochemistry demonstrated immunostaining of neuronal cells and neuropile; sst2 and sst3 immunostaining was identified on glioblastoma proliferating vessels endothelial cells and on medulloblastomas tumoral cells. Faint sst2 immunostaining among glial tumoral cells was due to microglia, while glioma cells did not significantly stain. In summary, medulloblastoma tumoral cells express sst2/sst3 receptors at a high level while glioma cells do not. In gliomas, sst expression is restricted to endothelial cells on proliferating vessels (displaying both sst2 and sst3 receptors), including parenchyma and reactive microglia (only sst2). The differential expression of sst2/sst3 receptors on gliomas and medulloblastomas has implications for the therapy of these tumours.

Homologous upregulation of sst2 somatostatin receptor expression in the rat arcuate nucleus in vivo.

In vitro studies using various cell systems have provided conflicting results regarding homologous regulation of somatostatin (SRIH) receptors, and information on whether SRIH regulates the expression of its own receptors in vivo is lacking. In the present study we examined, by in situ hybridization, the effects of pretreatment with the sst2-preferring SRIH analog, octreotide, in vivo, on mRNA levels of two SRIH receptor subtypes, sst1 and sst2, in rat brain and pituitary. (125)I-[DTrp(8)]-SRIH binding was also measured in these regions. Three hours after the iv injection of 50 microg octreotide to conscious adult male rats, there was a 46% increase (p < 0.01) in the labeling density of sst2 mRNA-expressing cells in the hypothalamic arcuate nucleus compared to normal saline-pretreated controls, but not in any of the other brain regions examined. Computer-assisted image analysis revealed that 3 h exposure to octreotide significantly (p < 0.01) augmented both the number and labeling density of sst2 mRNA-expressing cells in the arcuate nucleus, compared to those in saline-treated controls. By contrast, within the anterior pituitary gland, in vivo exposure to octreotide did not affect the expression of sst2 mRNA. No changes in sst1 mRNA-expressing cells were observed after octreotide treatment in any of the regions measured, indicating that the observed effects were homologous, i.e. specific of the receptor subtype stimulated. Octreotide pretreatment was also without effect on the density of (125)I-[DTrp(8)]-SRIH binding in either the arcuate nucleus or pituitary. These results demonstrate, for the first time, that SRIH preexposure in vivo upregulates the expression of a subtype of its own receptors, sst2, within the central nervous system. They further suggest that pretreatment with SRIH in vivo does not cause sst2 receptor desensitization in arcuate nucleus and pituitary. Such homologous regulatory mechanisms may play an important role in the neuroendocrine control of growth hormone (GH) secretion by the arcuate nucleus.

Somatostatin and behaviour: the need for genetically engineered models.

Somatostatin was originally characterised as a hypothalamic neurohormone responsible for the inhibition of pituitary Growth Hormone secretion. In mammals two genes encode for somatostatin-related peptides, somatostatin 14 and 28, and cortistatins, respectively. All peptides bind with similar affinities to the five cloned somatostatin receptors (sst), which belong to the GPCR family. Despite numerous studies, no clear behavioural function has yet been attributed to somatostatin-related peptides. This is due to the lack of good pharmacological tools (selective antagonists) and animal models. This review will focus on the recent development of such tools.

Involvement of sst2 somatostatin receptor in locomotor, exploratory activity and emotional reactivity in mice.

Somatostatin (SRIF) controls many physiological and pathological processes in the central nervous system but the respective roles of the five receptor isotypes (sst1-5) that mediate its effects are yet to be defined. In the present study, we attempted to identify functions of the sst2 receptor using mice with no functional copy of this gene (sst2 KO mice). In contrast with control 129Sv/C57Bl6 mice, sst2 mRNA was no longer detectable in the brain of sst2 KO mice; 125I-labeled Tyr0DTrp8-SRIF14 binding was also greatly reduced in almost all brain structures except for the hippocampal CA1 area, demonstrating that sst2 accounts for most SRIF binding in mouse brain. Invalidation of this subtype generated an increased anxiety-related behaviour in a number of behavioural paradigms, while locomotor and exploratory activity was decreased in stress-inducing situations. No major motor defects could be detected. sst2 KO mice also displayed increased release of pituitary ACTH, a main regulator of the stress response. Thus, somatostatin, via sst2 receptor isotype pathways, appears involved in the modulation of locomotor, exploratory and emotional reactivity in mice.

Distribution and pharmacological characterization of somatostatin receptor binding sites in the sheep brain.

Somatostatin binding sites have been localized and quantified in the sheep brain using 125I-Tyr0-DTrp8-somatostatin, by quantitative high resolution light microscopic autoradiography. Sections were analyzed by densitometry on radioautographic film, and subsequently on slides coated with photoemulsion. Specific somatostatin binding sites were concentrated in the medial habenula, superior colliculus, dorsal motor nucleus of the vagus nerve, inferior olive, spinal trigeminal nucleus, and cerebellum. In competition experiments, octreotide, a sst2/sst3/sst5 selective agonist only partially displaced 125I-Tyr0-DTrp8-somatostatin in the three cerebellar layers while it was fully active as compared to somatostatin 14 and 28 in the deeper layers of the parietal cortex. Moderate to low somatostatin receptor densities were present in the mesencephalic periaqueductal gray, dorsal raphe, thalamic paraventricular nucleus, interpeduncular nucleus, pineal gland, dorsal tegmental, dorsolateral tegmental and parabrachial nuclei, nucleus of the solitary tract. The distribution of somatostatin binding sites generally correlates with the data obtained on slides dipped in photoemulsion which provided better resolution and more precise localization. In most of the labeled areas, 125I-Tyr0-DTrp8-somatostatin receptor binding was distributed between both neuropil and perikarya. Perikarya bearing 125I-Tyr0-DTrp8-somatostatin receptors were observed in areas which did not display detectable binding sites on film such as the preoptic-anterior hypothalamic complex and arcuate nucleus and in the locus coeruleus. In conclusion, the distribution of 125I-Tyr0-DTrp8-somatostatin binding sites in sheep brain is very reminiscent of other mammals being closer to the human than to rodents.

Sex-related alterations in hypothalamic growth hormone-releasing hormone mRNA-but not somatostatin mRNA-expressing cells in genetically obese Zucker rats.

The possibility that the growth hormone (GH) suppression associated with obesity is due to alterations in hypothalamic GH-releasing hormone (GHRH) and/or somatostatin (SRIH) has been considered, but the data are not consistent. In the present study, we sought to clarify the roles of GHRH and SRIH in obesity by using in situ hybridization to localize and quantify the level of expression of GHRH mRNA- and SRIH mRNA-containing neurons in the hypothalamus of male and female lean and obese Zucker rats (12 weeks of age; n = 6 per group). In lean animals, the number of GHRH mRNA-expressing cells in the arcuate nucleus and SRIH mRNA-containing neurons in the periventricular nucleus was 2- to 3-fold higher in males compared to females. The obese phenotype in the male was associated with a striking reduction in arcuate GHRH mRNA expression, both in terms of number of cells (-71%; p < 0.01) and grains/cell (-44%; p < 0.05). In contrast, in obese females, there was a marked augmentation (+ 175%; p < 0.05) in the number of GHRH mRNA-containing cells in the arcuate nucleus compared to their lean littermates. The small population of GHRH mRNA-containing neurons of the ventromedial nucleus was not modified in male obese rats, while it was considerably increased (p < 0.05) in obese females. Neither the number of labeling density of SRIH mRNA-containing neurons in the periventricular and arcuate nuclei of obese rats of either sex was changed when compared to their sex-matched lean counterparts. These results demonstrate that: (1) the obese male Zucker rat exhibits a marked diminution in hypothalamic GHRH mRNA expression, while a reverse pattern is evident in the obese female; (2) hypothalamic SRIH mRNA-containing neurons are not significantly altered in obese rats of both sexes. Our findings suggest that the impaired GH secretion of the obese Zucker rat is due, at least in part, to alterations in hypothalamic GHRH gene expression and that SRIH does not play a major role.

Effects of neonatal administration of octreotide, a long-lasting somatostatin analogue, on growth hormone regulation in the adult rat.

The pulsatile pattern of GH secretion slowly develop in the postnatal period concomitantly with the dual network of GHRH and somatostatin (SRIH) hypothalamic neurons. We investigated whether an early postnatal treatment with a long acting SRIH analogue, octreotide, could affect maturation and subsequent operation of those networks in the adult rat. Octreotide administration (5 mu g/rat SC) every other day during the first 10 days of life resulted in growth retardation in the adult. In parallel, the amplitude of plasma GH secretory episodes in free moving unanesthetized animals was markedly reduced. The numbers of arcuate GHRH mRNA-containing and periventricular SRIH-mRNA containing neurons were not affected by the treatment. GHRH mRNA levels per neuron however was decreased by 30%, and median eminence GHRH stores by 50%. SRIH expression in the arcuate nucleus was also diminished, as was the number of 125I-SRIH labeled neurons in that nucleus. The effects of octreotide were compared to the hyposomatotropinemia induced by administration of monosodium glutamate (MSG), every other day during the first 10 days of life. Growth retardation and inhibition of GH secretory episodes in adult rats neonatally treated with MSG were slightly more pronounced than after octreotide. In contrast to octreotide, MSG induced a massive loss of GHRH neurons and a concomitant decrease in 125I-SRIH binding. Somatostatin did not protect GHRH neurons against the neurotoxic action of MSG since octreotide treatment did not further affect any of the parameters impaired by MSG. In conclusion, these experiments demonstrate that neonatally injected octreotide cannot counteract the toxic effect of MSG on arcuate neurons. However, a neonatal treatment with the SRIH agonist affects permanently growth rate and GH pulsatility. This effect is mediated in the hypothalamus by permanently impairing the neural networks that control GH secretion.

The increase in growth hormone secretion in experimentally induced arthritic rats is an adaptive process involved in the regulation of inflammation.

In Sprague-Dawley rats, Freund-adjuvant-induced arthritis (AIA) results in an increase in the amplitude of ultradian growth-hormone (GH)-secretory episodes without modification of their frequency. This is most apparent at the time of maximal inflammation, i.e. 14-21 days after inoculation of the adjuvant. GH responsiveness to a maximal dose of clonidine (10 micrograms/100 g body weight, BW), a secretagogue known to act at the hypothalamic level, is comparable in AIA and control rats. In contrast, GH response to a maximal dose of GH-releasing hormone (GHRH, 1 microgram/100 g BW), a peptide acting directly on pituitary somatotropes, is greater in AIA than in control rats. Furthermore AIA affects significantly neither hypothalamic somatostatin and GHRH mRNA levels nor pituitary GH content. In adult rats treated neonatally with monosodium glutamate (MSG), a neurotoxin which destroys the majority of GHRH neurons of the arcuate nucleus and reduces considerably plasma GH levels, clinical symptoms observed 14 days after inoculation of the Freund adjuvant are more marked than in AIA. The MSG-treated rats exhibit in particular a significantly higher increase in hindpaw diameter. Pulsatile administration of GH (40 micrograms/day/rat, with successive periods of 2 h of GH and 4 h of mineral oil) restoring the endogenous GH-secretory pattern throughout the 15-day period of arthritis development prevents hindpaw diameter increase. These results indicate that the impact of AIA on GH regulation occurs at the pituitary but not the hypothalamic level and suggest that increased GH secretion observed in AIA rats is an adaptive mechanism involved in the regulation of the inflammatory process.

125I-somatostatin-labeled cells in the anterior arcuate nucleus mediate somatostatin effects on growth hormone but not prolactin secretion.

The regional brain distribution of 125I-somatostatin (SRIH) binding sites was determined by quantitative radioautography in neonatally monosodium glutamate (MSG) treated adult male rats, a procedure which selectively destroys most neurons of the arcuate nucleus. Neonatal MSG treatment did not modify the extrahypothalamic distribution of 125I-SRIH-binding sites. In contrast, the number of 125I-SRIH-labeled cells in the ventrolateral part of the arcuate nucleus was strongly reduced in MSG-treated animals. The effect was selective for the anterior part of the arcuate nucleus and was not found in its posterior part or in the cells located more dorsally, beneath the ependymal zone of the periventricular nucleus. Intracerebroventricular SRIH injections, which increased growth hormone levels in control rats, were totally ineffective in MSG-treated animals. In contrast, the prolactin levels were equally stimulated by intracerebroventricular injections in control and MSG-treated animals. These results demonstrate that extrahypothalamic SRIH-binding sites are not located on neurons originating in the anterior arcuate nucleus neurons. In addition, 125I-SRIH-labeled cells in the ventrolateral part of the arcuate nucleus are necessary for the paradoxical stimulation of growth hormone secretion induced by intracerebroventricular SRIH injection, but do not seem to be essential for the increased prolactin secretion observed under these conditions.

Estradiol regulation of somatostatin receptors in the arcuate nucleus of the female rat.

Somatostatin receptors on lactotroph cells of the anterior pituitary are positively regulated by estradiol. In the present work, we investigated whether estradiol regulation of somatostatin receptors also occurred in the female rat brain. 125I-Tyr0-DTrp8-somatostatin (125I-SRIF: 780 Ci/mM) was used as a ligand. Female adult rats were ovariectomized and treated or not with estradiol benzoate (20 micrograms/day for 1 or 8 days). In female brains, 125I-SRIF binding, as assessed by film radioautography, was high in the basolateral amygdala, CA1 field and dentate gyrus of the hippocampus and locus coeruleus, moderate in the median habenula and deep layers all through the cortex. Castration or estradiol treatment did not modify 125I-SRIF binding in these regions. By light-microscopic radioautography, a subpopulation of 125I-SRIF-labeled cells was localized in the ventrolateral portion of the arcuate nucleus. Ovariectomy alone did not significantly affect the number and binding density of 125I-SRIF-labeled cells in the arcuate nucleus. However, estradiol treatment in ovariectomized animals significantly increased both parameters. Along the estrus cycle, the number of 125I-SRIF-labeled cells was not significantly modified but 125I-SRIF binding density was significantly higher in proestrus as compared to diestrus I, diestrus II and estrus. These results demonstrate that brain 125I-SRIF binding sites are positively regulated by estradiol only in the arcuate nucleus of the hypothalamus.

6-hydroxydopamine lesions of the locus coeruleus induce a paradoxical increase in growth hormone secretion in male rats.

While the pharmacology of noradrenaline effects on growth hormone (GH) secretion has been extensively studied, the precise localization of noradrenergic neurons involved remains unclear. In the present work, we investigated whether A6 noradrenergic neurons located in the locus coeruleus can play a role in the rhythmic pattern of GH secretion or in the sensitivity of the hormone response to different external challenges. Three weeks after bilateral 6-hydroxydopamine injections (8µg/3µl) into the locus coeruleus, hypothalamic noradrenaline concentrations were reduced by 60%. Pulsatile GH secretory patterns were observed in unanaesthetized, freely moving control, sham-operated or locus coeruleus-lesioned male rats. The amplitude of the pulses and the area under the curves during the 6- or 12-h sampling period were twice as high in locus coeruleus-lesioned than in control and sham-operated rats. In contrast, trough levels of GH and intervals between GH peaks were similar in all groups. Prolactin, adrenocorticotrophin, thyroid-stimulating hormone and luteinizing hormone plasma levels were not affected by the lesion. GH responses to two centrally acting drugs i.e. clonidine (2.5, 5 and 10µg/100g body wt) and morphine (200µg/100g body wt) were also highly amplified in locus coeruleus-lesioned rats. In contrast, GH responses to two peptides directly acting on somatotrophs i.e. GH-releasing factor (0.05 and 1.25µg/100g body wt) and vasoactive intestinal peptide (1.5µg/100g body wt) were the same in sham-operated and lesioned animals. These data suggest that noradrenergic inputs from the locus coeruleus exert a selective inhibitory influence on GH secretion through centrally mediated mechanisms.

Regional distribution of somatostatin binding sites in the human hypothalamus: a quantitative autoradiographic study.

Using in vitro quantitative autoradiography and [125I]Tyr0-D-Trp8SRIF 14 as radioligand, we characterized the detailed distribution of somatostatin binding sites in human hypothalamus of both infants and adults. Guanosine triphosphate pretreatment, before incubation, allowed us to detect higher [125I]Tyr0-D-Trp8SRIF 14 binding site densities in hypothalamic structures such as preoptic and anterior hypothalamic areas and ventromedial and dorsomedial nuclei. In contrast, guanosine triphosphate was without effect in the other hypothalamic regions. The regional effects of guanosine triphosphate pretreatment were not different in infant and adult hypothalamus. Scatchard analysis showed that in a guanosine triphosphate-sensitive region (preoptic area) and a guanosine triphosphate-insensitive area (infundibular nucleus), [125I]Tyr0-D-Trp8SRIF 14 bound to a single class of binding sites. Affinities were similar in both regions, not modified by guanosine triphosphate pretreatment and not different in the adult (1.5 +/- 1.2 nM vs 3.2 +/- 2.1 nM for preoptic area and infundibular nucleus, respectively) and infant (0.9 +/- 0.5 nM vs 2.4 +/- 1.7 nM for preoptic area and infundibular nucleus). [125I]Tyr0-D-Trp8SRIF 14 binding sites were widely distributed in the anterior, mediobasal and posterior hypothalamus. Somatostatin 28 was twice as potent as somatostatin 14 to displace [125I]Tyr0-D-Trp8SRIF 14 binding in the preoptic area and infundibular nucleus. However, IC50s were 30 times lower in the preoptic area as compared with the infundibular nucleus. In adult as well as in infant, high densities were found mainly in the diagonal band of Broca, preoptic area and infundibular nucleus. Intermediate densities were localized in the anterior hypothalamic area, ventromedial, dorsomedial and lateral mammillary nuclei. The dorsal hypothalamic area, the paraventricular and medial mammillary nuclei displayed low but measurable densities. The only marked difference in the distribution of [125I]Tyr0-D-Trp8SRIF 14 binding sites in adult vs infant was observed in the medial and tuberal nuclei where the concentrations were seven-fold higher in adult hypothalamus.

Characterization of pericellular [125I]Tyr0 DTrp8 somatostatin binding sites in the rat arcuate nucleus by a newly developed method: quantitative high-resolution light microscopic radioautography.

In the present work we characterized the kinetic properties of [125I]somatostatin pericellular binding sites in the arcuate nucleus of the hypothalamus of the rat by quantitative high-resolution light microscopic radioautography. In order to determine whether these pericellular binding sites corresponded to functional receptors, their properties were compared with those of previously well-characterized [125I]somatostatin binding sites present on neuronal processes on the same sections in the stratum radiatum of the CA1 of the hippocampus. Radiolabelled sections were analysed by densitometry using a Biocom image analysis system coupled with a Leitz orthoplan microscope. The linear relationship between optical densities and radioactive standards allowed us to quantitate [125I]somatostatin-specific binding. Binding was time- and temperature-dependent, and saturable and specific in the arcuate nucleus as in the CA1 of the hippocampus. Saturation experiments indicated a single receptor population of binding sites with KD values of 0.2 +/- 0.1 nM in the arcuate nucleus and 0.6 +/- 0.4 nM in the CA1. In both structures, displacement curves obtained with somatostatin 14 and somatostatin 28 were monophasic, but shallow, while the somatostatin analogue SMS 201-995 induced a biphasic displacement, suggesting two populations of binding sites. In both regions binding was GTP-dependent. Desaturation procedures (in vivo by cysteamine and in vitro by preincubating with GTP) resulted in an increase in the number of measurable binding sites.(ABSTRACT TRUNCATED AT 250 WORDS)

125I-[Tyr0,D-Trp8]somatostatin-14 binding sites in the locus coeruleus of the rat are located on both ascending and descending projecting noradrenergic cells.

Radioautographic determinations of 125I-[Tyr0,D-Trp8]somatostatin-14 (125I-SRIF) binding sites were performed on frozen serial sections of the locus coeruleus (LC) of control rats and of rats subjected to either bilateral microinjections of 6 hydroxydopamine (6-OHDA) into the LC or unilateral microinjection into the ascending noradrenergic bundles. These experiments were performed in order to determine whether 125I-SRIF binding was localized to noradrenergic-containing cells and in which regions the cells which contain the binding sites are projecting. The extent of the lesions was assessed by measuring norepinephrine (NE) levels in the hippocampus (88% decrease as compared to sham-operated animals) for bilateral LC lesions and in the frontal cortex (87% reduction vs. contralateral side) for unilateral bundle lesions. In control rats, 125I-SRIF binding sites were restricted to the boundaries of the LC and followed closely the distribution of tyrosine hydroxylase-labeled cells. Three weeks after bilateral injections of 6-OHDA, 125I-SRIF binding decreased by 79% in all regions of the LC. In contrast, unilateral destruction of the ascending noradrenergic bundles resulted in a moderate decrease only in the middle part of the LC with a more important effect in the dorsal (55%) than in the ventral (24%) portion of the nucleus. These data demonstrate that: 1) most SRIF receptors in the LC are located in the vicinity of NE-containing cell bodies and 2) NE-containing cells bearing SRIF receptors project to the forebrain as well as to other terminal areas located more caudally in the brain. These data suggest a general role for SRIF in the control of the multiple functions of the LC.

Regional distribution of somatostatin receptor affinity states in rat brain: effects of divalent cations and GTP.

In the present work, we have characterized by film radioautography the effects of divalent cations and guanine nucleotide on specific receptor for somatostatin (SRIF) using 125I-TyrO-DTrp8-SRIF14 (125I-ToD8-SRIF) as a ligand. The experiments were performed on coronal 20-microM-thick sections cut at the level of the amygdala, thus allowing to study binding sites in several regions enriched in binding sites (frontal cortex, hippocampus CA1 and dentate gyrus, habenula, basolateral nucleus of the amygdala). In a preliminary set of experiments using brain cortical membranes it was found that 3 mM Mg2+ ions doubled the specific binding of 125I-ToD8-SRIF. However, Mg2+ enhanced equally by a factor of 3 affinities of high- and low-affinity binding sites as evidenced by SMS 201.995 displacement curves without modifying the ratio between high and low affinity sites. In radioautographic studies while SRIF14 and SRIF28 elicited monophasic displacement curves, SMS 201.995 displaced 125I-ToD8-SRIF binding in a biphasic manner in all regions tested but the baso-lateral nucleus of the amygdala. Radioautographic distribution of 125I-ToD8-SRIF binding sites was identical whether the sections were incubated with MgCl2 or with MnCl2 and almost undetectable in the absence of ions. In all structures investigated increasing concentrations of GTP totally inhibited 125I-ToD8-SRIF binding with an IC50 value of 3 microM. In conclusion, our results demonstrate that 125I-ToD8-SRIF-binding sites in brain occur on two different affinity states as assessed by a displacement curve using endogenous ligands and SMS 201.995. According to the comparable effects of divalent cations and GTP, the two subtypes of 125I-ToD8-SRIF-binding sites discriminated by SMS 201.995 are likely to correspond to interconvertible forms of the same receptor coupled to a G protein-transducing system.

Circadian patterns of growth hormone, prolactin and corticotrophin secretion in hypothalamic and extra-hypothalamic lesions localized by stereotactic neuro-radiology in humans.

The integrity of mediobasal hypothalamus for the maintenance of normal circadian rhythms has been clearly established in the animal. Using precise stereotactic neuroradiologic localisations of various brain tumors, the authors demonstrate that in human, cortical lesions induce no significant modifications of the circadian variation of GH, prolactin or ACTH. In contrast subcortical lesions displacing the 3rd ventricle result in an inhibiton of the nocturnal rise in GH and prolactin, whereas the circadian rhythm of ACTH remains normal.